User:WindBOT/Filters
How to disable a filter
If the bot is malfunctioning, chances are that the problem lies in one of these blocks of code. Thus, instead of shutting down the whole bot, it would be wiser to disable only the chunk of code that is misbehaving. To make the bot ignore a certain line, add a "#" in front of it:
# This line will be ignored
If there are multiple lines, wrap them inside triple-quotes (you still need to put the two spaces at the beginning of the line):
"""This line will be ignored and this one as well and this one is cake and the previous one was a lie but it was still ignored"""
If all else fails, you can simply delete the block from the page. The bot can't come up with code by itself yet, so it won't run anything. Or, if the problem really is elsewhere, block the bot.
Contents
- 1 How to disable a filter
- 2 Global filters
- 3 Page filters
- 4 Semantic filters
- 5 Language-specific filters
- 6 Link filters
- 7 Template filters
- 7.1 Template renaming
- 7.2 Manage all {{Item infobox}}es
- 7.3 Remove useless templates
- 7.4 Remove manual video IDs from {{Weapon Demonstration}}
- 7.5 Filter parameters of certain templates
- 7.6 Remove obsolete parameters
- 7.7 Fix map infobox attribute
- 7.8 Implement Backpack Item Link template
- 7.9 Implement {{Dictionary}}
- 8 File filters
- 9 Additional tasks
Global filters
File categorization (Add {{No category}})
def fileCategorization(content, **kwargs):
badLoneCategories = [u'Category:Screenshot images', u'Category:Artwork images', u'Category:Multimedia files', u'Category:Extracted media', u'Category:Images without license tags', u'Category:Images that need improving', u'Category:Images using ((Another license)) incorrectly']
extraStuff = u'{{No category}}'
regLang = compileRegex('/[^/]+$')
if 'article' not in kwargs or content.find(u'#REDIRECT ') != -1:
return content
title = u(kwargs['article'].title)
if title[:5].lower() != u'file:':
return content
for c in kwargs['article'].getCategories():
if u(regLang.sub(u'', c)) not in badLoneCategories:
return content
if not len(content):
return extraStuff
if content.find(extraStuff) != -1:
return content
return content.strip() + u'\n' + extraStuff
addFilter(fileCategorization)
Page filters
addPageFilter(r'^user:', r'(?:talk|help|wiki|template):')
Semantic filters
Item names
categories = [ # These are categories that contain that name of things that should be capitalized
'Weapons',
'Hats',
'Miscellaneous items',
'Buildings'
]
exceptions = [ # Those are page names that are in the above categories but should not count as capitalized weapon names
'Direct Hit',
'Hats', 'Weapons', 'Miscellaneous items', 'Buildings',
'Item levels',
'Developer weapons',
'Self-made items',
'Unused content',
'Promotional items',
'Mercenary', 'Taunts', 'Decapitation', 'Fists', 'Cloak', # Too common to be reliably replaced
'Club', 'Bat', 'Knife', 'Syringe', 'Bottle', 'Original', # Ditto (weapons)
'Classified', # Not capitalized even when referring to the hat
'Attendant', # Avoids issue with French
'Buff Banner', 'Medi Gun', 'Bonk Helm', # Avoids issue with German
'Reskins' # Not capitalized
]
for c in categories:
c = wikitools.category.Category(wiki(), u(c))
for p in pageFilter(c.getAllMembers(titleonly=True)):
if p not in exceptions:
enforceCapitalization(p)
# Special case: Direct Hit
addSafeFilter(wordFilter('Direct Hit', r'(?<!a )(?<!one )(?<!on )(?<!\()(?<!placed )(?<!single )\bDirect Hit'))
# Special case: Attendant
enforceCapitalization('Attendant', languageBlacklist=['fr'])
# Special cases: "Buff-Banner"/"Medigun" in German
enforceCapitalization('Buff Banner', 'Medi Gun', languageBlacklist=['de'])
enforceCapitalization('Buff-Banner', 'Medigun', language='de')
# Put "an" in front of Electro Sapper
addSafeFilter(
dumbReplace('a Electro Sapper', 'an Electro Sapper'),
dumbReplace('A Electro Sapper', 'An Electro Sapper')
)
# Flare Gun
addSafeFilter(wordFilter('Flare Gun', 'flaregun'))
Classes
# Not "Heavy" because the word can be used as an adjective:
classes = ['Scout', 'Soldier', 'Pyro', 'Demoman', 'Heavy Weapons Guy', 'Engineer', 'Medic', 'Sniper'] # Spy added later
classesPlurals = ['Scouts', 'Pyros', 'Demomen', 'Heavy Weapons Guys', 'Heavies', 'Engineers', 'Snipers', 'Spies']
enforceCapitalization(*classes)
enforceCapitalization(*classesPlurals)
# Special Spy case:
addSafeFilter(wordFilter('Spy', '(?<!I )Spy'))
classes.append('Spy') # Used in later functions with Spy included
BLU/RED
DISABLED because of TFC's team names
"""addSafeFilter(wordFilter('BLU', 'BLU(?![- ]+ray)'))
# Can't force simply "red" because it would ruin things like "this red hat" or something. However, the following can probably be assumed to be correct:
for c in classes + classesPlurals + ['Soldiers']:
enforceCapitalization('RED ' + c) # Capitalize "RED Scout", "RED Demoman", etc
enforceCapitalization('RED team', 'RED Heavy', 'RED Intelligence')
# Similarly, "blue" can be converted to "BLU" in the same cases:
for c in classes + classesPlurals + ['Soldiers']:
addSafeFilter(wordFilter('BLU ' + c, 'blue? ' + c)) # Convert "blue scout" -> "BLU Scout", "blue demoman" -> "BLU Demoman", etc.
