closes#704, EndNote to Zotero style converter (won't actually convert styles due to copyright concerns, but will load them into the DB)
also adds CSL style manager
- Inspired by Dan Chudnov's Python/MODS-based Zeroconf demo at THATcamp
- Enabled by extensions.zotero.zeroconf.enabled (off by default)
- Currently supports only OS X (tested on Leopard, not sure about earlier versions)
- Uses Apple's dns-sd and mDNS command-client clients, but should be able to be extended to other clients, though a native library would be far superior
- Discovery is on-demand for now via Actions menu ("Search for Shared Libraries")
- Includes rudimentary web server (code copied from integration.js) that serves items as sync XML -- no authentication yet!
- Only supports top-level items
- Remote libraries show up in left pane (under remote computer name, for now)
- Items can be dragged into collections (but not the library yet, for some reason)
- On first run, might cause a long pause and the "This file was downloaded from the Internet" message on Leopard -- can't manage to get around the quarantine for the script file that we need to access stdout from Firefox
- Needs a lot of work, and without a real JS (or otherwise Mozilla-native) Zeroconf library we can't do proper discovery without intermittent polling
- But it works, at least for me
Also includes some data/sync-layer changes that I needed along the way (and that we'll need for shared collections of any type)
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.
Closes#285, Hitting Esc should close tags popup
Fixes various UI issues and problems adding tags on Minefield
Still getting autocomplete self-destruction
This is a bit uglier, and the large markers are kinda annoying once you know the drop-down is there, but perhaps this will stop the forum posts. Please let them stop.
If anyone's out there, now would be the time to test this.
Missing icons for the new types -- I could try to get some of those now, or we could just launch. Maybe we'll just launch. But really, might want to test this. Just sayin'.
Use automaticSnapshots pref (which defaults to on and is changeable in the prefs window) rather than downloadAssociatedFiles (which defaults to off and is only settable through about:config at the moment) for now in translate.js
downloadAssociatedFiles should eventually be used for PDFs and other large files, whereas automaticSnapshots will be for HTML and the like -- in the meantime, I think it's OK for scrapers to just follow the visible pref for both, since otherwise they'd be totally confused when the NIFP button took a snapshot and the scrapers didn't
Simon, if there's any problem I'm not aware of with switching this for now (other than people getting some large PDFs on JStor), let me know.
closes#165, verify import/export can carry all data for all fields and item types
closes#168, make sure MODS import works with files from external sources
Attach Snapshot of Current Page
Attach Link to Current Page
I also removed "New Item" from the context menu, as it was the only option in the menu that wasn't specific to the selected item, it made the whole menu a lot more confusing, and it doesn't take all that much longer to click the (+) button and choose Book (assuming that's even the item you want).
Addresses #155, Localize strings
- Added "Search subfolders"
- Localized strings in search dialog
Known issue:
#342, "Search subfolders" checkbox should be greyed out until applicable
Including in the DB, which it turns out isn't really all that bad (thanks, among other things, to SQLite's ability to DROP tables within transactions without autocommitting (which MySQL can't do))
At least some scrapers (NYT and WashPo, for sure) should be updated to follow this pref
Addresses #327, Scrapers should either take snapshots or use URL field
Changed creator type <label> to <toolbarbutton> -- hopefully won't have any ill effects with other platforms' default themes (I haven't checked)
Moved some metadata pane CSS from overlay.css to zotero.css
Currently uses an annoying popup window, since the code used in the regular metadata pane has too much baggage to use here -- it really should be abstracted into an XBL binding, but that won't be an easy feat
Also fixes SQL error opening popup notes and JS strict warnings when saving them
- Currently requires user to enter dates in SQL format if they want to change the access date, but at least it doesn't mangle the dates anymore
- Uses new function ScholardammitZotero.Date.dateToSQL(Date date [, Boolean toUTC])
- Utilities.lpad() now forces _string_ to a string so that .length exists
- Unrelated: Item.save() now returns false if the item didn't change
closes#313, Blacklist known ad sites from scraper detection
closes#306, some New York Times ads prevent page from being recognized
closes#308, attachment import bug
currently, the ad site blacklist is located at the top of ingester/browser.js. at some point, we may want to switch this to a database table.
addresses #327, Scrapers should either take snapshots or use URL field
closes#309, Integration server prevents Zotero from loading in multiple instances of Firefox
In other words, show both "Shakespeare" and "Shakespeare, William" in the drop-down, and if the latter is chosen, save both fields
One issue is that since the autocomplete is by default limited to the width of the textbox, longer entries get truncated (though you can see them with a mouseover), and that may not be easy to fix.
