suzanne.soy/zotero
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

migrate sciencedirect to use processAsycn(), something strange with the DOI xpath, using the data from RIS and regex instead

Browse Source
This commit is contained in:
Matt Burton 2009-06-30 17:20:10 +00:00
parent 5e080b78f0
commit fb104160d5
1 changed files with 172 additions and 84 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

110
translators/ScienceDirect.js
View File

@ -58,9 +58,17 @@ function doWeb(doc, url) {
Zotero.debug('no items');
return;
}
Zotero.Utilities.processDocuments(articles, function(newDoc) {
var doi = newDoc.evaluate('//div[@class="articleHeaderInner"][@id="articleHeader"]/a[contains(text(), "doi")]', newDoc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent.substr(4);
var sets = [];
for each (article in articles) {
sets.push({article:article});
}
var first = function(set, next) {
var article = set.article;
Zotero.Utilities.processDocuments(article, function(newDoc) {
var tempPDF = newDoc.evaluate('//a[@class="noul" and div/div[contains(text(), "PDF")]]', newDoc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext();
if (!tempPDF) { // PDF xpath failed, lets try another
tempPDF = newDoc.evaluate('//a[@class="noul" and contains(text(), "PDF")]', newDoc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext();
@ -86,6 +94,27 @@ function doWeb(doc, url) {
{url:url, title:"ScienceDirect Snapshot", mimeType:"text/html"},
];
}
// This does not work, not sure why.
//var doi = newDoc.evaluate('//a[contains(text(), "doi")]/text()', newDoc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext();
//Zotero.debug(doi);
//doi = doi.textContent.substr(4);
// pass these values to the next function
//set.doi = doi;
set.url = url;
set.get = get;
set.attachments = attachments;
next();
});
};
var second = function(set, next) {
var url = set.url;
var get = set.get;
Zotero.Utilities.HTTP.doGet(get, function(text) {
var md5 = text.match(/<input type=hidden name=md5 value=([^>]+)>/)[1];
var acct = text.match(/<input type=hidden name=_acct value=([^>]+)>/)[1];
@ -101,6 +130,23 @@ function doWeb(doc, url) {
}
var post = "_ob=DownloadURL&_method=finish&_acct=" + acct + "&_userid=" + userid + "&_docType=FLA&" + docID + "&md5=" + md5 + "&count=1&JAVASCRIPT_ON=Y&format=cite-abs&citation-type=RIS&Export=Export&x=26&y=17";
var baseurl = url.match(/https?:\/\/[^/]+\//)[0];
set.post = post;
set.baseurl = baseurl;
next();
});
};
var third = function(set, next) {
var baseurl = set.baseurl;
var post = set.post;
var attachments = set.attachments;
//var doi = set.doi;
Zotero.Utilities.HTTP.doPost(baseurl + 'science', post, function(text) {
var translator = Zotero.loadTranslator("import");
translator.setTranslator("32d59d2d-b65a-4da4-b0a3-bdd3cfb979e7");
@ -112,15 +158,22 @@ function doWeb(doc, url) {
item.abstractNote = item.notes[0].note;
item.notes = new Array();
}
if (doi) {
item.DOI = doi;
}
item.DOI = item.DOI.substr(10);
//if (doi) {
// item.DOI = doi;
//}
item.complete();
});
translator.translate();
next();
}, false, 'windows-1252');
});
}, function() {Zotero.done();});
};
var functioncallbacks = [first, second, third];
Zotero.Utilities.processAsync(sets, functioncallbacks, function() {Zotero.done()});
} else {
var articles = new Array();
if (detectWeb(doc, url) == "multiple") {
@ -149,7 +202,19 @@ function doWeb(doc, url) {
Zotero.debug('no items');
return;
}
Zotero.Utilities.processDocuments(articles, function(doc2) {
var sets = [];
for each (article in articles) {
sets.push({article:article});
}
var first = function(set, next) {
var article = set.article;
Zotero.Utilities.processDocuments(article, function(doc2) {
var item = new Zotero.Item("journalArticle");
item.repository = "ScienceDirect";
item.url = doc2.location.href;
@ -161,11 +226,12 @@ function doWeb(doc, url) {
if (voliss.match(/Issues?\s+[^,]+/)) item.issue = voliss.match(/Issues?\s+([^,]+)/)[1];
if (voliss.match(/(J|F|M|A|S|O|N|D)\w+\s+\d{4}/)) item.date = voliss.match(/(J|F|M|A|S|O|N|D)\w+\s+\d{4}/)[0];
if (voliss.match(/Pages?\s+[^,^\s]+/)) item.pages = voliss.match(/Pages?\s+([^,^\s]+)/)[1];
item.DOI = doc2.evaluate('//div[@class="articleHeaderInner"][@id="articleHeader"]/a[contains(text(), "doi")]', doc2, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent.substr(4);
// why doesn't this work?
//item.DOI = doc2.evaluate('//a[contains(text(), "doi")]/text()', doc2, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent.substr(4);
var abspath = '//div[@class="articleHeaderInner"][@id="articleHeader"]/div[@class="articleText"]/p';
var absx = doc2.evaluate(abspath, doc2, nsResolver, XPathResult.ANY_TYPE, null);
var ab;
item.abstractNote = ""
item.abstractNote = "";
while (ab = absx.iterateNext()) {
item.abstractNote += Zotero.Utilities.trimInternal(ab.textContent) + " ";
}
@ -182,7 +248,22 @@ function doWeb(doc, url) {
}
}
item.attachments.push({url:doc2.location.href, title:"ScienceDirect Snapshot", mimeType:"text/html"});
set.item = item;
next();
});
};
var second = function(set, next) {
var item = set.item;
Zotero.Utilities.HTTP.doGet(item.url, function(text) {
item.DOI = text.match(/>doi:([^<]*)/)[1];
try {
var aus = text.match(/<strong>\s+<p>.*<\/strong>/)[0].replace(/<sup>/g, "$").replace(/<\/sup>/g, "$");
aus = aus.replace(/\$[^$]*\$/g, "");
aus = aus.replace(/<a[^>]*>/g, "$").replace(/<\/a[^>]*>/g, "$");
@ -194,9 +275,16 @@ function doWeb(doc, url) {
item.creators.push(Zotero.Utilities.cleanAuthor(Zotero.Utilities.unescapeHTML(Zotero.Utilities.trimInternal(aus[a]), "author")));
}
}
} catch(e) {
Zotero.debug("No Authors listed.");
}
item.complete();
next();
});
}, function() {Zotero.done();});
};
var functioncallbacks = [first, second];
Zotero.Utilities.processAsync(sets, functioncallbacks, function() {Zotero.done()});
}
Zotero.wait();
}