addSafeFilter(
wordFilter('BLU team', 'blue? team'),
wordFilter('BLU Heavy', 'blue? Heavy'),
wordFilter('BLU Intelligence', 'blue? intelligence')
)"""
Other capitalized words
enforceCapitalization('Team Fortress 2', 'TF2')
enforceCapitalization('Payload', 'Cloak', 'Overtime', 'Dispensers', 'Sappers')
enforceCapitalization('PlayStation', 'Xbox')
enforceCapitalization('iPod', 'iPhone')
Word aliases
addSafeFilter(
wordFilter(u'Übersaw', u'[üu]bersaw'), # Ubersaw
wordFilter(u'Force-a-Nature', u'Force of Nature'), # Force-A-Nature
wordFilter(u'ÜberCharges', u'[Üüu]ber(?!säge)(?!sage)(?! Entertainment)(?: ?charge)?s'), # ÜberCharges
wordFilter(u'Intelligence', u'Intel(?!\\s*(?:CPU|processor))'), # Intelligence (maybe adding "flag" in there would be too agressive)
wordFilter(u'Electro Sapper', u'(?<!Elektro.)(?<!Electro.)(?<!Ultra.)Sapper', u'Electro-?Sapper'),
wordFilter(u'Heavy Weapons Guy', u'Heavy Guy'),
wordFilter(u'Intelligence room', u'Intel(?:ligence)? room'),
wordFilter(u'Intelligence briefcase', u'Intel(?:ligence)? briefcase'),
wordFilter(u'K.G.B.', u'KGB'),
wordFilter(u'Chieftain', u'Chieftian', u'Chieftan'),
wordFilter(u'Batallion\'s Backup', u'Batallion\'?s? Back-?up'),
wordFilter(u'Mann-Conomy', u'Mann?-?Conomy'),
wordFilter(u'First-person view', u'1(?:st)? person view'),
wordFilter(u'Über Update', u'[UÜü]ber update')
#wordFilter(u'Stickybomb', u'Stickybomb', u'Sticky bomb'),
#wordFilter(u'Stickybombs', u'Stickybombs', u'Sticky bombs')
)
addSafeFilter(
wordFilter(u'Medi Gun', u'Medi-?Gun'),
languageBlacklist=['fr', 'de', 'pl']
)
Sentry Gun
addSafeFilter(
wordFilter(u'Sentry Gun', r"(?<!Mini.)(?<!Mah.)(?<!My.)Sentry(?![-\s]*(?:Gun|here|ahead|forward|up there|right up|, right up|'{2,}|jump)|-)"),
wordFilter(u'Sentry Guns', r'Sentry Guns', r'Sentries(?!\s+Gun)'),
wordFilter(u'Combat Mini-Sentry Gun', r'Combat Mini-Sentry Gun', r'(?:(?:Mini|Combat)\s+)+Sentry(?!\s+Gun|-)'),
wordFilter(u'Combat Mini-Sentry Guns', r'(?:(?:Mini|Combat)[-\s]+)+Sentr(?:y[-\s]+Guns|ies)')
)
addFilter(regex(r'\[\[([^][|]+\|)?(?:Sentry Guns|Sentries)\]\]', '$1Sentry Guns')) # Put the "s" out of the link
Common misspellings
addSafeFilter(
wordFilter('Huntsman', 'Hunstman'),
wordFilter('Dispenser', 'Dis?pen[sc][eo]r'),
wordFilter('Heavy', 'Hevy'),
wordFilter('Engineer', 'Enginer'),
wordFilter('Soldier', 'Solider'),
wordFilter('Mini-Crit', 'Minicrit'),
wordFilter('Mini-Crits', 'Minicrits'),
wordFilter('Chargin\' Targe', 'charg[ei][-\'n\\s]*targe?'),
wordFilter('Kritzkrieg', 'Kritzkreig'),
wordFilter('screenshot', 'screen shot'),
wordFilter('screenshots', 'screen shots'),
wordFilter('in-game', 'ingame'),
wordFilter('team-colored', 'team colou?red', keepcapitalization=True)
# wordFilter('color', '(?<!Rustic )colour')
)
addSafeFilter(
wordFilter('Natascha', 'Natas?c?ha'),
language='en'
)
addSafeFilter(
wordFilter('Spies', 'Spys'),
languageBlacklist=['de', 'es']
)
Map names
addSafeFilter(
wordFilter('Gravel Pit', '(?<!cp_)Gravelpit', 'Gravel pit'),
wordFilter('Double Cross', 'Double Cross', '(?<!ctf_)Doublecross'),
wordFilter('Badwater Basin', '(?<!pl_)Badwater Basin', '(?<!pl_)Badwater(?!\s+Basin)'),
wordFilter('Gold Rush', '(?<!pl_)Goldrush')
)
# Capitalisation rules for map names are whitelist-based rather than Category:Maps+blacklist based, because lots of maps are common nouns.
# As such, if a map with a non-common name is added, please add it to this list.
enforceCapitalization(
'2Fort', 'Badlands', 'Coldfront', 'Dustbowl', 'Egypt', 'Fastlane', 'Granary', 'Gullywash',
'Hightower', 'Hoodoo', 'Landfall', 'Offblast', 'Thunder Mountain',
'Viaduct', 'Wildfire', 'Yukon'
)
Section headers
addSafeFilter(
wordFilter(u'== Update history ==', u'==+ ?(?:Update history|Previous changes) ?==+'),
wordFilter(u'== See also ==', u'==+ ?See also ?==+'),
wordFilter(u'== External links ==', u'==+ ?External links ?==+'),
wordFilter(u'== Painted variants ==', u'==+ ?Painted variants ?==+'),
wordFilter(u'== Item set ==', u'==+ ?Item set ?==+'),
wordFilter(u'== Damage and function times ==', u'==+ ?Damage and function times ?==+'),
wordFilter(u'=== As a crafting ingredient ===', u'==+ ?As a crafting ingredient ?==+'),
wordFilter(u'== Unused content ==', u'==+ ?Unused content ?==+'),
wordFilter(u'== See also ==', u'==+ ?See also ?==+'),
wordFilter(u'== Related achievements ==', u'==+ ?Related achievements ?==+'),
wordFilter(u'== Strange variant ==', u'==+ ?Strange variant ?==+')
)
Language-specific filters
Language-agnostic
# Language-agnostic achievements auto-translate
def translateAchievements(language, pageSuffix):
try:
tf = readLocaleFile(urllib2.urlopen(page('File:Tf_' + language + '.txt').getDownloadUrl()).read(-1))
except:
print 'Downloading failed.'