Changed "Scholar" to "Zotero", everywhere
Apologies to anyone with working copy changes, but there are probably the fewer at this moment than there will be again.
Hopefully this won't break anything, though existing prefs will be lost. I avoided scholar.google.com--if you know any other legitimate "scholar"s in the code, be sure to fix them once I'm done here.
This is a multi-commit change--there's at least one more coming. *Do not update to this version! It won't work!*
Closes#304, change references to "website" to "web page"
More changes as per discussions with Dan:
- Linked URLs have been given a second chance at life, though they still shouldn't be used for (most, if any) scrapers (which should use snapshots or the URL field instead)
- Renamed the "website" item type to "webpage"
- Removed "web page" from the New Item menu
- Added Save Link To Current Page toolbar button
- Added toolbar separator between New Item buttons and link/attachment/note to differentiate
- Added limited metadata (URL and accessDate) for attachments
- URL for attachments now stored in itemData (itemAttachments.originalPath is no longer used, but I'm probably not gonna worry about it and just wait for SQLite to support dropping columns with ALTER TABLE) -- getURL() removed in favor of getField('url')
- Snapshots now say "View Snapshot"
- Added Show File button to file attachments to show in filesystem
- Added timed note field to attachments for single notes and adjusted Item.updateNote(), etc. to work with attachments
- Fixed bug with manually bound params in fulltext indexer and Item.save() (execute() vs. executeStep()) -- any recently added items probably aren't in the fulltext index because of this
Known bugs/issues:
- Attachment metadata and notes probably aren't properly imported/exported now (and accessDate definitely isn't)
- Scrapers don't save metadata properly
- Attachment title should be editable
- File attachments could probably use some more metadata (#275, more or less, though they won't be getting tabs)
And fixed a bug that caused the text of some search conditions to not display when editing a previously saved search
Also added autocomplete to more fields in metadata pane:
- creator
- publisher
- place
- publicationTitle
- journalAbbreviation
- seriesTitle
- seriesText
It should also do the type and medium fields, but they need to be separated out first--e.g. artworkType, interviewMedium--since they're describing different data
Overhaul of the item drag and drop functionality, allowing dragging notes and attachments into and out of other items and addressing all the related issues that arise when that's possible.
Should also be generally smarter about deciding what can be dragged and dropped where and in what modes.
Let me know if something doesn't work as you expect.
Icons from Ken
Of course, I realized just now that this is really quite similar to the stop sign, and anyone who puts this to the left of their URL bar and doesn't have a Home icon in their toolbar will get the two next to each other... Looks good otherwise, though.
Changes as per my discussions with Dan:
- Separated snapshot functionality into two individual buttons, Create New Item From Current Page and Take Snapshot of Current page
- Updated schema to support primary, secondary and hidden item types (and future user customizations)
- Reorganized New Item menu, moving secondary items into sub-menu
- Removed ability to create link attachments, since it never really made much sense -- will simply use the webpage item type instead. Underlying functionality still exists for the time being, as people have existing links in their libraries--I think we're gonna have to just warn beta testers and delete them in a transition step, as converting nested links really wouldn't be worth the effort.
- Moved file link/add functions into new item menu and removed attachment drop-down
- Large, prominent View and Locate buttons in edit pane for going to an associated URL and looking up in OpenURL, respectively -- buttons gray out as appropriate
- New Item from Page stores the URL and access date (Item.save() checks for the string "CURRENT_TIMESTAMP" for accessDate and doesn't bind it as a string)
- "Website" to "Web Page" (do we prefer "Webpage"? they both look a bit funky in uppercase)
More coming.
Bugs/Known Issues:
- Since snapshots from the toolbar are now top-level in the current collection, there needs to be a way to drag them into items
- The camera icon for adding snapshots, despite being a famfamfam icon, really doesn't read too well (or perhaps just clashes with the rest of our icons). Anybody have a better one? (It also may be able to just be lightened up a bit.)
