zotero/chrome/content/zotero/xpcom/collectionTreeView.js

2306 lines
60 KiB
JavaScript
Raw Normal View History

/*
***** BEGIN LICENSE BLOCK *****
2009-12-28 09:47:49 +00:00
Copyright © 2009 Center for History and New Media
George Mason University, Fairfax, Virginia, USA
http://zotero.org
2009-12-28 09:47:49 +00:00
This file is part of Zotero.
2009-12-28 09:47:49 +00:00
Zotero is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
2009-12-28 09:47:49 +00:00
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
2009-12-28 09:47:49 +00:00
Zotero is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
2009-12-28 09:47:49 +00:00
You should have received a copy of the GNU Affero General Public License
2009-12-28 09:47:49 +00:00
along with Zotero. If not, see <http://www.gnu.org/licenses/>.
***** END LICENSE BLOCK *****
*/
////////////////////////////////////////////////////////////////////////////////
///
/// CollectionTreeView
/// -- handles the link between an individual tree and the data layer
/// -- displays only collections, in a hierarchy (no items)
///
////////////////////////////////////////////////////////////////////////////////
/*
* Constructor for the CollectionTreeView object
*/
Zotero.CollectionTreeView = function()
{
Zotero.LibraryTreeView.apply(this);
2007-10-23 07:11:59 +00:00
this.itemToSelect = null;
this.hideSources = [];
2007-10-23 07:11:59 +00:00
this._highlightedRows = {};
this._unregisterID = Zotero.Notifier.registerObserver(
this,
[
'collection',
'search',
'publications',
2015-02-03 11:57:32 -06:00
'feed',
'share',
'group',
'feedItem',
'trash',
'bucket'
],
2015-05-25 21:22:43 -04:00
'collectionTreeView',
25
);
this._containerState = {};
this._duplicateLibraries = [];
this._unfiledLibraries = [];
this._trashNotEmpty = {};
}
Zotero.CollectionTreeView.prototype = Object.create(Zotero.LibraryTreeView.prototype);
Zotero.CollectionTreeView.prototype.type = 'collection';
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Object.defineProperty(Zotero.CollectionTreeView.prototype, "selectedTreeRow", {
get: function () {
if (!this.selection || !this.selection.count) {
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return this.getRow(this.selection.currentIndex);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
/*
* Called by the tree itself
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.setTree = Zotero.Promise.coroutine(function* (treebox)
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
try {
if (this._treebox || !treebox) {
2007-10-23 07:11:59 +00:00
return;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
this._treebox = treebox;
// Add a keypress listener for expand/collapse
var tree = this._treebox.treeBody.parentNode;
tree.addEventListener('keypress', function(event) {
2015-03-16 23:40:57 -04:00
if (tree.editingRow != -1) return; // In-line editing active
var libraryID = this.getSelectedLibraryID();
if (!libraryID) return;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var key = String.fromCharCode(event.which);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (key == '+' && !(event.ctrlKey || event.altKey || event.metaKey)) {
this.expandLibrary(libraryID, true);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
else if (key == '-' && !(event.shiftKey || event.ctrlKey ||
event.altKey || event.metaKey)) {
this.collapseLibrary(libraryID);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
}.bind(this), false);
yield this.refresh();
if (!this._treebox.columns) {
2007-10-23 07:11:59 +00:00
return;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
this.selection.currentColumn = this._treebox.columns.getFirstColumn();
var lastViewedID = Zotero.Prefs.get('lastViewedFolder');
if (lastViewedID) {
var selected = yield this.selectByID(lastViewedID);
}
if (!selected) {
this.selection.select(0);
}
2015-06-11 16:06:00 -04:00
this.selection.selectEventsSuppressed = false;
yield this._runListeners('load');
this._initialized = true;
}
catch (e) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.debug(e, 1);
Components.utils.reportError(e);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (this.onError) {
this.onError(e);
}
throw e;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2015-06-11 16:06:00 -04:00
/**
* Rebuild the tree from the data access methods and clear the selection
*
* Calling code must invalidate the tree, restore the selection, and unsuppress selection events
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.refresh = Zotero.Promise.coroutine(function* ()
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.debug("Refreshing collections pane");
// Record open states before refreshing
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (this._rows) {
for (var i=0, len=this._rows.length; i<len; i++) {
var treeRow = this._rows[i];
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.ref && treeRow.ref.id == 'commons-header') {
var commonsExpand = this.isContainerOpen(i);
}
}
}
try {
this._containerState = JSON.parse(Zotero.Prefs.get("sourceList.persist"));
}
catch (e) {
this._containerState = {};
}
var userLibraryID = Zotero.Libraries.userLibraryID;
var readPref = function (pref) {
let ids = Zotero.Prefs.get(pref);
if (ids === "") {
this["_" + pref] = [];
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
}
else {
if (ids === undefined || typeof ids != 'string') {
ids = "" + userLibraryID;
Zotero.Prefs.set(pref, "" + userLibraryID);
}
this["_" + pref] = ids.split(',')
// Convert old id and convert to int
.map(id => id === "0" ? userLibraryID : parseInt(id));
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
}
}.bind(this);
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
if (this.hideSources.indexOf('duplicates') == -1) {
readPref('duplicateLibraries');
}
readPref('unfiledLibraries');
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var oldCount = this.rowCount || 0;
var newRows = [];
var added = 0;
//
// Add "My Library"
//
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('library', { libraryID: Zotero.Libraries.userLibraryID }),
added++
);
added += yield this._expandRow(newRows, 0);
this._addRowToArray(newRows, new Zotero.CollectionTreeRow('separator', false), added++);
// Add "My Publications"
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('publications', {
libraryID: Zotero.Libraries.publicationsLibraryID
}),
added++
);
// TODO: Unify feed and group adding code
// Add feeds
if (this.hideSources.indexOf('feeds') == -1) {
var feeds = Zotero.Feeds.getAll();
if (feeds.length) {
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('separator', false),
added++
);
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('header', {
id: "feed-libraries-header",
label: Zotero.getString('pane.collections.feedLibraries'),
libraryID: -1
}, 0),
added++
);
for (let feed of feeds) {
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('feed', feed),
added++
);
}
}
}
// Add groups
var groups = Zotero.Groups.getAll();
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
if (groups.length) {
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('separator', false),
added++
);
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('header', {
id: "group-libraries-header",
label: Zotero.getString('pane.collections.groupLibraries'),
libraryID: -1
}, 0),
added++
);
for (let group of groups) {
this._addRowToArray(
newRows,
new Zotero.CollectionTreeRow('group', group),
added++
);
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
2015-06-11 16:06:00 -04:00
this.selection.selectEventsSuppressed = true;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
this.selection.clearSelection();
this._rows = newRows;
this.rowCount = this._rows.length;
this._refreshRowMap();
2007-10-23 07:11:59 +00:00
var diff = this.rowCount - oldCount;
if (diff != 0) {
this._treebox.rowCountChanged(0, diff);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
/*
* Redisplay everything
*/
Zotero.CollectionTreeView.prototype.reload = function()
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return this.refresh()
.then(function () {
this._treebox.invalidate();
}.bind(this));
}
/**
* Select a row and wait for its items view to be created
*
* Note that this doesn't wait for the items view to be loaded. For that, add a 'load' event
* listener to the items view.
*
* @param {Integer} row
* @return {Promise}
*/
Zotero.CollectionTreeView.prototype.selectWait = Zotero.Promise.method(function (row) {
if (this.selection.selectEventsSuppressed) {
this.selection.select(row);
return;
}
if (this.selection.currentIndex == row) {
return;
};
var deferred = Zotero.Promise.defer();
this.addEventListener('select', () => deferred.resolve());
this.selection.select(row);
return deferred.promise;
});
/*
* Called by Zotero.Notifier on any changes to collections in the data layer
*/
Zotero.CollectionTreeView.prototype.notify = Zotero.Promise.coroutine(function* (action, type, ids, extraData) {
2015-02-03 11:57:32 -06:00
if (type == 'feed' && (action == 'unreadCountUpdated' || action == 'statusChanged')) {
for (let i=0; i<ids.length; i++) {
this._treebox.invalidateRow(this._rowMap['L' + ids[i]]);
}
return;
}
if ((!ids || ids.length == 0) && action != 'refresh' && action != 'redraw') {
2007-10-23 07:11:59 +00:00
return;
}
if (!this._rowMap) {
Zotero.debug("Row map didn't exist in collectionTreeView.notify()");
Initial Zotero 1.5 Megacommit Apologies for the massive (and, due to data_access.js splitting, difficult-to-follow) commit. Please note that external code that accesses the data layer may need to be tweaked for compatibility. Here's a comprehensive-as-possible changelog: - Added server sync functionality (incomplete) - Overhaul of data layer - Split data_access.js into separate files (item.js, items.js, creator.js, etc.) - Made creators and collections first-class objects, similar to items - Constructors now take id as first parameter, e.g. new Zotero.Item(1234, 'book'), to allow explicit id setting and id changing - Made various data layer operations (including attachment fields) require a save() rather than making direct DB changes - Better handling of unsaved objects - Item.setCreator() now takes creator objects instead of creator ids, and Item.save() will auto-save unsaved creators - clone() now works on unsaved objects - Newly created object instances are now disabled after save() to force refetch of globally accessible instance using Zotero.(Items|Creators|etc.).get() - Added secondary lookup key to data objects - Deprecated getID() and getItemType() methods in favor of .id and .itemTypeID properties - toArray() deprecated in favor of serialize(), which has a somewhat modified format - Added support for multiple creators with identical data -- currently unimplemented in interface and most of data layer - Added Item.diff() for comparing item metadata - Database changes - Added SQLite triggers to enforce foreign key constraints - Added Zotero.DB.transactionVacuum flag to run a VACUUM after a transaction - Added Zotero.DB.transactionDate, .transactionDateTime, and transactionTimestamp to retrieve consistent timestamps for entire transaction - Properly store 64-bit integers - Set PRAGMA locking_mode=EXCLUSIVE on database - Set SQLite page size to 4096 on new databases - Set SQLite page cache to 8MB - Do some database cleanup and integrity checking on migration from 1.0 branch - Removed IF NOT EXISTS from userdata.sql CREATE statements -- userdata.sql is now processed only on DB initialization - Removed itemNoteTitles table and moved titles into itemNotes - Abstracted metadata edit box and note box into flexible XBL bindings with various modes, including read-only states - Massive speed-up of item tree view - Several fixes from 1.0 branch for Fx3 compatibility - Added Notifier observer to log delete events for syncing - Zotero.Utilities changes - New methods getSQLDataType() and md5() - Removed onError from Zotero.Utilities.HTTP.doGet() - Don't display more than 1024 characters in doPost() debug output - Don't display passwords in doPost() debug output - Added Zotero.Notifier.untrigger() -- currently unused - Added Zotero.reloadDataObjects() to reset all in-memory objects - Added |chars| parameter to Zotero.randomString(len, chars) - Added Zotero.Date.getUnixTimestamp() and Date.toUnixTimestamp(JSDate) - Adjusted zotero-service.js to simplify file inclusion Various things (such as tags) are temporarily broken.
