http://www.mediawiki.org/wiki/Special:Code/pywikipedia/9622
Revision: 9622 Author: xqt Date: 2011-10-10 16:13:24 +0000 (Mon, 10 Oct 2011) Log Message: ----------- enable processing namespaces lists; get changes from family_check as generator, not as dictionary which leads to exhausted memory space
Modified Paths: -------------- trunk/pywikipedia/maintenance/update_namespaces.py
Modified: trunk/pywikipedia/maintenance/update_namespaces.py =================================================================== --- trunk/pywikipedia/maintenance/update_namespaces.py 2011-10-10 16:07:28 UTC (rev 9621) +++ trunk/pywikipedia/maintenance/update_namespaces.py 2011-10-10 16:13:24 UTC (rev 9622) @@ -1,3 +1,4 @@ +# -*- coding: utf-8 -*- """ Check the family files against the live site, and updates both the generic family.py and the site-specific family. @@ -5,9 +6,11 @@ options: -upmain Modify the main family.py, too. -wikimedia Update all the wikimedia families + <family> Work on a given wikimedia family file """ # -# (C) Pywikipedia bot team, 2003-2007 +# (C) xqt, 2010-2011 +# (C) Pywikipedia bot team, 2007-2009 # # Distributed under the terms of the MIT license. # @@ -31,14 +34,13 @@ r_namespace_def = re.compile(r'['"]([a-z_-]*)['"]\s*:\s*((?:%s)|(?:%s))\s*,' % (r_string, r_list)) def update_family(family, changes): global namespace_section_text, namespace_defs, new_defs - if family: - output(u'Updating family %s' % family.name) + output(u'\nUpdating family %s' % family.name) family_file_name = '../families/%s_family.py' % family.name r_namespace_section = r_namespace_section_sub base_indent = 8 else: - output(u'Updating family.py') + output(u'\nUpdating family.py') family_file_name = '../family.py' r_namespace_section = r_namespace_section_main base_indent = 12 @@ -46,11 +48,14 @@ old_family_text = family_text = family_file.read() family_file.close()
- for lang, namespaces in changes.iteritems(): - for namespace_id, namespace_name, predefined_namespace in namespaces: - output(u'Setting namespace[%s] for %s to %s' % (namespace_id, lang, namespace_name)) + for lang, namespaces in changes: + for namespace_id, namespace_list, predefined_namespace in namespaces: + msg = u'Setting namespace[%s] for %s to ' \ + + (u'[%s]' if len(namespace_list) > 1 else u'%s') + output(msg % (namespace_id, lang, ', '.join(namespace_list)))
- namespace_section = re.search(r_namespace_section % namespace_id, family_text) + namespace_section = re.search(r_namespace_section + % namespace_id, family_text) #namespace_section2 = re.search(r_namespace_section_once % (namespace_id, lang) ,family_text) if not namespace_section: continue @@ -58,17 +63,21 @@ namespace_defs = dict([(match.group(1), match.group(2)) for match in r_namespace_def.finditer(namespace_section_text)])
- if not namespace_defs.get(lang, '').startswith('['): - output(u'Updating namespace[%s] to %s' % (namespace_id, namespace_name)) - - namespace_defs[lang] = escape_string(namespace_name.encode('utf-8')) + msg = u'Updating namespace[%s] to ' \ + + (u'[%s]' if len(namespace_list) > 1 else u'%s') + output(msg % (namespace_id, ', '.join(namespace_list))) + if len(namespace_list) == 1: + namespace_defs[lang] = escape_string(namespace_list[0].encode('utf-8')) else: - output(u'Namespace[%s] definition is a list; not updating.' % namespace_id) + namespaces = u", ".join(escape_string(ns) for ns in namespace_list).encode('utf-8') + namespace_defs[lang] = '[%s]' % namespaces
new_defs = namespace_defs.items() new_defs.sort(key = lambda x: x[0]) - new_text = '\n' + ''.join([(base_indent + 4) * ' ' + "'%s': %s,\n" % i for i in new_defs]) + ' ' * base_indent - family_text = family_text.replace(namespace_section.group(1), new_text) + new_text = '\n' + ''.join([(base_indent + 4) * ' ' + "'%s': %s,\n" + % i for i in new_defs]) + ' ' * base_indent + family_text = family_text.replace(namespace_section.group(1), + new_text)
if family_text == old_family_text: output(u'No changes made')