- Trying the large View/Locate buttons after discussions with Dan, but this approach may not work -- 1) a large View button for the URL makes a lot less sense when you have a parent item with a child snapshot, since people will end up clicking it all the time when they really want to view the snapshot, and 2) the Locate button is awfully big for something that only applies to certain types of items, may not get used very often when it does, and probably won't work when it is
- The access date is stored in UTC and displayed with toLocaleString() like Date Added and Date Modified, but, unlike those two, it's also user-editable. This is clearly a problem. Probably need to parse to Date on blur() with strToDate() and insert as UTC, discarding anything left over.
- Item type itself is still "website" -- should probably change that while we still can
Closes#253, OpenURL arrow should provide visual feedback on mouseover and/or look more button-like
Addresses #304, change references to "website" to "web page"
Addresses #207, openurl arrow functionality
Differentiates between single and double fields for the search, but there's a problem in the current implementation in that only one field is editable at once, so displaying two-field names in a drop-down is a little problematic. While I could display the full names, comma-delimited, and get the discrete parts (which is what Scholar.Utilities.AutoComplete.getResultComment(), included in this commit, is for--the creatorID for the row would be hidden in the autocomplete drop-down comment field), it's a bit unclear what should happen when a user selects a comma-separated name from the drop-down of one of the fields. One option would be to have a row for the last name (in case that's all they want to complete) and other rows for "last, first" matches, and selecting one of the two-part names would replace whatever's in the opposite name field with the appropriate text (and save it to the DB, I'm afraid, unless I change how the creator fields work), keeping the focus in the current textbox for easy tabbing. Not great, but it might work.
Other ideas?
Adds tab handling to tags interface, which also addresses beta tester requests for a quicker way to enter multiple tags
This is getting pretty messy, but it's still probably better than repeating all the itemPane.js in tagsbox.xml.
One known issue is that, since it resorts the list of tags after a change, if you change an existing tag to a name that alters its current position and then tab away, you don't necessarily end up in the field you expect.
- Switched to manually repeated bound parameters in indexWords()
- Switched to the innerHTML regex used elsewhere instead of a more proper but nevertheless misguided DOM traverser to split elements in indexDocument
This may invalidate the fulltext progress indicator ticket
Still need #292, "Delete From Library" context menu option in collections
This may or may not help people understand that deleting from a collection doesn't by design delete from the library, but, regardless, this is the same behavior as iTunes.
Example usage:
var windowWatcher = Components.classes["@mozilla.org/embedcomp/window-watcher;1"].
getService(Components.interfaces.nsIWindowWatcher);
var progress = new Scholar.ProgressWindow(windowWatcher.activeWindow);
progress.changeHeadline('Indexing item...');
progress.addLines(['All About Foo'], ['chrome://scholar/skin/treeitem-book.png']);
progress.addDescription('Bar bar bar bar bar');
progress.show();
progress.fade();
There are currently two types of fulltext searching: an SQL-based word index and a file scanner. They each have their advantages and drawbacks.
The word index is very fast to search and is currently used for the find-as-you-type quicksearch. However, indexing files takes some time, so we should probably offer a preference to turn it off ("Index attachment content for quicksearch" or something). There's also an issue with Chinese characters (which are indexed by character rather than word, since there are no spaces to go by, so a search for a word with common characters could produce erroneous results). The quicksearch doesn't use a left-bound index (since that would probably upset German speakers searching for "musik" in "nachtmusik," though I don't know for sure how they think of words) but still seems pretty fast.
* Note: There will be a potentially long delay when you start Firefox with this revision as it builds a fulltext word index of your existing items. We obviously need a notification/option for this. *
The file scanner, used in the Attachment Content condition of the search dialog, offers phrase searching as well as regex support (both case-sensitive and not, and defaulting to multiline). It doesn't require an index, though it should probably be optimized to use the word index, if available, for narrowing the results when not in regex mode. (It does only scan files that pass all the other search conditions, which speeds it up considerably for multi-condition searches, and skips non-text files unless instructed otherwise, but it's still relatively slow.)
Both convert HTML to text before searching (with the exception of the binary file scanning mode).