2008-05-04 08:32:48 +00:00
return;
}
if (!this.selection) {
Zotero.debug("Selection didn't exist in collectionTreeView.notify()");
return;
}
//
// Actions that don't change the selection
//
if (action == 'redraw') {
this._treebox.invalidate();
return;
}
if (action == 'refresh') {
// If trash is refreshed, we probably need to update the icon from full to empty
if (type == 'trash') {
// libraryID is passed as parameter to 'refresh'
let deleted = yield Zotero.Items.getDeleted(ids[0], true);
this._trashNotEmpty[ids[0]] = !!deleted.length;
let row = this.getRowIndexByID("T" + ids[0]);
this._treebox.invalidateRow(row);
}
return;
}
//
// Actions that can change the selection
//
var currentTreeRow = this.getRow(this.selection.currentIndex);
2007-10-23 07:11:59 +00:00
this.selection.selectEventsSuppressed = true;
// If there's not at least one new collection to be selected, get a scroll position to restore later
var scrollPosition = false;
if (action != 'add' || ids.every(id => extraData[id] && extraData[id].skipSelect)) {
scrollPosition = this._saveScrollPosition();
}
if (action == 'delete') {
var selectedIndex = this.selection.count ? this.selection.currentIndex : 0;
// Since a delete involves shifting of rows, we have to do it in reverse order
let rows = [];
for (let i = 0; i < ids.length; i++) {
let id = ids[i];
switch (type) {
case 'collection':
if (this._rowMap['C' + id] !== undefined) {
rows.push(this._rowMap['C' + id]);
}
break;
case 'search':
if (this._rowMap['S' + id] !== undefined) {
rows.push(this._rowMap['S' + id]);
}
break;
case 'feed':
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
case 'group':
let row = this.getRowIndexByID("L" + extraData[id].libraryID);
let level = this.getLevel(row);
do {
rows.push(row);
row++;
}
while (row < this.rowCount && this.getLevel(row) > level);
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
break;
}
}
if (rows.length > 0) {
rows.sort(function(a,b) { return a-b });
for (let i = rows.length - 1; i >= 0; i--) {
let row = rows[i];
this._removeRow(row);
}
this._refreshRowMap();
2007-10-23 07:11:59 +00:00
}
if (!this.selection.count) {
// If last row was selected, stay on the last row
if (selectedIndex >= this.rowCount) {
selectedIndex = this.rowCount - 1;
};
this.selection.select(selectedIndex)
}
// Make sure the selection doesn't land on a separator (e.g. deleting last feed)
let index = this.selection.currentIndex;
while (index >= 0 && !this.isSelectable(index)) {
// move up, since we got shifted down
index--;
}
if (index >= 0) {
this.selection.select(index);
} else {
this.selection.clearSelection();
}
}
else if (action == 'modify') {
let row;
let id = ids[0];
let rowID = "C" + id;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
switch (type) {
case 'collection':
row = this.getRowIndexByID(rowID);
if (row !== false) {
// TODO: Only move if name changed
let reopen = this.isContainerOpen(row);
if (reopen) {
this._closeContainer(row);
}
this._removeRow(row);
yield this._addSortedRow('collection', id);
if (!extraData[id].skipSelect) {
yield this.selectByID(currentTreeRow.id);
if (reopen) {
let newRow = this.getRowIndexByID(rowID);
if (!this.isContainerOpen(newRow)) {
yield this.toggleOpenState(newRow);
}
}
}
2007-10-23 07:11:59 +00:00
}
break;
2007-10-23 07:11:59 +00:00
case 'search':
row = this.getRowIndexByID("S" + id);
if (row !== false) {
// TODO: Only move if name changed
this._removeRow(row);
yield this._addSortedRow('search', id);
yield this.selectByID(currentTreeRow.id);
}
break;
default:
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
yield this.reload();
yield this.selectByID(currentTreeRow.id);
break;
2010-03-29 03:44:26 +00:00
}
}
else if(action == 'add')
{
// skipSelect isn't necessary if more than one object
let selectRow = ids.length == 1 && (!extraData[ids[0]] || !extraData[ids[0]].skipSelect);
for (let id of ids) {
switch (type) {
case 'collection':
case 'search':
yield this._addSortedRow(type, id);
if (selectRow) {
if (type == 'collection') {
yield this.selectCollection(id);
}
else if (type == 'search') {
yield this.selectSearch(id);
}
}
break;
case 'group':
if (ids.length != 1) {
Zotero.logError("WARNING: Multiple groups shouldn't currently be added "
+ "together in collectionTreeView::notify()")
}
yield this.reload();
yield this.selectByID(currentTreeRow.id);
break;
case 'feed':
yield this.reload();
yield this.selectByID("L" + id);
break;
}
}
}
this._rememberScrollPosition(scrollPosition);
var deferred = Zotero.Promise.defer();
this.addEventListener('select', () => deferred.resolve());
2007-10-23 07:11:59 +00:00
this.selection.selectEventsSuppressed = false;
return deferred.promise;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
/**
* Add a row in the appropriate place
*
* This only adds a row if it would be visible without opening any containers
*
* @param {String} objectType
* @param {Integer} id - collectionID
* @return {Integer|false} - Index at which the row was added, or false if it wasn't added
*/
Zotero.CollectionTreeView.prototype._addSortedRow = Zotero.Promise.coroutine(function* (objectType, id) {
let beforeRow;
if (objectType == 'collection') {
let collection = yield Zotero.Collections.getAsync(id);
let parentID = collection.parentID;
// If parent isn't visible, don't add
if (parentID && this._rowMap["C" + parentID] === undefined) {
return false;
}
let libraryID = collection.libraryID;
let startRow;
if (parentID) {
startRow = this._rowMap["C" + parentID];
}
else {
startRow = this._rowMap['L' + libraryID];
}
// If container isn't open, don't add
if (!this.isContainerOpen(startRow)) {
return false;
}
let level = this.getLevel(startRow) + 1;
// If container is empty, just add after
if (this.isContainerEmpty(startRow)) {
beforeRow = startRow + 1;
}
else {
// Get all collections at the same level that don't have a different parent
startRow++;
loop:
for (let i = startRow; i < this.rowCount; i++) {
let treeRow = this.getRow(i);
beforeRow = i;
// Since collections come first, if we reach something that's not a collection,
// stop
if (!treeRow.isCollection()) {
break;
}
let rowLevel = this.getLevel(i);
if (rowLevel < level) {
break;
}
else {
// Fast forward through subcollections
while (rowLevel > level) {
beforeRow = ++i;
if (i == this.rowCount || !this.getRow(i).isCollection()) {
break loop;
}
treeRow = this.getRow(i);
rowLevel = this.getLevel(i);
}
if (Zotero.localeCompare(treeRow.ref.name, collection.name) > 0) {
break;
}
}
}
}
this._addRow(
new Zotero.CollectionTreeRow('collection', collection, level),
beforeRow
);
}
else if (objectType == 'search') {
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
let search = Zotero.Searches.get(id);
let libraryID = search.libraryID;
let startRow = this._rowMap['L' + libraryID];
// If container isn't open, don't add
if (!this.isContainerOpen(startRow)) {
return false;
}
let level = this.getLevel(startRow) + 1;
// If container is empty, just add after
if (this.isContainerEmpty(startRow)) {
beforeRow = startRow + 1;
}
else {
startRow++;
var inSearches = false;
for (let i = startRow; i < this.rowCount; i++) {
let treeRow = this.getRow(i);
beforeRow = i;
// If we've reached something other than collections, stop
if (treeRow.isSearch()) {
// If current search sorts after, stop
if (Zotero.localeCompare(treeRow.ref.name, search.name) > 0) {
break;
}
}
// If it's not a search and it's not a collection, stop
else if (!treeRow.isCollection()) {
break;
}
}
}
this._addRow(
new Zotero.CollectionTreeRow('search', search, level),
beforeRow
);
}
return beforeRow;
});
2007-10-23 07:11:59 +00:00
/*
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
* Set the rows that should be highlighted -- actual highlighting is done
2007-10-23 07:11:59 +00:00
* by getRowProperties based on the array set here
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.setHighlightedRows = Zotero.Promise.coroutine(function* (ids) {
2007-10-23 07:11:59 +00:00
this._highlightedRows = {};
this._treebox.invalidate();
if (!ids || !ids.length) {
return;
}
// Make sure all highlighted collections are shown
for (let id of ids) {
if (id[0] == 'C') {
id = id.substr(1);
yield this.expandToCollection(id);
}
}
// Highlight rows
var rows = [];
for (let id of ids) {
2015-11-01 21:22:13 -05:00
let row = this._rowMap[id];
this._highlightedRows[row] = true;
this._treebox.invalidateRow(row);
rows.push(row);
}
rows.sort();
var firstRow = this._treebox.getFirstVisibleRow();
var lastRow = this._treebox.getLastVisibleRow();
var scrolled = false;
for (let row of rows) {
// If row is visible, stop
if (row >= firstRow && row <= lastRow) {
scrolled = true;
break;
}
2007-10-23 07:11:59 +00:00
}
// Select first collection
// TODO: Select closest? Select a few rows above or below?
