{ "translatorID":"c3edb423-f267-47a1-a8c2-158c247f87c2", "translatorType":4, "label":"Common-Place", "creator":"Frederick Gibbs", "target":"http://www.common-place\\.|historycooperative\\.org/journals/cp", "minVersion":"1.0.0b4.r5", "maxVersion":"", "priority":100, "inRepository":true, "lastUpdated":"2008-06-12 19:30:00" } function detectWeb(doc, url) { if(doc.title.indexOf("Previous Issues") != -1 || doc.title.indexOf("Search Site") != -1 ) { return "multiple"; } else { return "journalArticle"; } } function scrape(doc) { var namespace = doc.documentElement.namespaceURI; var nsResolver = namespace ? function(prefix) { if (prefix == 'x') return namespace; else return null; } : null; var newItem = new Zotero.Item("journalArticle"); newItem.publicationTitle = "Common-Place"; newItem.url = doc.location.href; //get issue year and month //these will determine what xpaths we use for title and author var pubDate; var dateRe = /(.*)<\/a><\/b>/; var m = dateRe.exec(Zotero.Utilities.trimInternal(doc.getElementsByTagName("body")[0].innerHTML)); if(m) { //newItem.title = Zotero.Utilities.trimInternal(Zotero.Utilities.unescapeHTML(m[1])); pubDate = m[1]; } else { pubDate = doc.evaluate('//div[@id="container"]/div[@id="top"]/p/b/a[2]', doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent; } var d; //Zotero.debug(pubDate); pubDateVolRE = /vol. (.*) · no. /; d = pubDateVolRE.exec(pubDate); newItem.volume = d[1]; pubDateVolRE = /no. (.*) ·/; d = pubDateVolRE.exec(pubDate); newItem.issue = d[1]; pubDateVolRE = /no. [0-9] · (.*)/; d = pubDateVolRE.exec(pubDate); newItem.date = d[1]; //usually the page begins with the article title or book title (of reviewed book) //some pages have an image just before them, so we need to skip it if it's there. var pLevel; var m=doc.evaluate('//div[@id="content"]/p[1]/img', doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext(); //if there is an image here, offset the xpath if (m == null) { pLevel = '//div[@id="content"]/p[1]'; } else { pLevel = '//div[@id="content"]/p[2]'; } //issues before 2004 have a table based layout, so a totally different xpath. //check to see if we have anything, then try again if we don't. var author; var title; author = doc.evaluate(pLevel+'/span[1]', doc, nsResolver, XPathResult.ANY_TYPE, null); author = author.iterateNext(); if (author != null) { //Zotero.debug("au"+author+"au"); var title = doc.evaluate(pLevel+'/span[2]', doc, nsResolver, XPathResult.ANY_TYPE, null); //Zotero.debug("ti"+title+"ti"); title = title.iterateNext().textContent; //determine if we have a book review // if so, get the publication information if (author.textContent.indexOf("Review by") != -1 ) { newItem.title = String.concat("Review of ", title); newItem.author = author.textContent.substring(10); } else { newItem.author = author.textContent; newItem.title = title; } } else { //we have older issue //check if we are on a review var review = doc.evaluate('/html/body/table/tbody/tr/td[2]/p[2]', doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext(); var temp = review.textContent; if (temp.indexOf("Review") != -1) { title = doc.evaluate('/html/body/table/tbody/tr/td[2]/p/i', doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent; title = "Review of " + title; author = review.textContent.substring(10); } else { //for articles title = doc.evaluate('/html/body/table/tbody/tr/td[2]/p/b', doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent; author = doc.evaluate('/html/body/table/tbody/tr/td[2]/p[1]', doc, null, XPathResult.ANY_TYPE, null).iterateNext().textContent.split(/\n/)[1]; //Zotero.debug(author); } newItem.author = author; newItem.title = title; } newItem.attachments.push({document:doc, title:doc.title}); newItem.complete(); } function doWeb(doc, url) { var type = detectWeb(doc, url); if (type == "multiple") { var namespace = doc.documentElement.namespaceURI; var nsResolver = namespace ? function(prefix) { if (prefix == 'x') return namespace; else return null; } : null; //create list of items //what about home page (current issue table of contents?) //for search result links: /html/body/table[2]/tbody/tr/td[2]/li[3]/a //for previous issues: //tr/td/p/a/b (but we need to strip out volume links (starts with 'Volume') var link; var title; var items = new Object(); var searchLinks = doc.evaluate('/html/body/table[2]/tbody/tr/td[2]/li/a', doc, nsResolver, XPathResult.ANY_TYPE, null); while (elmt = searchLinks.iterateNext()) { Zotero.debug(elmt.textContent); title = elmt.textContent; link = elmt.href; if (title && link){ items[link] = title; } } items = Zotero.selectItems(items); if(!items) { return true; } var uris = new Array(); for(var i in items) { uris.push(i); } Zotero.Utilities.processDocuments(uris, function(doc) { scrape(doc) }, function() { Zotero.done(); }, null); Zotero.wait(); } else { scrape(doc); } }