return
try:
languages = parseLocaleFile(tf, language=language)
except:
print 'Error while parsing tf_' + language + '.txt'
return
acceptPrefix = ['TF_SCOUT_', 'TF_SOLDIER_', 'TF_PYRO_', 'TF_DEMOMAN_', 'TF_HEAVY_', 'TF_ENGINEER_', 'TF_MEDIC_', 'TF_SNIPER_', 'TF_SPY_', 'TF_GET_', 'TF_KILL_', 'TF_BURN_', 'TF_WIN_', 'TF_PLAY_']
acceptSuffix = ['_NAME', '_DESC']
filteredAchievements = languagesFilter(languages, commonto=[language, 'english'], prefix=acceptPrefix, suffix=acceptSuffix, exceptions=['TF_SOLDIER_ASSIST_MEDIC_UBER_NAME'])
associateLocaleWordFilters(filteredAchievements, 'english', language, pageSuffix)
Romanian filters
# Romanian characters
addSafeFilter(
dumbReplaces({
u'ş': u'ș',
u'ţ': u'ț'
}), language='ro'
)
# Romanian achievements (disabled for now)
#translateAchievements('romanian', 'ro')
# Fix User:Vulturas's stoopid:
addSafeFilter(
wordFilter(u'batjocură', u'batjocoră'),
wordFilter(u'batjocura', u'batjocora'),
wordFilter(u'batjocureşte', u'batjocoreşte', u'batjocorește'),
wordFilter(u'batjocuri', u'batjocori'),
wordFilter(u'batjocurile', u'batjocorile'),
wordFilter(u'batjocurii', u'batjocorii'),
language='ro'
)
German filters
addSafeFilter( # Requested by Picard
wordFilter(u'Krit-\'n-Cola', u'[CK]rit-\'?[an]\'?-Cola'),
wordFilter(u'Krit', u'Crit'),
wordFilter(u'Krits', u'Crits'),
wordFilter(u'Sonstiger Gegenstand', u'Diverser Gegenstand', u'diverser Gegenstand', u'sonstiger Gegenstand'),
wordFilter(u'Sonstige Gegenstände', u'Diverse Gegenstände', 'diverse Gegenstände', 'sonstige Gegenstände'),
language='de'
)
# German achievements (disabled for now)
#translateAchievements('german', 'de')
addSafeFilter(
wordFilter(u'== Update-Verlauf ==', u'==+ *(?:Update Verlauf|Letzte [Ääa]nderungen) *==+'), language='de'
)
Spanish filters
addSafeFilter(
wordFilter(u'Forajido', u'Gunslinger'),
wordFilter(u'Pistolón', u'Gran Masacre'),
wordFilter(u'Medic', u'M[eé]dico(?! Medieval)'),
wordFilter(u'Medics', u'M[eé]dicos(?! Medieval)'),
wordFilter(u'Sniper', u'(?<!Rifle de )Francotirador'),
wordFilter(u'Snipers', u'(?<!Rifle de )Francotiradore?s'),
wordFilter(u'Spy', u'Esp[ií]a'),
wordFilter(u'Spies', u'Spys', u'Esp[ií]as'),
wordFilter(u'Soldier', u'Soldado(?! de Fortuna)'),
wordFilter(u'Soldiers', u'Soldados'),
wordFilter(u'Engineer', u'(?<!Gorra de )Ingeniero'),
wordFilter(u'Engineers', u'Ingenieros'),
wordFilter(u'Curiosidades', u'Trivia'),
wordFilter(u'Actualizacion Mann-Conomy', u'Mann-Conomy Update'),
language='es'
)
class spanishDateFilter: # Requested by BiBi def __init__(self): self.reg = compileRegex(r'(?:Parche|Patch) del? (\d{1,2}) (?:del?)? (Enero|Febrero|Marzo|Abril|Mayo|Ju[nl]io|Agosto|Septiembre|Oct[ou]bre|Noviembre|Diciembre) de (\d{4})') self.filterName = u'Spanish date consistency filter' def replace(self, match): return u'Parche del ' + u(match.group(1)) + u' de ' + u(match.group(2)).title() + u' de ' + u(match.group(3)) def __call__(self, content, **kwargs): return self.reg.sub(self.replace, content) addSafeFilter(spanishDateFilter(), language='es')
addSafeFilter( # Requested by Dio
wordFilter(u'== Variaciones de Colores ==', u'==+ *Variantes +pintadas *==+'), language='es'
)
addSafeFilter( # Requested by Dio
wordFilter(u'La trampa', u'El cheto', keepcapitalization=True),
wordFilter(u'Las trampas', u'Los chetos', keepcapitalization=True),
wordFilter(u'Trampas', u'Chetos', keepcapitalization=True),
wordFilter(u'Trampa', u'Cheto', keepcapitalization=True),
language='es'
)
addSafeFilter( # Requested by Jagoba RL, then flipped by request of Dio
wordFilter(u'SuperActualización', u'Actualizaci[oó]n [Üüu]ber'),
language='es'
)
French filters
enforceCapitalization(
u'janvier', u'février', u'mars', u'avril', u'mai', u'juin', u'juillet', u'août', u'septembre', u'octobre', u'novembre', u'décembre', language='fr'
) # Do not capitalize months
addSafeFilter(wordFilter('janvier', 'janviers'), language='fr')
addSafeFilter(
wordFilter(u'== Historique des mises à jour ==', u'==+ *(?:Historique des? mises? [aà] jour|Changements? pr.c.dents?) *==+'), language='fr'
)
Brazilian Portuguese filters
def portugueseInfoboxAvailabilityFilter(t, **kwargs): # Requested by Cructo
if t.getName().lower() == u'item infobox':
return t # Skip
if t.getParam('availability'):
t.setParam('availability', t.getParam('availability').replace(u'craft', u'fabricação').replace(u'Craft', u'Fabricação'))
return t
addTemplateFilter(portugueseInfoboxAvailabilityFilter, language='pt-br')
addSafeFilter( # Requested by Cructo
wordFilter(u'Fabricação', u'Crafting', keepcapitalization=True),
wordFilter(u'Conquista', u'Achievement', keepcapitalization=True),
wordFilter(u'Conquistas', u'Achievements', keepcapitalization=True),
language='pt-br'
)
addSafeFilter( # Requested by EpicEric
wordFilter(u'Demomen', u'Demomans'),