if (!scrolled) {
this._treebox.ensureRowIsVisible(rows[0]);
2007-10-23 07:11:59 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2007-10-23 07:11:59 +00:00
/*
* Unregisters view from Zotero.Notifier (called on window close)
*/
Zotero.CollectionTreeView.prototype.unregister = function()
{
2007-10-23 07:11:59 +00:00
Zotero.Notifier.unregisterObserver(this._unregisterID);
}
////////////////////////////////////////////////////////////////////////////////
///
/// nsITreeView functions
/// http://www.xulplanet.com/references/xpcomref/ifaces/nsITreeView.html
///
////////////////////////////////////////////////////////////////////////////////
Zotero.CollectionTreeView.prototype.getCellText = function(row, column)
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var obj = this.getRow(row);
if (column.id == 'zotero-collections-name-column') {
return obj.getName();
}
else
return "";
}
Zotero.CollectionTreeView.prototype.getImageSrc = function(row, col)
{
var suffix = Zotero.hiDPISuffix;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row);
var collectionType = treeRow.type;
if (collectionType == 'group') {
collectionType = 'library';
}
// Show sync icons only in library rows
if (collectionType != 'library' && col.index != 0) {
return '';
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
switch (collectionType) {
case 'library':
case 'feed':
// Better alternative needed: https://github.com/zotero/zotero/pull/902#issuecomment-183185973
/*
2015-02-03 11:57:32 -06:00
if (treeRow.ref.updating) {
collectionType += '-updating';
} else */if (treeRow.ref.lastCheckError) {
2015-02-03 11:57:32 -06:00
collectionType += '-error';
}
break;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
case 'trash':
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (this._trashNotEmpty[treeRow.ref.libraryID]) {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
collectionType += '-full';
}
break;
case 'header':
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.ref.id == 'group-libraries-header') {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
collectionType = 'groups';
}
2015-02-03 11:57:32 -06:00
else if (treeRow.ref.id == 'feed-libraries-header') {
collectionType = 'feedLibrary';
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
else if (treeRow.ref.id == 'commons-header') {
collectionType = 'commons';
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
break;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
collectionType = 'library';
break;
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
case 'collection':
case 'search':
if (Zotero.isMac) {
return "chrome://zotero-platform/content/treesource-" + collectionType + ".png";
}
break;
case 'publications':
return "chrome://zotero/skin/treeitem-journalArticle" + suffix + ".png";
}
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
return "chrome://zotero/skin/treesource-" + collectionType + suffix + ".png";
}
Zotero.CollectionTreeView.prototype.isContainer = function(row)
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row);
return treeRow.isLibrary(true) || treeRow.isCollection() || treeRow.isPublications() || treeRow.isBucket();
}
2007-10-23 07:11:59 +00:00
/*
* Returns true if the collection has no child collections
*/
Zotero.CollectionTreeView.prototype.isContainerEmpty = function(row)
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row);
if (treeRow.isLibrary()) {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isBucket()) {
return true;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isGroup()) {
var libraryID = treeRow.ref.libraryID;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return !treeRow.ref.hasCollections()
&& !treeRow.ref.hasSearches()
&& this._duplicateLibraries.indexOf(libraryID) == -1
&& this._unfiledLibraries.indexOf(libraryID) == -1
&& this.hideSources.indexOf('trash') != -1;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isCollection()) {
return !treeRow.ref.hasChildCollections();
2007-10-23 07:11:59 +00:00
}
return true;
}
Zotero.CollectionTreeView.prototype.getParentIndex = function(row)
{
var thisLevel = this.getLevel(row);
if(thisLevel == 0) return -1;
for(var i = row - 1; i >= 0; i--)
if(this.getLevel(i) < thisLevel)
return i;
return -1;
}
Zotero.CollectionTreeView.prototype.hasNextSibling = function(row, afterIndex)
{
var thisLevel = this.getLevel(row);
for(var i = afterIndex + 1; i < this.rowCount; i++)
{
var nextLevel = this.getLevel(i);
if(nextLevel == thisLevel) return true;
else if(nextLevel < thisLevel) return false;
}
}
/*
* Opens/closes the specified row
*/
Zotero.CollectionTreeView.prototype.toggleOpenState = Zotero.Promise.coroutine(function* (row) {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
if (this.isContainerOpen(row)) {
return this._closeContainer(row);
}
var count = 0;
var treeRow = this.getRow(row);
if (treeRow.isLibrary(true) || treeRow.isCollection()) {
count = yield this._expandRow(this._rows, row, true);
}
this.rowCount += count;
this._treebox.rowCountChanged(row + 1, count);
this._rows[row].isOpen = true;
this._treebox.invalidateRow(row);
this._refreshRowMap();
this._startRememberOpenStatesTimer();
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2007-10-23 07:11:59 +00:00
Zotero.CollectionTreeView.prototype._closeContainer = function (row) {
if (!this.isContainerOpen(row)) return;
var count = 0;
var level = this.getLevel(row);
var nextRow = row + 1;
// Remove child rows
while ((nextRow < this._rows.length) && (this.getLevel(nextRow) > level)) {
this._removeRow(nextRow);
count--;
}
this._rows[row].isOpen = false;
this._treebox.invalidateRow(row);
this._refreshRowMap();
this._startRememberOpenStatesTimer();
}
/**
* After a short delay, persist the open states of the tree, or if already queued, cancel and requeue.
* This avoids repeated saving while opening or closing multiple rows.
*/
Zotero.CollectionTreeView.prototype._startRememberOpenStatesTimer = function () {
if (this._rememberOpenStatesTimeoutID) {
clearTimeout(this._rememberOpenStatesTimeoutID);
}
this._rememberOpenStatesTimeoutID = setTimeout(() => {
this._rememberOpenStates();
this._rememberOpenStatesTimeoutID = null;
}, 250)
};
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Zotero.CollectionTreeView.prototype.isSelectable = function (row, col) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row);
switch (treeRow.type) {
case 'separator':
case 'header':
return false;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
return true;
}
/**
* Tree method for whether to allow inline editing (not to be confused with this.editable)
*/
Zotero.CollectionTreeView.prototype.isEditable = function (row, col) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return this.selectedTreeRow.isCollection() && this.editable;
}
Zotero.CollectionTreeView.prototype.setCellText = function (row, col, val) {
val = val.trim();
if (val === "") {
return;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row);
treeRow.ref.name = val;
treeRow.ref.saveTx();
}
/**
* Returns TRUE if the underlying view is editable
*/
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Zotero.CollectionTreeView.prototype.__defineGetter__('editable', function () {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return this.getRow(this.selection.currentIndex).editable;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
});
/**
* @param {Integer} libraryID
* @param {Boolean} [recursive=false] - Expand all collections and subcollections
*/
Zotero.CollectionTreeView.prototype.expandLibrary = Zotero.Promise.coroutine(function* (libraryID, recursive) {
var row = this._rowMap['L' + libraryID]
if (row === undefined) {
return false;
}
if (!this.isContainerOpen(row)) {
yield this.toggleOpenState(row);
}
if (recursive) {
for (let i = row; i < this.rowCount && this.getRow(i).ref.libraryID == libraryID; i++) {
if (this.isContainer(i) && !this.isContainerOpen(i)) {
yield this.toggleOpenState(i);
}
}
}
return true;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
Zotero.CollectionTreeView.prototype.collapseLibrary = function (libraryID) {
var row = this._rowMap['L' + libraryID]
if (row === undefined) {
return false;
2007-10-23 07:11:59 +00:00
}
var closed = [];
var found = false;
for (let i = this.rowCount - 1; i >= row; i--) {
let treeRow = this.getRow(i);
if (treeRow.ref.libraryID !== libraryID) {
// Once we've moved beyond the original library, stop looking
if (found) {
break;
}
continue;
}
found = true;
if (this.isContainer(i) && this.isContainerOpen(i)) {
closed.push(treeRow.id);
this._closeContainer(i);
}
}
// Select the collapsed library
this.selection.select(row);
// We have to manually delete closed rows from the container state object, because otherwise
// _rememberOpenStates() wouldn't see any of the rows under the library (since the library is now
// collapsed) and they'd remain as open in the persisted object.
