commit 66a5b1ab14ee2826df008e6e6d72458d4afb1365
parent 961c53c5fe7b0c4faf59a1999f88f97d0a09466c
Author: Avram Lyon <ajlyon@gmail.com>
Date: Sun, 15 May 2011 20:23:59 +0000
Trans: PNN update from Martin Meyerhoff
Diffstat:
1 file changed, 33 insertions(+), 63 deletions(-)
diff --git a/translators/Potsdamer Neueste Nachrichten.js b/translators/Potsdamer Neueste Nachrichten.js
@@ -8,11 +8,11 @@
"priority": 100,
"inRepository": "1",
"translatorType": 4,
- "lastUpdated": "2011-03-26 13:42:35"
+ "lastUpdated": "2011-05-06 11:34:44"
}
/*
-Potsdamer Neueste Nachrichten Translator
+Potsdamer Neueste Nachrichten Translator 1.1
Copyright (C) 2011 Martin Meyerhoff
This program is free software: you can redistribute it and/or modify
@@ -30,10 +30,9 @@ along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
/*
-The articles themselves are quite badly tagged, so that the translator sometimes doesn't capture the summary or the authors.
Test it with:
-http://www.pnn.de/archiv/?type=archiv&phrase=Krise
+http://www.pnn.de/
http://www.pnn.de/zeitung/
http://www.pnn.de/zeitung/12.01.2011/
http://www.pnn.de/titelseite/364860/
@@ -48,8 +47,8 @@ function detectWeb(doc, url) {
if (prefix == 'x') return namespace; else return null;
} : null;
- var PNN_Article_XPath = ".//a[contains(@class, 'print')]"; //only articles have a print button.
- var PNN_Multiple_XPath = ".//ul/li/h2/a"
+ var PNN_Article_XPath = ".//div[contains (@class, 'um-article')]/h1"; //only articles have a print button.
+ var PNN_Multiple_XPath = "//div[contains(@class, 'um-teaser')]/h2/a"
if (doc.evaluate(PNN_Article_XPath, doc, null, XPathResult.ANY_TYPE, null).iterateNext() ){
Zotero.debug("newspaperArticle");
@@ -70,76 +69,47 @@ function scrape(doc, url) {
newItem.url = doc.location.href;
// Title
- var title_XPath = '//title'
+ var title_XPath = "//div[contains (@class, 'um-article')]/h1"
var title = doc.evaluate(title_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- title = title.split("—")[0]; // split at mdash
- title = title.replace(/\„|\“/g, '"'); // standard quotation marks
- title = title.replace(/|^\s*|\s*$/, ''); // remove whitespace
+ title = title.replace(/\s+|\n/g, ' ');
newItem.title = title;
// Summary
- var summary_XPath = ".//p[contains(@class, 'teaser')]";
- var summary = doc.evaluate(summary_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- summary=summary.replace(/\(.*\)/, ''); // No date in the summary.
- summary=Zotero.Utilities.trimInternal(summary); //remove white space
- newItem.abstractNote = summary;
+ var summary_XPath = "//div[contains (@class, 'um-article')]/p[@class='um-first']";
+ if (doc.evaluate(summary_XPath, doc, null, XPathResult.ANY_TYPE, null).iterateNext() ){
+ var summary = doc.evaluate(summary_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
+ newItem.abstractNote = summary;
+ }
// Date
- var date_XPath = "//*[contains(@class, 'teaser')]/span[contains(@class, 'date')]";
+ var date_XPath = "//div[contains (@class, 'um-article')]/div[@class='um-metabar']/ul/li[contains(@class, 'um-first')]";
var date = doc.evaluate(date_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- date = date.replace(/\(|\)|^\s*|\s*$/g, ''); // remove whitespace and braces
- newItem.date = date;
+ newItem.date = date.replace(/(\d+)\.(\d+).(\d+)/, '$3-$2-$1');;
- // Authors. Tricky. Actually, horrible. I hope they change their site at some point and this mess can be cleaned up.
- var temp = new Array();
- temp[0] = ""
- var author_XPath = ".//*[@id='teaser']/p/i"; // Sometimes, the author is in italics in the paragraph. Easy Case, really.
