From 10118480ccd373f5bc09e9ef10298ad99e79f3b2 Mon Sep 17 00:00:00 2001 From: Glandos Date: Sat, 2 Nov 2013 09:36:16 +0100 Subject: Update lemonde.fr.txt Authors have now their own tag. Put modification date first, and published date if modification date not found. Use id for body. Fix the stripped components to really strip unwanted links. --- lemonde.fr.txt | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/lemonde.fr.txt b/lemonde.fr.txt index 7aaa77a..1bf54b5 100644 --- a/lemonde.fr.txt +++ b/lemonde.fr.txt @@ -1,13 +1,18 @@ title: //h1 -# they have a single component containing both author and date -#author: //p[@class='source'] -#date: //p[@class='source'] +# We can have multiple authors +author: //a[@class='auteur'] -body: //div[@class='contenu_article'] +# Last edition date (if any) +date: //time[@itemprop='dateModified']/@datetime +# Publication date +date: //time[@itemprop='datePublished']/@datetime + + +body: //div[@id='articleBody'] #Shoot the insane "conjugaison.lemonde.fr" links : -strip: //a[contains(@class, 'listLink')] +strip: //a[contains(@class, 'conjug')] prune: no -test_url: http://www.lemonde.fr/economie/article/2011/07/05/moody-s-abaisse-la-note-du-portugal-de-quatre-crans_1545237_3234.html \ No newline at end of file +test_url: http://www.lemonde.fr/economie/article/2011/07/05/moody-s-abaisse-la-note-du-portugal-de-quatre-crans_1545237_3234.html -- cgit v1.3.1