wordFilter(u'Heavies', u'Heavys'),
wordFilter(u'N/D', u'N/A'),
wordFilter(u'Atualização', u'Patch', keepcapitalization=True),
wordFilter(u'Atualizações', u'Patches', keepcapitalization=True),
wordFilter(u'Desempenho', u'Performance', keepcapitalization=True),
wordFilter(u'Motor Source', u'Source Engine'),
wordFilter(u'Spies', u'Spys'),
wordFilter(u'ÜberCarga', u'Sobrecarga', u'Übercharge'),
wordFilter(u'Mochila', u'Backpack', keepcapitalization=True),
wordFilter(u'Comunidade Steam', u'Steam Community'),
wordFilter(u'Corrigido um bug', u'Corrigido um defeito', keepcapitalization=True),
wordFilter(u'== Bugs ==', u'==+ *Defeitos *==+'),
wordFilter(u'Sala de renascimento', u'Respawn room', u'Sala de respawn', keepcapitalization=True),
wordFilter(u'Área de renascimento', u'Respawn area', u'Área de respawn', keepcapitalization=True),
wordFilter(u'Über Atualização', u'Atualização Über', u'Atualização do Über', u'Über Update'),
wordFilter(u'Sistema de obtenção de itens', u'Sistema de drop de itens', u'Item drop system', u'Sistema de queda de itens', keepcapitalization=True),
wordFilter(u'Rajada de ar', u'Compression blast', u'Airblast', keepcapitalization=True),
wordFilter(u'minicrits', u'Mini-Crits', u'Mini-Críticos', u'Minicríticos'),
wordFilter(u'Blog Oficial do TF2', u'TF2 Official Blog', u'Blog Oficial de TF2'),
wordFilter(u'Personalizada', u'Customized', keepcapitalization=True),
wordFilter(u'Genuína', u'Genuine', keepcapitalization=True),
wordFilter(u'Feito por mim', u'Self-Made', keepcapitalization=True),
wordFilter(u'Padrão', u'Stock', keepcapitalization=True),
wordFilter(u'Estranha', u'Strange', keepcapitalization=True),
wordFilter(u'Única', u'Unique', keepcapitalization=True),
wordFilter(u'Incomum', u'Unusual', keepcapitalization=True),
wordFilter(u'Compartimento', u'Slot', keepcapitalization=True),
wordFilter(u'Jogabilidade', u'Gameplay', keepcapitalization=True),
wordFilter(u'Rei do Pedaço', u'King of the Hill'),
wordFilter(u'Disparo-Alt', u'Alt-Fire', u'Fogo Alternativo', u'Disparo Alternativo', keepcapitalization=True),
wordFilter(u'Botão de Disparo-Alt', u'Botão Direito do Mouse', keepcapitalization=True),
wordFilter(u'Golden Wrench', u'Golden Chave Inglesa'),
wordFilter(u'Habilidades', u'Abilidades', keepcapitalization=True),
wordFilter(u'Austrálio', u'Australium', keepcapitalization=True),
language='pt-br'
)
enforceCapitalization( # Do not capitalize languages, nationalities, months or days of the week (will not include abbrevs. of weekdays, as they are also ordinal numbers) u'inglês', u'russo', u'russa', u'francês', u'francesa', u'alemão', u'alemã', u'polonês', u'polonesa', u'brasileiro', u'brasileira', u'finlandês', u'finlandesa', u'castelhano', u'espanhol', u'espanhola', u'holandês', u'holandesa', u'chinês', u'chinesa', u'árabe', u'tcheco', u'tcheca', u'dinamarquês', u'dinamarquesa', u'húngaro', u'húngara', u'italiano', u'italiana', u'japonês', u'japonesa', u'coreano', u'coreana', u'norueguês', u'norueguesa', u'português', u'portuguesa', u'romeno', u'romena', u'sueco', u'sueca', u'turco', u'egípcio', u'egípcia', u'janeiro', u'fevereiro', u'março', u'abril', u'maio', u'junho', u'julho', u'agosto', u'setembro', u'outubro', u'novembro', u'dezembro', u'segunda-feira', u'terça-feira', u'quarta-feira', u'quinta-feira', u'sexta-feira', u'sábado', u'domingo', language='pt-br' )
addSafeFilter( # Requested by Cructo
wordFilter(u'Espingarda', u'Scattergun'),
wordFilter(u'Pistola', u'Pistol'),
wordFilter(u'Taco', u'Bat'),
wordFilter(u'Lança-Foguetes', u'Rocket Launcher'),
wordFilter(u'Escopeta', u'Shotgun'),
wordFilter(u'Pá', u'Shovel'),
wordFilter(u'Lança-Chamas', u'Flamethrower'),
wordFilter(u'Machado de Incêndio', u'Fire Axe'),
wordFilter(u'Lança-Granadas', u'Grenade Launcher'),
wordFilter(u'Lança-Stickybombs', u'Stickybomb Launcher'),
wordFilter(u'Garrafa', u'Bottle'),
wordFilter(u'Metralhadora Giratória', u'Minigun'),
wordFilter(u'Punhos', u'Fists'),
wordFilter(u'Ferramenta de Construção', u'Build PDA'),
wordFilter(u'Ferramenta de Demolição', u'Destroy PDA'),
wordFilter(u'Arma de Seringas', u'Syringe Gun'),
wordFilter(u'Arma Médica', u'Medi Gun'),
wordFilter(u'Serra de Ossos', u'Bone Saw'),
wordFilter(u'Rifle de Sniper', u'Sniper Rifle'),
wordFilter(u'Submetralhadora', u'SMG'),
wordFilter(u'Revólver', u'Revolver'),
wordFilter(u'Faca', u'Knife'),
wordFilter(u'Relógio de Invisibilidade', u'Invis Watch'),
wordFilter(u'Kit de Disfarce', u'Disguise Kit'),
language='pt-br'
)
Russian filters
addSafeFilter(
regexes({
(u'^([^<>]*)(?<!=)([""])((?:(?!\\2|[=<>]).)+)\\2([^<>]*)$', re.MULTILINE): u'$1«$3»$4',
u'«([^»]*)(?<!\')\'([^»\']+)\'(?!\')': u'«$1„$2“'
}),
language='ru'
)
Korean filters
addSafeFilter(dumbReplace(u'솔져', u'솔저'), language='ko') # Requested by Cyrus H.
