mirrors/zotero
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

Don't truncate HTML file content in full-text sync

Browse source 
Also write a cache file for HTML files at index time
This commit is contained in:
Dan Stillman 2013-11-04 01:41:04 -05:00
parent 62840f3b93
commit ccf2a05c49
1 changed files with 33 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

39
chrome/content/zotero/xpcom/fulltext.js
View file

@ -370,9 +370,22 @@ Zotero.Fulltext = new function(){
text = text.replace(/(>)/g, '$1 ');
text = this.HTMLToText(text);
this.indexString(text, document.characterSet, itemID);
var charsIndexed = Math.min(maxLength, text.length);
this.setChars(itemID, { indexed: charsIndexed, total: text.length });
// Write the converted text to a cache file
Q.fcall(function () {
let cacheFile = self.getItemCacheFile(itemID);
Zotero.debug("Writing converted full-text HTML content to " + cacheFile.path);
if (!cacheFile.parent.exists()) {
Zotero.Attachments.createDirectoryForItem(itemID);
}
return Zotero.File.putContentsAsync(cacheFile, text);
})
.catch(function (e) {
Zotero.debug(e, 1);
Components.utils.reportError(e);
})
}
@ -597,6 +610,7 @@ Zotero.Fulltext = new function(){
* @return {Array<Object>}
*/
this.getUnsyncedContent = function (maxChars) {
var maxLength = Zotero.Prefs.get('fulltext.textMaxLength');
var first = true;
var chars = 0;
var contentItems = [];
@ -632,15 +646,18 @@ Zotero.Fulltext = new function(){
}
Zotero.debug("Adding full-text content from file for item " + libraryKey);
text = Zotero.File.getContents(
file, item.attachmentCharset, row.indexedChars
);
text = Zotero.File.getContents(file, item.attachmentCharset, maxLength);
// Split elements to avoid word concatentation
// If HTML, convert to plain text first, and cache the result
if (item.attachmentMIMEType == 'text/html') {
// Split elements to avoid word concatentation
text = text.replace(/(>)/g, '$1 ');
text = this.HTMLToText(text);
// Include in the cache file only as many characters as we've indexed
text = text.substr(0, row.indexedChars);
// Write the converted text to a cache file
Zotero.debug("Writing converted full-text HTML content to "
+ cacheFile.path);
@ -651,7 +668,11 @@ Zotero.Fulltext = new function(){
.catch(function (e) {
Zotero.debug(e, 1);
Components.utils.reportError(e);
})
});
}
else {
// Include only as many characters as we've indexed
text = text.substr(0, row.indexedChars);
}
}
}
@ -1067,6 +1088,12 @@ Zotero.Fulltext = new function(){
content = this.HTMLToText(content);
// Include in the cache file only as many characters as we've indexed
let chars = this.getChars(itemID);
if (chars && chars.indexedChars) {
content = content.substr(0, chars.indexedChars);
}
// Write the converted text to a cache file for future searches
Zotero.debug("Writing converted full-text content to " + cacheFile.path);
if (!cacheFile.parent.exists()) {