closed.forEach(id => { delete this._containerState[id]; });
this._rememberOpenStates();
return true;
};
2007-10-23 07:11:59 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.expandToCollection = Zotero.Promise.coroutine(function* (collectionID) {
var col = yield Zotero.Collections.getAsync(collectionID);
2008-12-29 07:57:20 +00:00
if (!col) {
Zotero.debug("Cannot expand to nonexistent collection " + collectionID, 2);
return false;
}
// Open library if closed
var libraryRow = this._rowMap['L' + col.libraryID];
if (!this.isContainerOpen(libraryRow)) {
yield this.toggleOpenState(libraryRow);
}
var row = this._rowMap["C" + collectionID];
if (row !== undefined) {
2008-12-29 07:57:20 +00:00
return true;
}
var path = [];
var parentID;
while (parentID = col.parentID) {
path.unshift(parentID);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
col = yield Zotero.Collections.getAsync(parentID);
2008-12-29 07:57:20 +00:00
}
2016-05-03 23:51:12 +02:00
for (let id of path) {
row = this._rowMap["C" + id];
2008-12-29 07:57:20 +00:00
if (!this.isContainerOpen(row)) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
yield this.toggleOpenState(row);
2008-12-29 07:57:20 +00:00
}
}
return true;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2008-12-29 07:57:20 +00:00
2007-10-23 07:11:59 +00:00
////////////////////////////////////////////////////////////////////////////////
///
/// Additional functions for managing data in the tree
///
////////////////////////////////////////////////////////////////////////////////
Zotero.CollectionTreeView.prototype.selectByID = Zotero.Promise.coroutine(function* (id) {
var type = id[0];
id = ('' + id).substr(1);
switch (type) {
case 'L':
return yield this.selectLibrary(id);
case 'C':
var found = yield this.expandToCollection(id);
break;
case 'S':
var search = yield Zotero.Searches.getAsync(id);
var found = yield this.expandLibrary(search.libraryID);
break;
2015-05-24 22:04:40 -04:00
case 'T':
return yield this.selectTrash(id);
2016-03-12 05:03:14 -05:00
}
var row = this._rowMap[type + id];
if (!row) {
return false;
}
this._treebox.ensureRowIsVisible(row);
yield this.selectWait(row);
return true;
});
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
/**
* @param {Integer} libraryID Library to select
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
*/
Zotero.CollectionTreeView.prototype.selectLibrary = Zotero.Promise.coroutine(function* (libraryID) {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
// Select local library
if (!libraryID) {
this._treebox.ensureRowIsVisible(0);
yield this.selectWait(0);
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
return true;
}
// Check if library is already selected
if (this.selection && this.selection.count && this.selection.currentIndex != -1) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(this.selection.currentIndex);
if (treeRow.isLibrary(true) && treeRow.ref.libraryID == libraryID) {
this._treebox.ensureRowIsVisible(this.selection.currentIndex);
return true;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
// Find library
var row = this._rowMap['L' + libraryID];
if (row !== undefined) {
this._treebox.ensureRowIsVisible(row);
yield this.selectWait(row);
return true;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
return false;
});
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
Zotero.CollectionTreeView.prototype.selectCollection = function (id) {
return this.selectByID('C' + id);
}
Zotero.CollectionTreeView.prototype.selectSearch = function (id) {
return this.selectByID('S' + id);
}
Zotero.CollectionTreeView.prototype.selectTrash = Zotero.Promise.coroutine(function* (libraryID) {
// Check if trash is already selected
if (this.selection && this.selection.count && this.selection.currentIndex != -1) {
2015-05-24 22:04:40 -04:00
let itemGroup = this.getRow(this.selection.currentIndex);
if (itemGroup.isTrash() && itemGroup.ref.libraryID == libraryID) {
this._treebox.ensureRowIsVisible(this.selection.currentIndex);
return true;
}
}
// Find library trash
for (let i = 0; i < this.rowCount; i++) {
2015-05-24 22:04:40 -04:00
let itemGroup = this.getRow(i);
2015-05-24 22:04:40 -04:00
// If library is closed, open it
if (itemGroup.isLibrary(true) && itemGroup.ref.libraryID == libraryID
&& !this.isContainerOpen(i)) {
yield this.toggleOpenState(i);
continue;
}
if (itemGroup.isTrash() && itemGroup.ref.libraryID == libraryID) {
this._treebox.ensureRowIsVisible(i);
this.selection.select(i);
return true;
}
}
return false;
});
/*
* Delete the selection
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.deleteSelection = Zotero.Promise.coroutine(function* (deleteItems)
{
if(this.selection.count == 0)
return;
//collapse open collections
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
for (let i=0; i<this.rowCount; i++) {
if (this.selection.isSelected(i) && this.isContainer(i)) {
this._closeContainer(i);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
}
this._refreshRowMap();
//create an array of collections
var rows = new Array();
var start = new Object();
var end = new Object();
for (var i=0, len=this.selection.getRangeCount(); i<len; i++)
{
this.selection.getRangeAt(i,start,end);
for (var j=start.value; j<=end.value; j++)
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if(!this.getRow(j).isLibrary())
rows.push(j);
}
//iterate and erase...
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
//this._treebox.beginUpdateBatch();
for (var i=0; i<rows.length; i++)
{
//erase collection from DB:
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(rows[i]-i);
if (treeRow.isCollection() || treeRow.isFeed()) {
yield treeRow.ref.eraseTx({
deleteItems: true
});
}
if (treeRow.isCollection() || treeRow.isFeed()) {
yield treeRow.ref.erase(deleteItems);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
else if (treeRow.isSearch()) {
yield Zotero.Searches.erase(treeRow.ref.id);
}
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
//this._treebox.endUpdateBatch();
if (end.value < this.rowCount) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var row = this.getRow(end.value);
if (row.isSeparator()) {
return;
}
this.selection.select(end.value);
}
else {
this.selection.select(this.rowCount-1);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
/**
* Expand row based on last state
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype._expandRow = Zotero.Promise.coroutine(function* (rows, row, forceOpen) {
var treeRow = rows[row];
var level = rows[row].level;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var isLibrary = treeRow.isLibrary(true);
var isCollection = treeRow.isCollection();
var libraryID = treeRow.ref.libraryID;
if (treeRow.isPublications() || treeRow.isFeed()) {
return false;
}
if (isLibrary) {
var collections = Zotero.Collections.getByLibrary(libraryID, treeRow.ref.id);
}
else if (isCollection) {
var collections = Zotero.Collections.getByParent(treeRow.ref.id);
}
if (isLibrary) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var savedSearches = yield Zotero.Searches.getAll(libraryID);
var showDuplicates = (this.hideSources.indexOf('duplicates') == -1
&& this._duplicateLibraries.indexOf(libraryID) != -1);
var showUnfiled = this._unfiledLibraries.indexOf(libraryID) != -1;
var showTrash = this.hideSources.indexOf('trash') == -1;
}
else {
var savedSearches = [];
var showDuplicates = false;
var showUnfiled = false;
var showTrash = false;
}
// If not a manual open and either the library is set to be collapsed or this is a collection that isn't explicitly opened,
// set the initial state to closed
if (!forceOpen &&
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
(this._containerState[treeRow.id] === false
|| (isCollection && !this._containerState[treeRow.id]))) {
rows[row].isOpen = false;
return 0;
}
var startOpen = !!(collections.length || savedSearches.length || showDuplicates || showUnfiled || showTrash);
// If this isn't a manual open, set the initial state depending on whether
// there are child nodes
if (!forceOpen) {
rows[row].isOpen = startOpen;
}
if (!startOpen) {
return 0;
}
var newRows = 0;
// Add collections
for (var i = 0, len = collections.length; i < len; i++) {
let beforeRow = row + 1 + newRows;
this._addRowToArray(
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
rows,
new Zotero.CollectionTreeRow('collection', collections[i], level + 1),
beforeRow
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
);
newRows++;
// Recursively expand child collections that should be open
newRows += yield this._expandRow(rows, beforeRow);
}
if (isCollection) {
return newRows;
}
// Add searches
for (var i = 0, len = savedSearches.length; i < len; i++) {
this._addRowToArray(
rows,
new Zotero.CollectionTreeRow('search', savedSearches[i], level + 1),
row + 1 + newRows
);
newRows++;
}
// Duplicate items
if (showDuplicates) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
let d = new Zotero.Duplicates(libraryID);
this._addRowToArray(
rows,
new Zotero.CollectionTreeRow('duplicates', d, level + 1),
row + 1 + newRows
);
newRows++;
}
// Unfiled items
if (showUnfiled) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
let s = new Zotero.Search;
s.libraryID = libraryID;
s.name = Zotero.getString('pane.collections.unfiled');
s.addCondition('libraryID', 'is', libraryID);
s.addCondition('unfiled', 'true');
this._addRowToArray(
rows,
new Zotero.CollectionTreeRow('unfiled', s, level + 1),
row + 1 + newRows
);
newRows++;
}
if (showTrash) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
let deletedItems = yield Zotero.Items.getDeleted(libraryID);
if (deletedItems.length || Zotero.Prefs.get("showTrashWhenEmpty")) {
var ref = {
libraryID: libraryID
};
this._addRowToArray(
rows,
new Zotero.CollectionTreeRow('trash', ref, level + 1),
row + 1 + newRows
);
newRows++;
}
this._trashNotEmpty[libraryID] = !!deletedItems.length;
}
return newRows;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
/**
* Returns libraryID or FALSE if not a library
*/
Zotero.CollectionTreeView.prototype.getSelectedLibraryID = function() {
if (!this.selection || !this.selection.count || this.selection.currentIndex == -1) return false;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(this.selection.currentIndex);
return treeRow && treeRow.ref && treeRow.ref.libraryID !== undefined
&& treeRow.ref.libraryID;
}
Zotero.CollectionTreeView.prototype.getSelectedCollection = function(asID) {
if (this.selection
&& this.selection.count > 0
&& this.selection.currentIndex != -1) {
var collection = this.getRow(this.selection.currentIndex);
if (collection && collection.isCollection()) {
return asID ? collection.ref.id : collection.ref;
}
}
return false;
}