Dutch filters
#addSafeFilter(wordFilter(u'Niveau', u'Level', keepcapitalization=True), language='nl') # Requested by Apparition; too broad
addSafeFilter( # Requested by Warlike and Heifastus
wordFilter(u'Updateverleden', u'Updategeschiedenis', keepcapitalization=True),
wordFilter(u'Galerij', u'Gallerij', keepcapitalization=True),
language='nl'
)
Link filters
Moved links
def movedLinks(link, **kwargs):
movedLinks = {
u'Quality#Normal': u'Normal',
u'Quality#Unique': u'Unique',
u'Quality#Vintage': u'Vintage',
u'Quality#Genuine': u'Genuine',
u'Quality#Strange': u'Strange',
u'Quality#Unusual': u'Unusual',
u'Quality#Community': u'Community (quality)',
u'Quality#Self-Made': u'Self-Made',
u'Quality#Valve': u'Valve (quality)'
}
if link.getType() == u'internal' and link.getLink() in movedLinks:
link.setLink(movedLinks[link.getLink()])
return link
addLinkFilter(movedLinks)
Wikipedia links filter
def wikipediaLinks(link, **kwargs):
wikipediaRegex = compileRegex(r'^https?://(?:(\w+)\.)?wikipedia\.org/wiki/(\S+)')
if link.getType() == u'external':
linkInfo = wikipediaRegex.search(link.getLink())
if linkInfo:
link.setType(u'internal')
try:
wikiPage = urllib2.unquote(str(linkInfo.group(2))).decode('utf8', 'ignore').replace(u'_', ' ')
except:
wikiPage = u(linkInfo.group(2)).replace(u'_', ' ')
if not linkInfo.group(1) or linkInfo.group(1).lower() == u'en':
link.setLink(u'Wikipedia:' + wikiPage) # English Wikipedia
else:
link.setLink(u'Wikipedia:' + linkInfo.group(1).lower() + u':' + wikiPage) # Non-english Wikipedia
if link.getLabel() is None:
link.setLabel(u'(Wikipedia)')
return link
addLinkFilter(wikipediaLinks)
TF2Wiki links filter
def tf2wikiLinks(link, **kwargs):
tf2wikiRegex1 = compileRegex(r'^https?://[-.\w]*tf2wiki\.net/wiki/(\S+)$')
tf2wikiRegex2 = compileRegex(r'^https?://[-.\w]*tf2wiki\.net/w[-_/\w]+?/([^/\s]+)$')
if link.getType() == 'external':
linkInfo = tf2wikiRegex1.search(link.getLink())
isMedia = False
if not linkInfo:
linkInfo = tf2wikiRegex2.search(link.getLink())
isMedia = True
if linkInfo:
link.setType('internal')
try:
wikiPage = u(urllib2.unquote(str(linkInfo.group(1))).decode('utf8', 'ignore').replace(u'_', ' '))
except:
wikiPage = u(linkInfo.group(1)).replace(u'_', ' ')
label = wikiPage
if isMedia:
if wikiPage[-4:].lower() == '.wav':
wikiPage = 'Media:' + wikiPage
else:
wikiPage = ':File:' + wikiPage
link.setLink(wikiPage)
if link.getLabel() is None:
link.setLabel(label)
return link
addLinkFilter(tf2wikiLinks)
Category removal on pages using {{Item infobox}}
def removeCategory(l, **kwargs):
catsToRemove = [u'Category:Weapons', u'Category:Hats', u'Category:Primary weapons', u'Category:Secondary weapons', u'Category:Melee weapons', u'Category:PDA1 weapons', u'Category:PDA2 weapons', u'Category:Miscellaneous items', u'Category:Tools', u'Category:Action items', u'Category:Taunts', u'Community-contributed items']
regLang = compileRegex('/[^/]+$')
if 'article' not in kwargs or regLang.sub(u, l.getLink()) not in catsToRemove:
return l
if u'Category:Item infobox usage' not in kwargs['article'].getCategories():
return l
return None
addLinkFilter(removeCategory)
FPSBanana to GameBanana
addLinkFilter(linkDomainFilter('fpsbanana.com', 'gamebanana.com'))
Remove trailing slashes from internal links
def removeTrailingSlash(l, **kwargs):
if l.getType() != u'internal':
return l
if l.getLink()[-1] == '/':
l.setLink(l.getLink()[:-1])
return l
addLinkFilter(removeTrailingSlash)
Convert patch links to {{Patch name}}
def patchNameLinkFilter(l, **kwargs):
if l.getType() != u'internal':
return l
regPachName = compileRegex(u'(January|February|March|April|May|June|July|August|September|October|November|December)\\s+(\\d+),\\s+(\\d{4,})\\s+Patch(?:/\\w+)?')
result = regPachName.match(l.getLink())
if result is None or l.getLabel().find(result.group(2)) == -1 or l.getLabel().find(result.group(3)) == -1:
return l
monthNames = ('january', 'february', 'march', 'april', 'may', 'june', 'july', 'august', 'september', 'october', 'november', 'december')
beta = u
if l.getLink().lower().find(u'beta') != -1:
beta = u'|beta'
return template(u'{{Patch name|' + u(monthNames.index(result.group(1).lower()) + 1) + u'|' + u(result.group(2)) + u'|' + u(result.group(3)) + beta + u'}}')
addLinkFilter(patchNameLinkFilter)
Template filters
Template renaming
def templateRenameMapping(t, **kwargs):
templateMap = {
# Format goes like this (without the "#" in front obviously):
#'Good template name': ['Bad template lowercase name 1', 'Bad template lowercase name 2', 'Bad template lowercase name 3'],
# Last line has no comma at the end
'Scout Nav': ['scout nav/ro'],
'Soldier Nav': ['soldier nav/ro'],
'Pyro Nav': ['pyro nav/ro'],
'Heavy Nav': ['heavy nav/ro'],
'Demoman Nav': ['demoman nav/ro'],
'Engineer Nav': ['engineer nav/ro'],
'Medic Nav': ['medic nav/ro'],
'Sniper Nav': ['sniper nav/ro'],
'Spy Nav': ['spy nav/ro'],
'Promo nav': ['mncnav', 'pokernightnav', 'l4dnav', 'kfnav'],
'Crush': ['pngcrush']
}
for n in templateMap:
if t.getName().lower() in templateMap[n]:
t.setName(n)
return t
addTemplateFilter(templateRenameMapping)
Manage all {{Item infobox}}es
def infoboxFilter(t, **kwargs):
filteredTemplates = ('weapon infobox', 'hat infobox', 'tool infobox', 'item infobox') # Only do stuff to these templates
if t.getName().lower() not in filteredTemplates:
return t # Skip
t.setName('Item infobox')
t.indentationMatters(True) # Reindents every time, not only when modifying values
paramAliases = { # Parameter alias 'goodParam': 'badParam', or 'goodParam': [list of bad params].