There are some issues with which files get indexed and which don't that we can't do much about and that will probably confuse users immensely. Dan C. suggested some sort of indicator (say, a green dot) to show which files are indexed.
Also added (very ugly) charset detection (anybody want to figure out getCharsetFromString(str)?), a setTimeout() replacement in the XPCOM service, an arrayToHash() method, and a new header to timedtextarea.xml, since it's really not copyright CHNM (it's really just a few lines off from the toolkit timed-textbox binding--I tried to change it to extend timed-textbox and just ignore Return keypress events so that we didn't need to duplicate the Mozilla code, but timed-textbox's reliance on html:input instead of html:textarea made things rather difficult).
To do:
- Pref/buttons to disable/clear/rebuild fulltext index
- Hidden prefs to set maximum file size to index/scan
- Don't index words of fewer than 3 non-Asian characters
- MRU cache for saved searches
- Use word index if available to narrow search scope of fulltext scanner
- Cache attachment info methods
- Show content excerpt in search results (at least in advanced search window, when it exists)
- Notification window (a la scraping) to show when indexing
- Indicator of indexed status
- Context menu option to index
- Indicator that a file scanning search is in progress, if possible
- Find other ways to make it index the NYT front page in under 10 seconds
- Probably fix lots of bugs, which you will likely start telling me about...now.
Note that there's no code for user types and fields yet -- just the schema (actually there's a tiny bit of code in the item type manager, since we'll probably use some of the same methods for managing user types, but not much)
Templates for primary item types are currently only used by the item type manager to make creating new types easier and to prevent the removal of fields from an item type that are associated with its template item type -- the fields are all still recorded in itemTypeFields, since they might have different orders or default visibility settings from their templates
This lets us add tables to user.sql without writing migration steps for them yet still have the ability to change existing user tables and migrate data if necessary.
Also added _getDropCommands() to do a regex on the SQL file and create the DROP TABLE|INDEX steps necessary to use the DB_REBUILD flag without DROP commands in the SQL file itself, before I realized that it probably made the most sense to just delete the SQL file and storage directory. Changed _initializeSchema() to do that instead. Leaving _getDropCommands() in, in case there's ever a need for it.
Using Shift-Enter as the save keystroke within the Extra textbox so that people can use Enter to create multiple lines of text. Shift-Enter would normally be the newline command, but that's probably a convention that non-technical users of Zotero wouldn't know... Tab (and other triggers for blur()) also saves, and since Extra is the last field, tabbing away functions the same as hitting Enter does for other fields, so it's probably not that big of a deal.
This should make development much easier, as we can, for example, add 80 item types without having to write transition steps
Pretty sure this won't delete anyone's data. Might want to test that theory, though.
- closes#217, ability to exclude notes/attachments from select items window
- closes#244, ability to quick search from select items window
- fixes a bug with footnotes in Word integration
- fixes a bug in InnoPAC translator where items would sometimes appear twice
Here's a shot at a single/double creator field toggle switch -- let me know what you think
A few issues:
- There's currently no comma between the last name and first name when in two-field mode -- I removed it to greatly simplify the code, hoping to be able to use the CSS :after pseudo-element, but that seems to not work with XUL -- I'll figure out a clean solution and add it back ( refs #288 )
- It's not very smart about switching between single-field mode and two-field mode, as it currently just keeps the last word (even if it's "Jr." or "III") as the last name and puts the rest in the first name field -- not a big deal, but it should at least be a bit smarter about this ( refs #289 )
- There are probably some other bugs
New logic in itemTreeView notify() target:
- Items are only selected on add in the active window -- this fixes a fairly major flaw in the previous system that would cause new items to be selected in all open windows
- If a quicksearch is open in the active window and a new item is added, clear the search
- If quicksearch and active window and item modify, rerun search
- If quicksearch and not active window, rerun search
- If not quicksearch and not active window, update list but retain previous selection
Addresses #260, Add auto-complete to search window
- New XPCOM autocomplete component for Zotero data -- can be used by setting the autocompletesearch attribute of a textbox to 'zotero' and passing a search scope with the autocompletesearchparam attribute. Additional parameters can be passed by appending them to the autocompletesearchparam value with a '/', e.g. 'tag/2732' (to exclude tags that show up in item 2732)
- Tag entry now uses more or less the same interface as metadata -- no more popup window -- note that tab isn't working properly yet, and there's no way to quickly enter multiple tags (though it's now considerably quicker than it was before)
- Autocomplete for tags, excluding any tags already set for the current item
- Standalone note windows now register with the Notifier (since tags needed item modification notifications to work properly), which will help with #282, "Notes opened in separate windows need item notification"
- Tags are now retrieved in alphabetical order
- Scholar.Item.replaceTag(oldTagID, newTag), with a single notify
- Scholar.getAncestorByTagName(elem, tagName) -- walk up the DOM tree from an element until an element with the specified tag name is found (also checks with 'xul:' prefix, for use in XBL), or false if not found -- probably shouldn't be used too widely, since it's doing string comparisons, but better than specifying, say, nine '.parentNode' properties, and makes for more resilient code
A few notes:
- Autocomplete in Minefield seems to self-destruct after using it in the same field a few times, taking down saving of the field with it -- this may or may not be my fault, but it makes Zotero more or less unusable in 3.0 at the moment. Sorry. (I use 3.0 myself for development, so I'll work on it.)