Duplicate detection: - Adds a per-library "Duplicate Items" virtual search to the source list -- shows up by default for "My Library" but can be added to and removed from all libraries - Current matching algorithm is very basic: finds exact title matches (after normalizing case/diacritics/punctuation/spacing) and DOI/ISBN matches (untested) - In duplicates view, sets are selected automatically; in other views, duplicate items can be selected manually and the merge interface can be brought up with "Merge Items" in the context menu - Can select a master item and individual fields to merge from other versions - Word processor integration code will automatically find mapped replacements and update documents with new item keys Possible future improvements: - Improved detection algorithms - UI tweaks - Currently if any items differ, all available versions will be shown as master item options, even if only one item is different; probably the earliest equivalent item should be shown for each distinct version - Caching of results for performance - Confidence scale - Creator version selection (currently the creators from the chosen master item are kept) - Merging of matching child items - Better sorting of duplicates if not clustered together by the selected sort column - Relation path compression when merging items that are already mapped to previously removed duplicates Other changes in this commit: - Don't show Trash in word processor integration windows - Consider items in trash to be missing in word processor documents - Selection of special views (Trash, Unfiled, Duplicates) is now restored properly in new windows - Disabled field transform context menu when item isn't editable - Left/right arrow now expands/collapses all selected items instead of just the last-selected row - Relation deletions are now synced - The same items row is now reselected after item deletion - (dev) Zotero.Item.getNotes(), Zotero.Item.getAttachments(), and Zotero.Item.getTags() now return empty arrays rather than FALSE if no matches -- tests on those return values in third-party code will need to be changed - (dev) New function Zotero.Utilities.removeDiacritics(str, lowercaseOnly) -- could be used to generate ASCII BibTeX keys - (dev) New 'tempTable' search condition can take a table to join against -- useful for implementing virtual source lists - (dev) Significant UI code cleanup - (dev) Moved all item pane content into itemPane.xul - Probably various other things Needless to say, this needs testing.
2011-07-22 21:24:38 +00:00
/**
* Creates mapping of item group ids to tree rows
*/
Zotero.CollectionTreeView.prototype._refreshRowMap = function() {
this._rowMap = {};
for (let i = 0, len = this.rowCount; i < len; i++) {
this._rowMap[this.getRow(i).id] = i;
}
}
/**
* Persist the current open/closed state of rows to a pref
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype._rememberOpenStates = Zotero.Promise.coroutine(function* () {
var state = this._containerState;
// Every so often, remove obsolete rows
if (Math.random() < 1/20) {
Zotero.debug("Purging sourceList.persist");
for (var id in state) {
var m = id.match(/^C([0-9]+)$/);
if (m) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!(yield Zotero.Collections.getAsync(m[1]))) {
delete state[id];
}
continue;
}
var m = id.match(/^G([0-9]+)$/);
if (m) {
if (!Zotero.Groups.get(m[1])) {
delete state[id];
}
continue;
}
}
}
for (var i = 0, len = this.rowCount; i < len; i++) {
if (!this.isContainer(i)) {
continue;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(i);
if (!treeRow.id) {
continue;
}
var open = this.isContainerOpen(i);
// Collections and feeds default to closed
if ((!open && treeRow.isCollection()) || treeRow.isFeed()) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
delete state[treeRow.id];
continue;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
state[treeRow.id] = open;
}
this._containerState = state;
Zotero.Prefs.set("sourceList.persist", JSON.stringify(state));
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
////////////////////////////////////////////////////////////////////////////////
///
/// Command Controller:
/// for Select All, etc.
///
////////////////////////////////////////////////////////////////////////////////
Zotero.CollectionTreeCommandController = function(tree)
{
this.tree = tree;
}
Zotero.CollectionTreeCommandController.prototype.supportsCommand = function(cmd)
{
}
Zotero.CollectionTreeCommandController.prototype.isCommandEnabled = function(cmd)
{
}
Zotero.CollectionTreeCommandController.prototype.doCommand = function(cmd)
{
}
Zotero.CollectionTreeCommandController.prototype.onEvent = function(evt)
{
}
////////////////////////////////////////////////////////////////////////////////
///
/// Drag-and-drop functions:
/// canDrop() and drop() are for nsITreeView
/// onDragStart() and onDrop() are for HTML 5 Drag and Drop
///
////////////////////////////////////////////////////////////////////////////////
2008-11-30 20:18:48 +00:00
/*
* Start a drag using HTML 5 Drag and Drop
2008-11-30 20:18:48 +00:00
*/
Zotero.CollectionTreeView.prototype.onDragStart = function(event) {
// See note in LibraryTreeView::_setDropEffect()
if (Zotero.isWin) {
event.dataTransfer.effectAllowed = 'move';
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.selectedTreeRow;
if (!treeRow.isCollection()) {
return;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
event.dataTransfer.setData("zotero/collection", treeRow.ref.id);
2008-11-30 20:18:48 +00:00
}
/**
* Called by treechildren.onDragOver() before setting the dropEffect,
* which is checked in libraryTreeView.canDrop()
*/
Zotero.CollectionTreeView.prototype.canDropCheck = function (row, orient, dataTransfer) {
2008-11-30 20:18:48 +00:00
//Zotero.debug("Row is " + row + "; orient is " + orient);
var dragData = Zotero.DragDrop.getDataFromDataTransfer(dataTransfer);
2008-11-30 20:18:48 +00:00
if (!dragData) {
Zotero.debug("No drag data");
return false;
}
2008-11-30 20:18:48 +00:00
var dataType = dragData.dataType;
var data = dragData.data;
// Empty space below rows
if (row == -1) {
return false;
}
2007-10-23 07:11:59 +00:00
// For dropping collections onto root level
if (orient == 1 && row == 0 && dataType == 'zotero/collection') {
return true;
}
// Directly on a row
else if (orient == 0) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var treeRow = this.getRow(row); //the collection we are dragging over
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (dataType == 'zotero/item' && treeRow.isBucket()) {
return true;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!treeRow.editable) {
Zotero.debug("Drop target not editable");
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
return false;
}
if (treeRow.isFeed()) {
Zotero.debug("Cannot drop into feeds");
return false;
}
2007-10-23 07:11:59 +00:00
if (dataType == 'zotero/item') {
2008-11-30 20:18:48 +00:00
var ids = data;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
var items = Zotero.Items.get(ids);
var skip = true;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
for each(var item in items) {
// Can only drag top-level items
if (!item.isTopLevelItem()) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.debug("Can't drag child item");
return false;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isWithinGroup() && item.isAttachment()) {
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
// Linked files can't be added to groups
if (item.attachmentLinkMode == Zotero.Attachments.LINK_MODE_LINKED_FILE) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.debug("Linked files cannot be added to groups");
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!treeRow.filesEditable) {
Zotero.debug("Drop target does not allow files to be edited");
return false;
}
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
skip = false;
continue;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
if (treeRow.isPublications() && treeRow.ref.libraryID != item.libraryID) {
if (item.isAttachment() || item.isNote()) {
Zotero.debug("Top-level attachments and notes cannot be added to My Publications");
return false;
}
if(item instanceof Zotero.FeedItem) {
Zotero.debug("FeedItems cannot be added to My Publications");
return false;
}
skip = false;
continue;
}
// Cross-library drag
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.ref.libraryID != item.libraryID) {
// Only allow cross-library drag to root library and collections
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!(treeRow.isLibrary(true) || treeRow.isCollection())) {
Zotero.debug("Cross-library drag to non-collection not allowed");
return false;
}
skip = false;
continue;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
// Intra-library drag
// Don't allow drag onto root of same library
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isLibrary(true)) {
Zotero.debug("Can't drag into same library root");
return false;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Allow drags to collections. Item collection membership is an asynchronous
// check, so we do that on drop()
if (treeRow.isCollection()) {
skip = false;
}
}
if (skip) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.debug("Drag skipped");
return false;
}
return true;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
else if (dataType == 'text/x-moz-url' || dataType == 'application/x-moz-file') {
if (treeRow.isSearch() || treeRow.isPublications()) {
2007-10-23 07:11:59 +00:00
return false;
}
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
if (dataType == 'application/x-moz-file') {
// Don't allow folder drag
if (data[0].isDirectory()) {
return false;
}
// Don't allow drop if no permissions
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!treeRow.filesEditable) {
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
return false;
}
2007-10-23 07:11:59 +00:00
}
Zotero File Storage megacommit - Group file sync via Zotero File Storage - Split file syncing into separate modules for ZFS and WebDAV - Dragging items between libraries copies child notes, snapshots/files, and links based on checkboxes for each (enabled by default) in the Zotero preferences - Sync errors now trigger an exclamation/error icon separate from the sync icon, with a popup window displaying the error and an option to report it - Various errors that could cause perpetual sync icon spinning now stop the sync properly - Zotero.Utilities.md5(str) is now md5(strOrFile, base64) - doPost(), doHead(), and retrieveSource() now takes a headers parameter instead of requestContentType - doHead() can now accept an nsIURI (with login credentials), is a background request, and isn't cached - When library access or file writing access is denied during sync, display a warning and then reset local group to server version - Perform additional steps (e.g., removing local groups) when switching sync users to prevent errors - Compare hash as well as mod time when checking for modified local files - Don't trigger notifications when removing groups from the client - Clear relation links to items in removed groups - Zotero.Item.attachmentHash property to get file MD5 - importFromFile() now takes libraryID as a third parameter - Zotero.Attachments.getNumFiles() returns the number of files in the attachment directory - Zotero.Attachments.copyAttachmentToLibrary() copies an attachment item, including files, to another library - Removed Zotero.File.getFileHash() in favor of updated Zotero.Utilities.md5() - Zotero.File.copyDirectory(dir, newDir) copies all files from dir into newDir - Preferences shuffling: OpenURL to Advanced, import/export character set options to Export, "Include URLs of paper articles in references" to Styles - Other stuff I don't remember Suffice it to say, this could use testing.
