Fixes #1400 RHS has updated their XML format, also added support for bookSections

This commit is contained in:
Matt Burton 2009-03-04 01:57:09 +00:00
parent 4ac98e19a3
commit c781705793

View file

@ -12,10 +12,10 @@
}
function detectWeb(doc, url) {
if (doc.evaluate('//tr/td[3][@class="bib_data"]/a[@class="bibref"]', doc, null, XPathResult.ANY_TYPE, null).iterateNext()) {
if (doc.evaluate('//div[@class="bib_data"]/a[contains(@title, "view in more detail")]', doc, null, XPathResult.ANY_TYPE, null).iterateNext()) {
return "multiple";
} else if (doc.evaluate('//a[text() = "View records in XML"]', doc, null, XPathResult.ANY_TYPE, null).iterateNext()) {
return "book";
return "document";
}
}
@ -23,7 +23,7 @@ function doWeb(doc, url) {
var books = new Array();
if (detectWeb(doc, url) == "multiple") {
var items = new Object();
var xpath = '//a[@class="bibref"]';
var xpath = '//div[@class="bib_data"]/a[contains(@title, "view in more detail")]';
var results = doc.evaluate(xpath, doc, null, XPathResult.ANY_TYPE, null);
var result;
while (result = results.iterateNext()) {
@ -41,12 +41,14 @@ function doWeb(doc, url) {
text = text.replace(/<\?[^?]+\?>/, "");
var xml = new XML(text);
xml = xml..recordList;
var itemtype = "book";
if (xml..journal_title.length() != 0) itemtype = "journalArticle";
if (xml..journal.length() != 0) itemtype = "journalArticle";
if (xml..in.length() != 0) itemtype = "bookSection";
var item = new Zotero.Item(itemtype);
item.title = Zotero.Utilities.capitalizeTitle(xml..title.toString());
item.title = Zotero.Utilities.cleanTags(xml..title.toString());
item.date = xml..date.toString().match(/\d+/);
item.pages = xml..length.toString();
for (var i = 0; i < xml..author.length(); i++) {
var name = xml..author[i].toString().match(/^[^,]+,[^,]+/)[0].split(/,\s+/);
item.creators.push({lastName:name[0], firstName:name[1], creatorType:"author"});
@ -54,20 +56,33 @@ function doWeb(doc, url) {
if (item.itemType == "book") {
item.place = xml..place_of_publication.toString();
item.publisher = xml..publisher.toString();
item.date = xml..publication_year.toString();
item.ISBN = xml..ISBN.toString();
} else if (item.itemType == "journalArticle") {
item.publicationTitle = xml..journal_title.toString();
var voliss = xml..journal_number.split(":");
Zotero.debug(voliss);
item.volume = voliss[0];
item.issue = voliss[1];
item.date = xml..journal_issue_year;
item.pages = xml..journal_pages;
item.publicationTitle = xml..journal.toString();
item.volume = xml..journal_volume.toString();
item.issue = xml..journal_issue.toString();
item.series = xml..series.toString();
item.ISSN = xml..ISSN.toString();
} else if (item.itemType == "bookSection") {
if (xml..title_interpolation.toString()) {
item.bookTitle = xml..title_interpolation.toString();
} else {
item.bookTitle = xml..in.toString().match(/\),\s(.+)\s\(/)[1];
}
item.ISBN = xml..ISBN.toString();
try { // this regex is iffy
bookInfo = xml..in.toString().match(/\(([A-Za-z;\s]+:[A-Za-z\s']+),\s\d+\)/)[1];
} catch (e) {
Zotero.debug("regex didn't work");
}
item.place = bookInfo.split(": ")[0]
item.publisher = bookInfo.split(": ")[1]
}
item.complete();
});
Zotero.done();
}, function(){Zotero.done();});
}
Zotero.wait();
}