Trans: New version of NZZ.ch translator by Ibex

This commit is contained in:
Avram Lyon 2010-12-20 08:21:27 +00:00
parent e681b4698f
commit ef28a3a705

View file

@ -1,14 +1,14 @@
{
"translatorID":"61ffe600-55e0-11df-bed9-0002a5d5c51b",
"translatorType":4,
"label":"nzz.ch",
"creator":"ibex",
"target":"^http://((www\\.)?nzz\\.ch/.)",
"minVersion":"2.0",
"maxVersion":"",
"priority":100,
"inRepository":false,
"lastUpdated":"2010-09-08 12:00:00"
"translatorID":"61ffe600-55e0-11df-bed9-0002a5d5c51b",
"label":"nzz.ch",
"creator":"ibex",
"target":"^http://((www\\.)?nzz\\.ch/.)",
"minVersion":"2.0",
"maxVersion":"",
"priority":100,
"inRepository":"0",
"translatorType":4,
"lastUpdated":"2010-12-20 11:17:03"
}
/*
@ -41,20 +41,20 @@ function getXPath(xpath, doc) {
/* Zotero API */
function detectWeb(doc, url) {
//Zotero.debug("ibex detectWeb URL= "+ url);
if (doc.title.substr(0, 6) == "Suche " && getXPath('//div[@class = "searchdetails"]', doc)) {
//Zotero.debug("ibex detectWeb URL= " + url);
if (doc.title.substr(0, 6) == "Suche " && getXPath('//ul[@class = "berichte"]', doc)) {
return "multiple";
} else if (doc.location.href.match(/\.\d+\.html/) && getXPath('//li[@id = "article"]/div[@class = "article"]', doc)) {
} else if (doc.location.href.match(/\.\d+\.html/) && getXPath('/html/body[@class = "artikel"]', doc)) {
return "newspaperArticle";
}
}
/* Zotero API */
function doWeb(doc, url) {
//Zotero.debug("ibex doWeb URL= "+ url);
//Zotero.debug("ibex doWeb URL= " + url);
var urls = new Array();
if (detectWeb(doc, url) == "multiple") {
var items = Zotero.Utilities.getItemArray(doc, doc.getElementById("searchresult").getElementsByTagName("h3"), '\\.\\d+\\.html');
var items = Zotero.Utilities.getItemArray(doc, doc.getElementById("content").getElementsByClassName('berichte'), '\\.\\d+\\.html');
if (!items || countObjectProperties(items) == 0) {
return true;
}
@ -75,59 +75,65 @@ function doWeb(doc, url) {
/* Three types of articles: "Neue Zürcher Zeitung", "NZZ Online" and "NZZ am Sonntag" */
function scrape(doc) {
//Zotero.debug("ibex scrape URL = "+ doc.location.href);
var newArticle = new Zotero.Item('newspaperArticle');
newArticle.url = doc.location.href;
newArticle.title = Zotero.Utilities.trimInternal(getXPath('//li[@id = "article"]/div[@class = "article"]/div[@class = "header"]//h1', doc).textContent);
//Zotero.debug("ibex scrape URL = " + doc.location.href);
var newItem = new Zotero.Item('newspaperArticle');
newItem.url = doc.location.href;
newItem.title = Zotero.Utilities.trimInternal(getXPath('//div[@id = "content"]//h1', doc).textContent);
var publ = Zotero.Utilities.trimInternal(getXPath('//li[@id = "article"]/div[@class = "article"]/div[@class = "header"]/div[@class = "pubication"]', doc).textContent);
var publ = Zotero.Utilities.trimInternal(getXPath('//div[@id = "content"]//p[@class = "dachzeile"]', doc).textContent);
publ = publ.split(',');
newArticle.date = Zotero.Utilities.trimInternal(publ[0]);
newArticle.publicationTitle = Zotero.Utilities.trimInternal(publ[publ.length - 1]);
if (newArticle.publicationTitle.match(/^\d/)) {
newItem.date = Zotero.Utilities.trimInternal(publ[0]);
newItem.publicationTitle = Zotero.Utilities.trimInternal(publ[publ.length - 1]);
if (newItem.publicationTitle.match(/^\d/)) {
//set a publication title if there is only a number (date)
newArticle.publicationTitle = "NZZ";
} else if (newArticle.publicationTitle == "Neue Zürcher Zeitung") {
newArticle.ISSN = "0376-6829";
} else if (newArticle.publicationTitle == "NZZ am Sonntag") {
newArticle.ISSN = "1660-0851";