2009-09-13 07:23:29 +00:00
2007-10-23 07:11:59 +00:00
return true;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
else if (dataType == 'zotero/collection') {
if (treeRow.isPublications()) {
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
let draggedCollectionID = data[0];
let draggedCollection = Zotero.Collections.get(draggedCollectionID);
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.ref.libraryID == draggedCollection.libraryID) {
// Collections cannot be dropped on themselves
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (draggedCollectionID == treeRow.ref.id) {
return false;
}
// Nor in their children
if (draggedCollection.hasDescendent('collection', treeRow.ref.id)) {
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
return false;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
// Dragging a collection to a different library
else {
// Allow cross-library drag only to root library and collections
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!treeRow.isLibrary(true) && !treeRow.isCollection()) {
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
return true;
}
}
return false;
};
/**
* Perform additional asynchronous drop checks
*
* Called by treechildren.drop()
*/
Zotero.CollectionTreeView.prototype.canDropCheckAsync = Zotero.Promise.coroutine(function* (row, orient, dataTransfer) {
//Zotero.debug("Row is " + row + "; orient is " + orient);
var dragData = Zotero.DragDrop.getDataFromDataTransfer(dataTransfer);
if (!dragData) {
Zotero.debug("No drag data");
return false;
}
var dataType = dragData.dataType;
var data = dragData.data;
if (orient == 0) {
var treeRow = this.getRow(row); //the collection we are dragging over
if (dataType == 'zotero/item' && treeRow.isBucket()) {
return true;
}
if (dataType == 'zotero/item') {
var ids = data;
var items = Zotero.Items.get(ids);
var skip = true;
for (let i=0; i<items.length; i++) {
let item = items[i];
// Cross-library drag
if (treeRow.ref.libraryID != item.libraryID) {
let linkedItem = item.getLinkedItem(treeRow.ref.libraryID, true);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (linkedItem && !linkedItem.deleted) {
// For drag to root, skip if linked item exists
if (treeRow.isLibrary(true)) {
Zotero.debug("Linked item " + linkedItem.key + " already exists "
+ "in library " + treeRow.ref.libraryID);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
continue;
}
// For drag to collection
else if (treeRow.isCollection()) {
// skip if linked item is already in it
if (treeRow.ref.hasItem(linkedItem.id)) {
Zotero.debug("Linked item " + linkedItem.key + " already exists "
+ "in collection");
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
continue;
}
// or if linked item is a child item
else if (!linkedItem.isTopLevelItem()) {
Zotero.debug("Linked item " + linkedItem.key + " already exists "
+ "as child item");
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
continue;
}
}
}
skip = false;
continue;
}
// Intra-library drag
// Make sure there's at least one item that's not already in this destination
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (treeRow.isCollection()) {
if (treeRow.ref.hasItem(item.id)) {
Zotero.debug("Item " + item.id + " already exists in collection");
continue;
}
skip = false;
continue;
}
}
if (skip) {
Zotero.debug("Drag skipped");
return false;
}
}
else if (dataType == 'zotero/collection') {
let draggedCollectionID = data[0];
let draggedCollection = Zotero.Collections.get(draggedCollectionID);
// Dragging a collection to a different library
if (treeRow.ref.libraryID != draggedCollection.libraryID) {
// Disallow if linked collection already exists
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (yield col.getLinkedCollection(treeRow.ref.libraryID)) {
return false;
}
var descendents = col.getDescendents(false, 'collection');
for each(var descendent in descendents) {
descendent = Zotero.Collections.get(descendent.id);
// Disallow if linked collection already exists for any subcollections
//
// If this is allowed in the future for the root collection,
// need to allow drag only to root
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (yield descendent.getLinkedCollection(treeRow.ref.libraryID)) {
return false;
}
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
2007-10-23 07:11:59 +00:00
}
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
return true;
});
/*
* Called when something's been dropped on or next to a row
*/
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeView.prototype.drop = Zotero.Promise.coroutine(function* (row, orient, dataTransfer)
{
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!this.canDrop(row, orient, dataTransfer)
|| !(yield this.canDropCheckAsync(row, orient, dataTransfer))) {
2007-10-23 07:11:59 +00:00
return false;
}
var dragData = Zotero.DragDrop.getDataFromDataTransfer(dataTransfer);
if (!dragData) {
Zotero.debug("No drag data");
return false;
}
var dropEffect = dragData.dropEffect;
2008-11-30 20:18:48 +00:00
var dataType = dragData.dataType;
var data = dragData.data;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var event = Zotero.DragDrop.currentEvent;
var sourceTreeRow = Zotero.DragDrop.getDragSource(dataTransfer);
var targetTreeRow = Zotero.DragDrop.getDragTarget(event);
2008-11-30 20:18:48 +00:00
var copyOptions = {
tags: Zotero.Prefs.get('groups.copyTags'),
childNotes: Zotero.Prefs.get('groups.copyChildNotes'),
childLinks: Zotero.Prefs.get('groups.copyChildLinks'),
childFileAttachments: Zotero.Prefs.get('groups.copyChildFileAttachments')
};
var copyItem = Zotero.Promise.coroutine(function* (item, targetLibraryID, options) {
var targetLibraryType = Zotero.Libraries.get(targetLibraryID).libraryType;
// Check if there's already a copy of this item in the library
var linkedItem = item.getLinkedItem(targetLibraryID, true);
if (linkedItem) {
// If linked item is in the trash, undelete it and remove it from collections
// (since it shouldn't be restored to previous collections)
if (linkedItem.deleted) {
linkedItem.setCollections();
linkedItem.deleted = false;
yield linkedItem.save({
skipSelect: true
});
}
return linkedItem.id;
/*
// TODO: support tags, related, attachments, etc.
