From cc9f064e51f31b553e067dc3f8b1bfc024094702 Mon Sep 17 00:00:00 2001 From: emijrp Date: Tue, 5 Apr 2011 22:08:53 +0000 Subject: [PATCH] suballpages git-svn-id: https://wikiteam.googlecode.com/svn/trunk@4 31edc4fc-5e31-b4c4-d58b-c8bc928bcb95 --- dumpgenerator.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/dumpgenerator.py b/dumpgenerator.py index 1553440..7774a8c 100644 --- a/dumpgenerator.py +++ b/dumpgenerator.py @@ -81,7 +81,7 @@ def getAllPageTitles(domain='', namespaces=[]): deep = 3 # 3 is the current deep of English Wikipedia for Special:Allpages, 3 levels c = 0 checked_suballpages = [] - rawacum = '' + rawacum = raw while re.search(r_suballpages, raw) and c < deep: #load sub-Allpages m = re.compile(r_suballpages).finditer(raw) @@ -150,8 +150,8 @@ def cleanXML(xml=''): return xml if __name__ == '__main__': - domain = 'http://archiveteam.org/index.php' # 'http://en.wikipedia.org/w' - #domain = 'http://wikanda.cadizpedia.eu/w/index.php' # 'http://en.wikipedia.org/w' + #domain = 'http://archiveteam.org/index.php' # 'http://en.wikipedia.org/w' + domain = 'http://wikanda.cadizpedia.eu/w/index.php' # 'http://en.wikipedia.org/w' curonly = False namespaces = [0]