- if (doc.evaluate(author_XPath, doc, null, XPathResult.ANY_TYPE, null).iterateNext()) {
- var author = doc.evaluate(author_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- temp[0] = author;
- } else {
- author_XPath = ".//*[@id='teaser']"; // basically, grab the entire article. no other chance.
+ // Authors
+ var author_XPath = "//div[contains (@class, 'um-article')]/span[@class='um-author']";
+ if (doc.evaluate(author_XPath, doc, null, XPathResult.ANY_TYPE, null).iterateNext() ){
var author = doc.evaluate(author_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- author = author.replace(/\s\s\s*/g, "|"); // replace lots of white space (indicative of a line break / paragraph)
- author = author.split("|");
-
-
- // Zotero.debug(author);
- var author_searchpattern1 = /^Von(.*)/; // These three patterns capture the majority of authors.
- var author_searchpattern2 = /^Das\sGespräch\sführte(.*)\.$/;
- var author_searchpattern3 = /^Interview\:\s(.*)Foto:.*/;
-
- for (var i in author) {
- if (temp[0] == "") {
- if (author[i].match(author_searchpattern1)) {
- var temp = author[i].match(author_searchpattern1);
- temp[0] = temp[0].replace(author_searchpattern1, "$1");
- }
- if (author[i].match(author_searchpattern2)) {
- var temp = author[i].match(author_searchpattern2);
- temp[0] = temp[0].replace(author_searchpattern2, "$1");
- }
- if (author[i].match(author_searchpattern3)) {
- var temp = author[i].match(author_searchpattern3);
- temp[0] = temp[0].replace(author_searchpattern3, "$1");
- }
+ author =author.replace(/^von\s|^\s*|\s*$/g, '');
+ author =author.split(/\sund\s|\su\.\s|\,\s/);
+ for (var i in author) {
+ if (author[i].match(/\s/)) { // only names that contain a space!
+ newItem.creators.push(Zotero.Utilities.cleanAuthor(author[i], "author"));
}
- }
- }
- var realauthor = temp[0].replace(/^\s*|\s*$/g, '');
- realauthor = realauthor.split(/\sund\s|\su\.\s|\,\s/);
- for (var i in realauthor) {
- if (realauthor[i].match(/\s/)) { // only names that contain a space!
- newItem.creators.push(Zotero.Utilities.cleanAuthor(realauthor[i], "author"));
}
}
+
newItem.attachments.push({url:doc.location.href, title:doc.title, mimeType:"text/html"});
newItem.publicationTitle = "Potsdamer Neueste Nachrichten"
+
// section
- var section_XPath = ".//*[@id='sidebar-left']/ul/li[contains(@class, 'active')]";
- var section = doc.evaluate(section_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
- newItem.section = section.replace(/^\s*|\s*$/g, '');
+ var section_XPath = "//div[@class='um-mainnav']/ul/li[@class='um-selected']/a";
+ if (doc.evaluate(section_XPath, doc, null, XPathResult.ANY_TYPE, null).iterateNext() ){
+ var section = doc.evaluate(section_XPath, doc, nsResolver, XPathResult.ANY_TYPE, null).iterateNext().textContent;
+ newItem.section = section.replace(/^\s*|\s*$/g, '');
+ }
+
newItem.complete();
}
@@ -155,19 +125,19 @@ function doWeb(doc, url) {
if (detectWeb(doc, url) == "multiple") {
var items = new Object();
- var titles = doc.evaluate(".//ul/li/h2/a", doc, nsResolver, XPathResult.ANY_TYPE, null);
+ var titles = doc.evaluate("//div[contains(@class, 'um-teaser')]/h2/a", doc, nsResolver, XPathResult.ANY_TYPE, null);
var next_title;
while (next_title = titles.iterateNext()) {
- items[next_title.href] = next_title.textContent;
+ items[next_title.href] = next_title.textContent.replace(/\s+/g, ' ');
}
items = Zotero.selectItems(items);
for (var i in items) {
articles.push(i);
}
Zotero.Utilities.processDocuments(articles, scrape, function() {Zotero.done();});
- Zotero.wait();
} else {
scrape(doc, url);
}
+ Zotero.wait();
}