'name': ['weapon-name-override', 'hat-name-override', 'tool-name-override', 'NAME', 'name-override'],
'image': ['weapon-image', 'hat-image', 'tool-image'],
'kill-text-1': 'kill-text',
'team-colors': 'has-team-colors',
'two-models': 'has-two-models',
'slot': ['weapon-slot', 'hat-slot', 'tool-slot'],
'trade': 'tradable',
'gift': 'giftable',
'craft': 'craftable',
'paint': ['paintable', 'Paint'],
'rename': ['name-tag', 'nametag'],
'loadout': 'display-loadout-stats',
'level': 'level-and-type',
'loadout-prefix': 'hide-loadout-prefix'
}
catstoCheck = { # Mapping 'templateAttribute': [List of 'Category|templateAttributeValue']
'type': ['Taunts|taunt', 'Action items|action', 'Taunts|action taunt', 'Hats|hat', 'Miscellaneous items|misc item', 'Tools|tools', 'Weapons|weapon'],
'slot': ['Primary weapons|primary', 'Secondary weapons|secondary', 'Melee weapons|melee', 'PDA1 weapons|pda 1', 'PDA2 weapons|pda 2']
}
catsDontCheck = [u'Category:Beta and unused content', 'Category:Taunts'] # Type and slot won't be modified on these pages
preferedOrder = [ # Prefered order of keys inside template
'name', 'game', 'type', 'beta', 'unused', 'image', 'imagewidth', '3d-team', '3d-alt', '3d-team-alt', 'number-of-3d-images', 'number-of-3d-team-images', 'number-of-3d-alt-images', 'view#', 'view#name', 'team-colors', 'team-colors-class#', 'two-models', 'skin-image-red', 'skin-image-blu', 'TFC-model', 'QTF-model', 'hide-kill-icon', 'kill-icon-#', 'kill-text-#', 'used-by', 'slot', 'crafting-slot', 'custom-slot', 'equip-region', 'equip-region-2', 'contributed-by', 'released', 'availability', 'trade', 'gift', 'craft', 'paint', 'rename', 'numbered', 'medieval', 'ammo-loaded', 'ammo-carried', 'ammo-type', 'show-ammo', 'reload', 'loadout', 'loadout-prefix', 'quality', 'level', '%ATTRIBUTES%', 'item-description', 'item-uses', 'item-flags', 'item-expiration'
]
checkEnglish = ['trade', 'gift', 'craft', 'paint', 'rename', 'numbered', 'medieval'] # If these attributes aren't yes or no, check the values on the english page
attributeTypes = ['neutral', 'positive', 'negative'] # Possible loadout attribute types
regLang = compileRegex('/[^/]+$')
# Step 0 - Check categories:
tCats = None
isTFC = False # Assume false by default
if 'article' in kwargs:
if kwargs['article'] is not None:
cats2 = kwargs['article'].getCategories()
tCats = []
for c in cats2:
tCats.append(u(regLang.sub(u, u(c))))
isTFC = u'Category:Weapons (Classic)' in tCats
# Step 1 - Rename obsolete attributes
for p in paramAliases:
if type(paramAliases[p]) is type([]):
for a in paramAliases[p]:
t.renameParam(a, p)
else:
t.renameParam(paramAliases[p], p)
# Step 2 - Fix ammo stuff, delete pricing attributes
t.delParam('ammo', 'price', 'show-price', 'purchasable', 'backpack-image')
# Step 3 - Fix reload stuff
if t.getParam('reload') is None:
t.renameParam('reload-type', 'reload')
# Step 4 - Count, split, order and fix loadout attributes
attrNumber = 1
regexAttrSplit = compileRegex(r'\s*<br[^<>]*>\s*')
for a in attributeTypes:
if t.getParam(a + '-attributes') is not None:
attrs = regexAttrSplit.split(t.getParam(a + '-attributes'))
for attr in attrs:
t.setParam('att-' + str(attrNumber) + '-' + a, attr)
attrNumber += 1
t.delParam(a + '-attributes')
if tCats is not None:
# Step 5 - Lookup english fallback on certain attributes
fetchEnglish = False
values = {}
for attr in checkEnglish:
if t.getParam(attr) is not None and t.getParam(attr).lower() not in (u'yes', u'no'):
fetchEnglish = True
elif t.getParam(attr) is not None:
values[attr] = t.getParam(attr).lower()
if regLang.search(kwargs['article'].title):
englishArticle = page(regLang.sub(, kwargs['article'].title))
try:
englishContent = englishArticle.getWikiText()
except:
englishContent = u
englishContent, englishTemplates = templateExtract(englishContent)
for englishT in englishTemplates:
if englishT.getName().lower() in filteredTemplates:
for p in paramAliases:
if type(paramAliases[p]) is type([]):
for a in paramAliases[p]:
englishT.renameParam(a, p)
else:
englishT.renameParam(paramAliases[p], p)
for attr in checkEnglish:
if englishT.getParam(attr) is not None and englishT.getParam(attr).lower() in (u'yes', u'no'):
values[attr] = englishT.getParam(attr).lower()
break
for attr in values:
t.setParam(attr, values[attr])
checkCats = True
for c in catsDontCheck:
if c in tCats:
checkCats = False
break
if not isTFC and not t.getParam('custom-slot') and checkCats:
# Step 6 - Set certains attributes based on page categories
for cname in catstoCheck:
cname = u(cname)
found = None
foundmultiple = False
for c in catstoCheck[cname]:
cat, val = u(c).split(u'|')
cat = u'Category:' + u(regLang.sub(u, cat))
if cat in tCats:
if found is not None:
foundmultiple = True
found = val
if not foundmultiple:
pass#t.setParam(cname, found) # Temporarily disabled for Made Man
if u'Category:Weapons' not in tCats:
t.delParam('slot')
# Step 7 - Convert neutral attributes - Disabled
"""desc = []
if t.getParam('item-description') is not None:
desc = regexAttrSplit.split(t.getParam('item-description'))
for i in range(1, 10):
if t.getParam('att-' + str(i) + '-neutral') is not None:
desc.append(t.getParam('att-' + str(i) + '-neutral'))
t.delParam('att-' + str(i) + '-neutral')