// Overlay source item fields on unsaved clone of linked item
var newItem = item.clone(false, linkedItem.clone(true));
newItem.setField('dateAdded', item.dateAdded);
newItem.setField('dateModified', item.dateModified);
var diff = newItem.diff(linkedItem, false, ["dateAdded", "dateModified"]);
if (!diff) {
// Check if creators changed
var creatorsChanged = false;
var creators = item.getCreators();
var linkedCreators = linkedItem.getCreators();
if (creators.length != linkedCreators.length) {
Zotero.debug('Creators have changed');
creatorsChanged = true;
}
else {
for (var i=0; i<creators.length; i++) {
if (!creators[i].ref.equals(linkedCreators[i].ref)) {
Zotero.debug('changed');
creatorsChanged = true;
break;
}
}
}
if (!creatorsChanged) {
Zotero.debug("Linked item hasn't changed -- skipping conflict resolution");
continue;
}
}
toReconcile.push([newItem, linkedItem]);
continue;
*/
}
// Standalone attachment
if (item.isAttachment()) {
var linkMode = item.attachmentLinkMode;
// Skip linked files
if (linkMode == Zotero.Attachments.LINK_MODE_LINKED_FILE) {
Zotero.debug("Skipping standalone linked file attachment on drag");
return false;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!targetTreeRow.filesEditable) {
Zotero.debug("Skipping standalone file attachment on drag");
return false;
}
return Zotero.Attachments.copyAttachmentToLibrary(item, targetLibraryID);
}
// Create new clone item in target library
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
var newItem = item.clone(targetLibraryID, false, !options.tags);
// Set Rights field for My Publications
if (options.license) {
if (!options.keepRights || !newItem.getField('rights')) {
newItem.setField('rights', options.licenseName);
}
}
var newItemID = yield newItem.save({
skipSelect: true
});
// Record link
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
yield newItem.addLinkedItem(item);
if (item.isNote()) {
return newItemID;
}
// For regular items, add child items if prefs and permissions allow
// Child notes
if (options.childNotes) {
var noteIDs = item.getNotes();
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
var notes = Zotero.Items.get(noteIDs);
for each(var note in notes) {
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
let newNote = note.clone(targetLibraryID);
newNote.parentID = newItemID;
yield newNote.save({
skipSelect: true
})
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
yield newNote.addLinkedItem(note);
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
// Child attachments
if (options.childLinks || options.childFileAttachments) {
var attachmentIDs = item.getAttachments();
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
var attachments = Zotero.Items.get(attachmentIDs);
for each(var attachment in attachments) {
var linkMode = attachment.attachmentLinkMode;
// Skip linked files
if (linkMode == Zotero.Attachments.LINK_MODE_LINKED_FILE) {
Zotero.debug("Skipping child linked file attachment on drag");
continue;
}
// Skip imported files if we don't have pref and permissions
if (linkMode == Zotero.Attachments.LINK_MODE_LINKED_URL) {
if (!options.childLinks) {
Zotero.debug("Skipping child link attachment on drag");
continue;
}
}
else {
if (!options.childFileAttachments
|| (!targetTreeRow.filesEditable && !targetTreeRow.isPublications())) {
Zotero.debug("Skipping child file attachment on drag");
continue;
}
}
yield Zotero.Attachments.copyAttachmentToLibrary(attachment, targetLibraryID, newItemID);
}
}
return newItemID;
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var targetLibraryID = targetTreeRow.ref.libraryID;
var targetCollectionID = targetTreeRow.isCollection() ? targetTreeRow.ref.id : false;
if (dataType == 'zotero/collection') {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var droppedCollection = yield Zotero.Collections.getAsync(data[0]);
// Collection drag between libraries
if (targetLibraryID != droppedCollection.libraryID) {
yield Zotero.DB.executeTransaction(function* () {
var copyCollections = Zotero.Promise.coroutine(function* (descendents, parentID, addItems) {
for (var desc of descendents) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Collections
if (desc.type == 'collection') {
var c = yield Zotero.Collections.getAsync(desc.id);
var newCollection = new Zotero.Collection;
newCollection.libraryID = targetLibraryID;
yield c.clone(false, newCollection);
if (parentID) {
newCollection.parentID = parentID;
}
var collectionID = yield newCollection.save();
// Record link
c.addLinkedCollection(newCollection);
// Recursively copy subcollections
if (desc.children.length) {
yield copyCollections(desc.children, collectionID, addItems);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Items
else {
var item = yield Zotero.Items.getAsync(desc.id);
var id = yield copyItem(item, targetLibraryID, copyOptions);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Standalone attachments might not get copied
if (!id) {
continue;
}
// Mark copied item for adding to collection
if (parentID) {
if (!addItems[parentID]) {
addItems[parentID] = [];
}
2016-03-06 01:20:51 -05:00
// If source item is a top-level non-regular item (which can exist in a
// collection) but target item is a child item (which can't), add
// target item's parent to collection instead
if (!item.isRegularItem()) {
let targetItem = yield Zotero.Items.getAsync(id);
let targetItemParentID = targetItem.parentItemID;
if (targetItemParentID) {
id = targetItemParentID;
}
}
2016-03-06 01:20:51 -05:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
addItems[parentID].push(id);
}
}
}
});
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var collections = [{
id: droppedCollection.id,
children: droppedCollection.getDescendents(true),
type: 'collection'
}];
var addItems = {};
yield copyCollections(collections, targetCollectionID, addItems);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
for (var collectionID in addItems) {
var collection = yield Zotero.Collections.getAsync(collectionID);
yield collection.addItems(addItems[collectionID]);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
}
// TODO: add subcollections and subitems, if they don't already exist,
// and display a warning if any of the subcollections already exist
});
}
// Collection drag within a library
else {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
droppedCollection.parentID = targetCollectionID;
yield droppedCollection.saveTx();
}
}
2007-10-23 07:11:59 +00:00
else if (dataType == 'zotero/item') {
2008-11-30 20:18:48 +00:00
var ids = data;
2007-10-23 07:11:59 +00:00
if (ids.length < 1) {
return;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (targetTreeRow.isBucket()) {
targetTreeRow.ref.uploadItems(ids);
return;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
var items = yield Zotero.Items.getAsync(ids);
if (items.length == 0) {
return;
}
if (items[0] instanceof Zotero.FeedItem) {
if (!(targetTreeRow.isCollection() || targetTreeRow.isLibrary() || targetTreeRow.isGroup())) {
return;
}
let promises = [];
for (let item of items) {
// No transaction, because most time is spent traversing urls
promises.push(item.translate(targetLibraryID, targetCollectionID))
}
return Zotero.Promise.all(promises);
}
if (targetTreeRow.isPublications()) {
Deasyncification :back: :cry: While trying to get translation and citing working with asynchronously generated data, we realized that drag-and-drop support was going to be...problematic. Firefox only supports synchronous methods for providing drag data (unlike, it seems, the DataTransferItem interface supported by Chrome), which means that we'd need to preload all relevant data on item selection (bounded by export.quickCopy.dragLimit) and keep the translate/cite methods synchronous (or maintain two separate versions). What we're trying instead is doing what I said in #518 we weren't going to do: loading most object data on startup and leaving many more functions synchronous. Essentially, this takes the various load*() methods described in #518, moves them to startup, and makes them operate on entire libraries rather than individual objects. The obvious downside here (other than undoing much of the work of the last many months) is that it increases startup time, potentially quite a lot for larger libraries. On my laptop, with a 3,000-item library, this adds about 3 seconds to startup time. I haven't yet tested with larger libraries. But I'm hoping that we can optimize this further to reduce that delay. Among other things, this is loading data for all libraries, when it should be able to load data only for the library being viewed. But this is also fundamentally just doing some SELECT queries and storing the results, so it really shouldn't need to be that slow (though performance may be bounded a bit here by XPCOM overhead). If we can make this fast enough, it means that third-party plugins should be able to remain much closer to their current designs. (Some things, including saving, will still need to be made asynchronous.)
2016-03-07 16:05:51 -05:00
let io = this._treebox.treeBody.ownerDocument.defaultView
.ZoteroPane.showPublicationsWizard(items);
if (!io) {
return;
}
copyOptions.childNotes = io.includeNotes;
copyOptions.childFileAttachments = io.includeFiles;
copyOptions.childLinks = true;
copyOptions.tags = true; // TODO: add checkbox
['keepRights', 'license', 'licenseName'].forEach(function (field) {
copyOptions[field] = io[field];
});
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
yield Zotero.DB.executeTransaction(function* () {
var newItems = [];
var newIDs = [];
var toMove = [];
// TODO: support items coming from different sources?