- This would have been much, much easier if having an autocomplete textbox (which uses an XBL-generated popup for the suggestions) within a popup (as it is in the independent note edit panes) didn't introduce all sorts of crazy bugs that had to be defeated with annoying hackery -- one side effect of this is that at the moment you can't close the tags popup with the Escape key
- Independent note windows now need to pull in itemPane.js to function properly, which is a bit messy and not ideal, but less messy and more ideal than duplicating all the dual-state editor and tabindex logic would be
- Hitting tab in a tag field not only doesn't work but also breaks things until the next window refresh.
- There are undoubtedly other bugs.
- import translators no longer fail when trying to import an item with no name
- the T2/BT field becomes the publication title when no JO/JF field is available (fixes newspaper issues)
- Y2 is now treated as part of the date if and only if it is improperly formatted (seriously, why can't Thomson get their own specs straight?)
- work around EndNote's strange behavior of putting article titles into notes for no apparent reason
- RIS export gives dates as per specification
- fixed a bug that could have (potentially) caused problems formatting "January"
- allow translators to access strToDate function
MODS uses the encoding as specified in the <?xml tag, or else UTF-8
RIS uses IBM850, since the spec says "IBM Extended Character Set" and it's the only code page Mozilla supports. (should I do this? or just use unicode?)
MARC uses UTF-8, since I don't think there's any way to get full MARC-8 support, and UTF-8 is now the preferred encoding anyway
- improved bibliography (especially Chicago Manual of Style)
- improved error handling for import/export/bibliography
- bibliographic export now ignores notes and standalone attachment (before, they made export silently fail). an error appears if you try to generate a bibliography from only notes or standalone attachments.
positions "saving item" window in a slightly better place on Windows
the UMich bug was actually bigger than I though. as it turns out, the HiddenDOMWindow in Windows is not a chrome window, so i had to modify createHiddenBrowser() to attach the hidden browser object to an existing browser window. i don't believe this should have any adverse effects for snapshots, etc., but Dan, correct me if i'm wrong. it would be nice to be able to create a real chrome instance instead of a XUL element, but all of my attempts at doing so have failed.
i've fixed the Amazon.com bug (i think) and made the translator show a "Could Not Save Item" prompt rather than show an empty list, but if you see any other pages where this happens, let me know
- make EBSCO scraper work better through a proxy
- shorten Accession Number -> Accession No, Journal Abbreviation -> Journal Abbr, Publication Title -> Publication. it does look a bit stranger, but it also makes the interface more functional (especially for those of us without giant widescreen LCDs ;-)
(Apparently 'columnName NONE' in SQLite is functionally equivalent to 'columnName ILUVTIDDLYWINKS' in setting column affinity--as in, they both use the default of NUMERIC--which frankly I feel the docs could've been slightly more clear on, since they scatter the word NONE in capital letters all about the page with the other affinity keywords...)
modifies scrapers to use dates in the format that comes out of the page, rather than converting to SQL
adds Scholar.Date.formatDate() to provide a pretty representation of dates