newItem.publicationTitle = "NZZ";
} else if (newItem.publicationTitle == "Neue Zürcher Zeitung") {
newItem.ISSN = "0376-6829";
} else if (newItem.publicationTitle == "NZZ am Sonntag") {
newItem.ISSN = "1660-0851";
}
var subtitle = getXPath('//li[@id = "article"]/div[@class = "article"]/div[@class = "header"]//h2', doc);
if (subtitle != null && newArticle.publicationTitle != "NZZ am Sonntag") {
newArticle.shortTitle = newArticle.title;
newArticle.title += ": " + Zotero.Utilities.trimInternal(subtitle.textContent);
var subtitle = getXPath('//div[@id = "content"]//h2', doc);
if ((subtitle != null) && (Zotero.Utilities.trimInternal(subtitle.textContent) != "")) {
newItem.shortTitle = newItem.title;
newItem.title += ": " + Zotero.Utilities.trimInternal(subtitle.textContent);
}
var teaser = getXPath('//li[@id = "article"]/div[@class = "article"]//div[@class = "body"]/h5', doc);
if (teaser != null) {
newArticle.abstractNote = Zotero.Utilities.trimInternal(teaser.textContent);
var teaser = getXPath('//div[@id = "content"]//h3', doc);
if ((teaser != null) && (Zotero.Utilities.trimInternal(teaser.textContent) != "")) {
newItem.abstractNote = Zotero.Utilities.trimInternal(teaser.textContent);
}
var authorline = getXPath('//li[@id = "article"]/div[@class = "article"]//div[@class = "body"]/p[contains(@class, "quelle")]', doc);
authorline = !authorline && newArticle.publicationTitle == "NZZ am Sonntag"? subtitle :authorline; // subtitle in some cases of "NZZ am Sonntag"
var authorline = getXPath('//div[@id = "content"]//p[@class = "autor"]', doc);
if (authorline != null) {
authorline = Zotero.Utilities.trimInternal(authorline.textContent);
//assumption of authorline: "[Interview:|Von ]name1 [und Name2][, location]"
authorline = authorline.replace(/^.*Von /, "");
authorline = authorline.replace(/Interview: /, "");
//assumption of authorline: "[Interview:|Von ]name1[, name2] [und Name3][, location]"
authorline = authorline.replace(/^Von /, "");
authorline = authorline.replace(/^Interview: /, "");
authorline = authorline.replace(/vor Ort /i, "");
//remove ", location"
authorline = Zotero.Utilities.trim(authorline.replace(/, .*$/, ""));
authorline = Zotero.Utilities.trim(authorline.replace(/, \S*$/, ""));
var authors = authorline.split(" und ");
var authors = authorline.split(/,|und/);
for (var i = 0; i < authors.length && authorline.length > 0; i++) {
newArticle.creators.push(Zotero.Utilities.cleanAuthor(authors[i], "author"));
newItem.creators.push(Zotero.Utilities.cleanAuthor(authors[i], "author"));
}
}
var section = getXPath('//ul[@id="navContent"]/li/a[@id="navContentSelected"]', doc);
var section = getXPath('//ul[@id="navi"]//ul[@id="submenu1"]/li[@class="selected"]/a', doc);
if (section != null) {
newArticle.section = Zotero.Utilities.trimInternal(section.textContent.replace(/·/,""));
newItem.section = Zotero.Utilities.trimInternal(section.textContent);
}
newArticle.attachments.push({title:"NZZ Online Article Snapshot", mimeType:"text/html", url:doc.location.href + "?printview=true", snapshot:true});
var source = getXPath('//div[@id = "content"]//span[@class="quelle"]', doc);
if (source != null) {
newItem.extra = Zotero.Utilities.trimInternal(source.textContent).replace(/^\(/,"").replace(/\)$/,"");
}
newArticle.complete();
newItem.attachments.push({title:"NZZ Online Article Snapshot", mimeType:"text/html", url:doc.location.href, snapshot:true});
newItem.complete();
}
/* There is no built-in function to count object properties which often are used as associative arrays.*/