if (items[0].libraryID == targetLibraryID) {
var sameLibrary = true;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
else {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var sameLibrary = false;
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
for each(var item in items) {
if (!item.isTopLevelItem()) {
continue;
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (sameLibrary) {
newIDs.push(item.id);
toMove.push(item.id);
}
else {
newItems.push(item);
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (!sameLibrary) {
var toReconcile = [];
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var newIDs = [];
2016-05-03 23:51:12 +02:00
for (let item of newItems) {
var id = yield copyItem(item, targetLibraryID, copyOptions)
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Standalone attachments might not get copied
if (!id) {
continue;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
newIDs.push(id);
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (toReconcile.length) {
var sourceName = Zotero.Libraries.getName(items[0].libraryID);
var targetName = Zotero.Libraries.getName(targetLibraryID);
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var io = {
dataIn: {
type: "item",
captions: [
// TODO: localize
sourceName,
targetName,
"Merged Item"
],
objects: toReconcile
}
};
/*
if (type == 'item') {
if (!Zotero.Utilities.isEmpty(changedCreators)) {
io.dataIn.changedCreators = changedCreators;
}
}
*/
var wm = Components.classes["@mozilla.org/appshell/window-mediator;1"]
.getService(Components.interfaces.nsIWindowMediator);
var lastWin = wm.getMostRecentWindow("navigator:browser");
lastWin.openDialog('chrome://zotero/content/merge.xul', '', 'chrome,modal,centerscreen', io);
for each(var obj in io.dataOut) {
yield obj.ref.save();
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
}
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// Add items to target collection
if (targetCollectionID) {
2015-03-16 23:40:57 -04:00
let ids = newIDs.filter(function (itemID) {
var item = Zotero.Items.get(itemID);
2015-05-07 00:59:53 -04:00
return item.isTopLevelItem();
2015-03-16 23:40:57 -04:00
});
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var collection = yield Zotero.Collections.getAsync(targetCollectionID);
2015-03-16 23:40:57 -04:00
yield collection.addItems(ids);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
// If moving, remove items from source collection
if (dropEffect == 'move' && toMove.length) {
if (!sameLibrary) {
throw new Error("Cannot move items between libraries");
}
if (!sourceTreeRow || !sourceTreeRow.isCollection()) {
throw new Error("Drag source must be a collection for move action");
}
yield sourceTreeRow.ref.removeItems(toMove);
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2007-10-23 07:11:59 +00:00
}
else if (dataType == 'text/x-moz-url' || dataType == 'application/x-moz-file') {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var targetLibraryID = targetTreeRow.ref.libraryID;
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
if (targetTreeRow.isCollection()) {
var parentCollectionID = targetTreeRow.ref.id;
2007-10-23 07:11:59 +00:00
}
else {
var parentCollectionID = false;
}
var commitNotifier = Zotero.Notifier.begin();
2007-10-23 07:11:59 +00:00
try {
2008-11-30 20:18:48 +00:00
for (var i=0; i<data.length; i++) {
var file = data[i];
2007-10-23 07:11:59 +00:00
if (dataType == 'text/x-moz-url') {
2008-11-30 20:18:48 +00:00
var url = data[i];
2007-10-23 07:11:59 +00:00
if (url.indexOf('file:///') == 0) {
var wm = Components.classes["@mozilla.org/appshell/window-mediator;1"]
.getService(Components.interfaces.nsIWindowMediator);
var win = wm.getMostRecentWindow("navigator:browser");
// If dragging currently loaded page, only convert to
// file if not an HTML document
if (win.content.location.href != url ||
win.content.document.contentType != 'text/html') {
var nsIFPH = Components.classes["@mozilla.org/network/protocol;1?name=file"]
.getService(Components.interfaces.nsIFileProtocolHandler);
try {
var file = nsIFPH.getFileFromURLSpec(url);
}
catch (e) {
Zotero.debug(e);
}
}
}
// Still string, so remote URL
if (typeof file == 'string') {
var wm = Components.classes["@mozilla.org/appshell/window-mediator;1"]
.getService(Components.interfaces.nsIWindowMediator);
var win = wm.getMostRecentWindow("navigator:browser");
win.ZoteroPane.addItemFromURL(url, 'temporaryPDFHack', null, row); // TODO: don't do this
2007-10-23 07:11:59 +00:00
continue;
}
// Otherwise file, so fall through
}
if (dropEffect == 'link') {
yield Zotero.Attachments.linkFromFile({
file: file,
collections: [parentCollectionID]
});
}
else {
yield Zotero.Attachments.importFromFile({
file: file,
libraryID: targetLibraryID,
collections: [parentCollectionID]
});
// If moving, delete original file
if (dragData.dropEffect == 'move') {
try {
file.remove(false);
}
catch (e) {
Components.utils.reportError("Error deleting original file " + file.path + " after drag");
2007-10-23 07:11:59 +00:00
}
}
}
2007-10-23 07:11:59 +00:00
}
}
finally {
Zotero.Notifier.commit(unlock);
}
}
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
});
2008-11-30 20:18:48 +00:00
////////////////////////////////////////////////////////////////////////////////
///
/// Functions for nsITreeView that we have to stub out.
///
////////////////////////////////////////////////////////////////////////////////
Zotero.CollectionTreeView.prototype.isSorted = function() { return false; }
2007-10-23 07:11:59 +00:00
/* Set 'highlighted' property on rows set by setHighlightedRows */
Zotero.CollectionTreeView.prototype.getRowProperties = function(row, prop) {
var props = [];
var treeRow = this.getRow(row);
if (treeRow.isHeader()) {
props.push("header");
}
else if (this._highlightedRows[row]) {
props.push("highlighted");
2007-10-23 07:11:59 +00:00
}
return props.join(" ");
2007-10-23 07:11:59 +00:00
}
Zotero.CollectionTreeView.prototype.getColumnProperties = function(col, prop) {}
Zotero.CollectionTreeView.prototype.getCellProperties = function(row, col, prop) {
var props = [];
var treeRow = this.getRow(row);
if (treeRow.isHeader()) {
props.push("header");
props.push("notwisty");
}
else if (treeRow.isPublications()) {
props.push("notwisty");
} else if (treeRow.ref && treeRow.ref.unreadCount) {
props.push('unread');
}
return props.join(" ");
}
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
Zotero.CollectionTreeView.prototype.isSeparator = function(index) {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
var source = this.getRow(index);
2.0b3 megacommit - Support for group libraries - General support for multiple libraries of different types - Streamlined sync support - Using solely libraryID and key rather than itemID, and removed all itemID-changing code - Combined two requests for increased performance and decreased server load - Added warning on user account change - Provide explicit error message on SSL failure - Removed snapshot and link toolbar buttons and changed browser context menu options and drags to create parent items + snapshots - Closes #786, Add numPages field - Fixes #1063, Duplicate item with tags broken in Sync Preview - Added better purging of deleted tags - Added local user key before first sync - Add clientDateModified to all objects for more flexibility in syncing - Added new triples-based Relation object type, currently used to store links between items copied between local and group libraries - Updated zotero.org translator for groups - Additional trigger-based consistency checks - Fixed broken URL drag in Firefox 3.5 - Disabled zeroconf menu option (no longer functional) Developer-specific changes: - Overhauled data layer - Data object constructors no longer take arguments (return to 1.0-like API) - Existing objects can be retrieved by setting id or library/key properties - id/library/key must be set for new objects before other fields - New methods: - ZoteroPane.getSelectedLibraryID() - ZoteroPane.getSelectedGroup(asID) - ZoteroPane.addItemFromDocument(doc, itemType, saveSnapshot) - ZoteroPane.addItemFromURL(url, itemType) - ZoteroPane.canEdit() - Zotero.CollectionTreeView.selectLibrary(libraryID) - New Zotero.URI methods - Changed methods - Many data object methods now take a libraryID - ZoteroPane.addAttachmentFromPage(link, itemID) - Removed saveItem and saveAttachments parameters from Zotero.Translate constructor - translate() now takes a libraryID, null for local library, or false to not save items (previously on constructor) - saveAttachments is now a translate() parameter - Zotero.flattenArguments() better handles passed objects - Zotero.File.getFileHash() (not currently used)
2009-05-14 18:23:40 +00:00
return source.type == 'separator';
}
Zotero.CollectionTreeView.prototype.performAction = function(action) { }
Zotero.CollectionTreeView.prototype.performActionOnCell = function(action, row, col) { }
Zotero.CollectionTreeView.prototype.getProgressMode = function(row, col) { }
Zotero.CollectionTreeView.prototype.cycleHeader = function(column) { }
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
Zotero.CollectionTreeCache = {
"lastTreeRow":null,
"lastTempTable":null,
"lastSearch":null,
"lastResults":null,
2015-05-25 21:22:43 -04:00
"clear": function () {
Async DB megacommit Promise-based rewrite of most of the codebase, with asynchronous database and file access -- see https://github.com/zotero/zotero/issues/518 for details. WARNING: This includes backwards-incompatible schema changes. An incomplete list of other changes: - Schema overhaul - Replace main tables with new versions with updated schema - Enable real foreign key support and remove previous triggers - Don't use NULLs for local libraryID, which broke the UNIQUE index preventing object key duplication. All code (Zotero and third-party) using NULL for the local library will need to be updated to use 0 instead (already done for Zotero code) - Add 'compatibility' DB version that can be incremented manually to break DB compatibility with previous versions. 'userdata' upgrades will no longer automatically break compatibility. - Demote creators and tags from first-class objects to item properties - New API syncing properties - 'synced'/'version' properties to data objects - 'etag' to groups - 'version' to libraries - Create Zotero.DataObject that other objects inherit from - Consolidate data object loading into Zotero.DataObjects - Change object reloading so that only the loaded and changed parts of objects are reloaded, instead of reloading all data from the database (with some exceptions, including item primary data) - Items and collections now have .parentItem and .parentKey properties, replacing item.getSource() and item.getSourceKey() - New function Zotero.serial(fn), to wrap an async function such that all calls are run serially - New function Zotero.Utilities.Internal.forEachChunkAsync(arr, chunkSize, func) - Add tag selector loading message - Various API and name changes, since everything was breaking anyway Known broken things: - Syncing (will be completely rewritten for API syncing) - Translation architecture (needs promise-based rewrite) - Duplicates view - DB integrity check (from schema changes) - Dragging (may be difficult to fix) Lots of other big and little things are certainly broken, particularly with the UI, which can be affected by async code in all sorts of subtle ways.
2014-08-06 17:38:05 -04:00
this.lastTreeRow = null;
this.lastSearch = null;
if(this.lastTempTable) {
// Drop the last temp table when we can. We don't wait on this because it can cause a
// deadlock: this waits on open transactions, but a transaction could be waiting on
// ItemTreeView::notify(), which waits on ItemTreeView::refresh(), which calls this.
Zotero.DB.queryTx("DROP TABLE IF EXISTS " + this.lastTempTable).done();
}
this.lastTempTable = null;
this.lastResults = null;
2015-05-25 21:22:43 -04:00
}
};