From a98dae78f1279397797c98f968bec2d94a2035bc Mon Sep 17 00:00:00 2001
From: emijrp <emijrp@gmail.com>
Date: Fri, 8 Apr 2011 20:35:05 +0000
Subject: [PATCH] image regexp for old wikis (wiki libsdl, wikijuegos)

git-svn-id: https://wikiteam.googlecode.com/svn/trunk@39 31edc4fc-5e31-b4c4-d58b-c8bc928bcb95
---
 dumpgenerator.py | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/dumpgenerator.py b/dumpgenerator.py
index 5671559..3aacf33 100644
--- a/dumpgenerator.py
+++ b/dumpgenerator.py
@@ -272,7 +272,16 @@ def getImageFilenamesURL(config={}, start='!'):
         raw = cleanHTML(raw)
         #archiveteam <td class="TablePager_col_img_name"><a href="/index.php?title=File:Yahoovideo.jpg" title="File:Yahoovideo.jpg">Yahoovideo.jpg</a> (<a href="/images/2/2b/Yahoovideo.jpg">file</a>)</td>
         #wikanda <td class="TablePager_col_img_user_text"><a href="/w/index.php?title=Usuario:Fernandocg&amp;action=edit&amp;redlink=1" class="new" title="Usuario:Fernandocg (página no existe)">Fernandocg</a></td>
-        m = re.compile(r'(?im)<td class="TablePager_col_img_name"><a href[^>]+title="[^:>]+:(?P<filename>[^>]+)">[^<]+</a>[^<]+<a href="(?P<url>[^>]+/[^>/]+)">[^<]+</a>[^<]+</td>\s*<td class="TablePager_col_img_user_text"><a[^>]+>(?P<uploader>[^<]+)</a></td>').finditer(raw)
+        r_images1 = r'(?im)<td class="TablePager_col_img_name"><a href[^>]+title="[^:>]+:(?P<filename>[^>]+)">[^<]+</a>[^<]+<a href="(?P<url>[^>]+/[^>/]+)">[^<]+</a>[^<]+</td>\s*<td class="TablePager_col_img_user_text"><a[^>]+>(?P<uploader>[^<]+)</a></td>'
+        #wikijuegos http://softwarelibre.uca.es/wikijuegos/Especial:Imagelist old mediawiki version
+        r_images2 = r'(?im)<td class="TablePager_col_links"><a href[^>]+title="[^:>]+:(?P<filename>[^>]+)">[^<]+</a>[^<]+<a href="(?P<url>[^>]+/[^>/]+)">[^<]+</a></td>\s*<td class="TablePager_col_img_timestamp">[^<]+</td>\s*<td class="TablePager_col_img_name">[^<]+</td>\s*<td class="TablePager_col_img_user_text"><a[^>]+>(?P<uploader>[^<]+)</a></td>'
+        m = []
+        #different mediawiki versions
+        if re.search(r_images1, raw):
+            m = re.compile(r_images1).finditer(raw)
+        elif re.search(r_images2, raw):
+            m = re.compile(r_images2).finditer(raw)
+        
         for i in m:
             url = i.group('url')
             if url[0] == '/' or not url.startswith('http://'): #relative URL