desc2 = []
for d in desc:
if d.strip():
desc2.append(d.strip())
if len(desc2):
t.setParam('item-description', u'
'.join(desc2))"""
# Step 7.5 TEMPORARY: Add numbered = no
if not isTFC and not t.getParam('numbered'):
t.setParam('numbered', 'no')
# Step 8 - Do TFC stuff
if isTFC:
t.setParam('type', 'weapon')
t.setParam('game', 'tfc')
# Step 9 - Set correct preferred indentation
for k in ['quality', 'level', 'item-description', 'item-uses', 'item-flags', 'level-and-type', 'loadout-name', 'loadout-prefix', 'hide-loadout-prefix']:
t.setPreferedIndentation(k, 2)
for i in range(1, 10):
for a in attributeTypes:
t.setPreferedIndentation('att-' + str(i) + '-' + a, 2)
# Step 10 - Build correct attribute order
newOrder = []
for o in preferedOrder:
if o.find('#') == -1:
newOrder.append(o)
continue
if o == '%ATTRIBUTES%':
for i in range(1, 10):
for a in attributeTypes:
newOrder.append('att-' + str(i) + '-' + a)
for i in range(1, 10):
newOrder.append(o.replace('#', str(i)))
t.setPreferedOrder(newOrder)
# Step 11 - There is no step 11
return t
addTemplateFilter(infoboxFilter)
Remove useless templates
def removeUselessTemplate(t, **kwargs):
if t.getName().lower() in (u'targeted', u'languages'):
return None # Delete template
return t
addTemplateFilter(removeUselessTemplate)
Remove manual video IDs from {{Weapon Demonstration}}
def weaponDemonstrationFilter(t, **kwargs):
if t.getName().lower() != 'weapon demonstration':
return t # Skip
t.delParam('1')
return t
addTemplateFilter(weaponDemonstrationFilter)
Filter parameters of certain templates
def templateParamFilter(t, **kwargs):
params = { # Map: 'lowercase template name': ['list', 'of', 'params', 'to', 'filter']
'patch layout': ['before', 'after', 'current'],
'item infobox': ['released']
}
if t.getName().lower() not in params:
return t
for p in params[t.getName().lower()]:
if t.getParam(p):
t.setParam(p, fixContent(t.getParam(p), **kwargs))
return t
addTemplateFilter(templateParamFilter)
Remove obsolete parameters
def obsoleteParameterFilter(t, **kwargs):
params = { # Map: 'lowercase template name': ['list', 'of', 'params', 'to', 'delete']
'blueprint': ['ingredient-#n-local', 'result-local', 'result-#n-local'],
'taunt': ['weapon-#n-local'],
}
if t.getName().lower() not in params:
return t
for p in params[t.getName().lower()]:
p = u(p)
if p.find(u'#n') != -1:
for i in range(10):
t.delParam(p.replace(u'#n', str(i)))
else:
t.delParam(p)
return t
addTemplateFilter(obsoleteParameterFilter)
Fix map infobox attribute
def mapTypeFix(t, **kwargs):
mapTypeDict = {
'Arena': ['Badlands (Arena)', 'Granary (Arena)', 'Hardhat', 'Lumberyard', 'Nucleus', 'Offblast', 'Ravine', 'Sawmill', 'Watchtower', 'Well (Arena)'],
'Capture the Flag': ['2Fort', 'Atrophy (Capture the Flag)', 'Bedrock', 'Cloudburst', 'Converge', 'Convoy', 'Deliverance', 'Double Cross', 'Fusion', 'HAARP', 'Landfall', 'Mercy', 'Overlook', 'Premuda', 'Sawmill (Capture the Flag)', 'Slate (Capture the Flag)', 'Stockpile', 'Turbine', 'Vector', 'Well (Capture the Flag)', 'Wildfire', 'Wiretap'],
'Control Point': ['5Gorge', 'Badlands', 'Coldfront', 'DeGroot Keep', 'Dustbowl', 'Egypt', 'Fastlane', 'Freight', 'Furnace Creek', 'Glacier', 'Gorge', 'Granary', 'Gravel Pit', 'Gullywash', 'Junction', 'Mann Manor', 'Mountain Lab', 'Obscure', 'Steel', 'Well', 'Yukon'],
'King of the Hill': ['Harvest', 'Harvest Event', 'Lakeside', 'Moonshine', 'Nucleus (King of the Hill)', 'Sawmill (King of the Hill)', 'Viaduct'],
'Payload': ['Badwater Basin', 'Cashworks', 'Cranetop', 'Frontier', 'Gold Rush', 'Hoodoo', 'Swiftwater', 'Thunder Mountain', 'Upward', 'Waste'],
'Payload Race': ['Animus', 'Cornfield', 'Hightower', 'Highwind', 'Nightfall', 'Panic', 'Pipeline', 'Scoville', 'Solitude'],
'Territorial Control': ['Hydro', 'Meridian'],
'Training mode': ['Dustbowl (Training)', 'Target', 'Walkway']
}
if 'article' not in kwargs:
return t
basetitle = u(kwargs['article'].title.split('/')[0])
if t.getName().lower() != 'map infobox':
return t
for n in mapTypeDict:
if basetitle in mapTypeDict[n]:
t.setParam('game-type', n)
return t
addTemplateFilter(mapTypeFix)
Implement Backpack Item Link template
def backpackLink(t, **kwargs):
if t.getName().lower() != 'item infobox':
return t
if t.getParam('contributed-by') is None:
return t
contributor = t.getParam('contributed-by')
optf2LinkReg = compileRegex(r'^\[http://optf2.com/item/(\S+)\s(.[^\[\]]+)]$')
result = optf2LinkReg.search(contributor)
if result:
t.setParam('contributed-by', '{{Backpack Item Link|' + result.group(1) + '|' + result.group(2) + '}}')
return t
addTemplateFilter(backpackLink)
Implement {{Dictionary}}
class DictionaryUpdater:
def __init__(self):
self.subpageTemplateLang = """{{#switch:{{{lang|{{SUBPAGENAME}}}}}|%options%}}"""
self.subpageTemplateParam = """{{#switch:{{{1|}}}|%options%}}"""
self.invalidParamError = """<span class="error">Error: invalid param.</span>[[Category:ERROR]]"""
self.subpageTemplateID = """%string%"""
self.dictionaries = {
u'Template:Dictionary/items': { # Dictionary page
'name': 'items', # Dictionary name (used for categorizing)
'sync': 'Template:Dictionary/items/Special:SyncData' # Page holding last sync data
},
u'Template:Dictionary/common strings': { # Warning: no underscore
'name': 'common strings',
'sync': 'Template:Dictionary/common strings/Special:SyncData'
},
u'Template:Dictionary/classes': {
'name': 'classes',
'sync': 'Template:Dictionary/classes/Special:SyncData'
},
u'Template:Dictionary/demonstrations': {
'name': 'demonstrations',
'sync': 'Template:Dictionary/demonstrations/Special:SyncData'
},
u'Template:Dictionary/price': {
'name': 'price',
'sync': 'Template:Dictionary/price/Special:SyncData',
'allTemplate': '{{{{{template|item price/fmt}}}|%options%|tt={{{tt|yes}}}}}'
},
u'Template:Dictionary/merchandise': {
'name': 'merchandise',
'sync': 'Template:Dictionary/merchandise/Special:SyncData'
},
u'Template:Dictionary/steam ids': {
'name': 'steam ids',
'sync': 'Template:Dictionary/steam ids/Special:SyncData'
},
u'Template:Dictionary/achievements/scout': {
'name': 'achievements/scout',
'sync': 'Template:Dictionary/achievements/scout/Special:SyncData'
},
u'Template:Dictionary/achievements/soldier': {
'name': 'achievements/soldier',
'sync': 'Template:Dictionary/achievements/soldier/Special:SyncData'
},
u'Template:Dictionary/achievements/pyro': {
'name': 'achievements/pyro',
'sync': 'Template:Dictionary/achievements/pyro/Special:SyncData'
},
u'Template:Dictionary/achievements/demoman': {
'name': 'achievements/demoman',
'sync': 'Template:Dictionary/achievements/demoman/Special:SyncData'
},
u'Template:Dictionary/achievements/heavy': {
'name': 'achievements/heavy',
'sync': 'Template:Dictionary/achievements/heavy/Special:SyncData'
},
u'Template:Dictionary/achievements/engineer': {
'name': 'achievements/engineer',
'sync': 'Template:Dictionary/achievements/engineer/Special:SyncData'
},
u'Template:Dictionary/achievements/medic': {
'name': 'achievements/medic',
'sync': 'Template:Dictionary/achievements/medic/Special:SyncData'
},
u'Template:Dictionary/achievements/sniper': {
'name': 'achievements/sniper',
'sync': 'Template:Dictionary/achievements/sniper/Special:SyncData'
},
u'Template:Dictionary/achievements/spy': {
'name': 'achievements/spy',
'sync': 'Template:Dictionary/achievements/spy/Special:SyncData'
},
u'Template:Dictionary/achievements/general': {
'name': 'achievements/general',
'sync': 'Template:Dictionary/achievements/general/Special:SyncData'
},
u'Template:Dictionary/achievements/halloween': {
'name': 'achievements/halloween',
'sync': 'Template:Dictionary/achievements/halloween/Special:SyncData'
},
u'Template:Dictionary/achievements/treasure hunt': {
'name': 'achievements/treasure hunt',
'sync': 'Template:Dictionary/achievements/treasure hunt/Special:SyncData'
},
u'Template:Dictionary/achievements/replay': {
'name': 'achievements/replay',
'sync': 'Template:Dictionary/achievements/replay/Special:SyncData'
},
u'Template:Dictionary/achievements/summer camp': {
'name': 'achievements/summer camp',
'sync': 'Template:Dictionary/achievements/summer camp/Special:SyncData'
},
u'Template:Dictionary/achievements/foundry': {
'name': 'achievements/foundry',
'sync': 'Template:Dictionary/achievements/foundry/Special:SyncData'
},
u'Template:Dictionary/achievements/christmas': {
'name': 'achievements/christmas',
'sync': 'Template:Dictionary/achievements/christmas/Special:SyncData'
},
u'Template:Dictionary/blueprints': {
'name': 'blueprints',
'sync': 'Template:Dictionary/achievements/blueprints/Special:SyncData'
},
u'Template:Dictionary/templatecore': {
'name': 'templatecore',
'sync': 'Template:Dictionary/templatecore/Special:SyncData'
},
u'Template:Dictionary/defindex': {
'name': 'defindex',
'sync': 'Template:Dictionary/defindex/Special:SyncData'
}
}
self.subpageSeparator = u'/'
# List of supported languages, in prefered order
self.languages = [u'en', u'ar', u'cs', u'da', u'de', u'es', u'fi', u'fr', u'hu', u'it', u'ja', u'ko', u'nl', u'no', u'pl', u'pt', u'pt-br', u'ro', u'ru', u'sv', u'tr', u'zh-hans', u'zh-hant']
self.defaultLang = u'en'
self.allKeyName = u'_all_'
self.filterName = u'Your friendly neighborhood dictionary updater'
self.commentsExtract = compileRegex(r)
self.stringsExtract = compileRegex(r'(?:^[ \t]*#[ \t]*([^\r\n]*?)[ \t]*$\s*)?^[ \t]*([^\r\n]+?[ \t]*(?:\|[ \t]*[^\r\n]+?[ \t]*)*):[ \t]*([^\r\n]+?[ \t]*$|\s*[\r\n]+(?:\s*[ \t]+[-\w]+[ \t]*:[ \t]*[^\r\n]+[ \t]*$)+)', re.IGNORECASE | re.MULTILINE)
self.translationExtract = compileRegex(r'^[ \t]+([-\w]+)[ \t]*:[ \t]*([^\r\n]+)[ \t]*$', re.IGNORECASE | re.MULTILINE)
self.scheduler = BatchScheduler(16)
addWhitelistPage(self.dictionaries.keys())
def generateSubpage(self, keyName, data, currentDict, syncData):
h = hashlib.md5()
if type(data) is type({}): # Subkeys (translations or not)
isTranslation = True
subpage = u(self.subpageTemplateLang)
for k in data:
if k not in self.languages:
isTranslation = False
subpage = u(self.subpageTemplateParam)
break
ordered = []
unordered = {}
if isTranslation:
missing = []
for lang in self.languages:
if lang in data:
ordered.append(lang + u'=' + data[lang])
unordered[lang] = data[lang]
h.update((lang + u'=' + data[lang]).encode('utf8'))
else:
missing.append(lang)
h.update((u'null-' + lang).encode('utf8'))
if self.defaultLang in data:
ordered.insert(0, u'#default=' + data[self.defaultLang])
if len(missing):
subpage = subpage.replace(u'%missing%', u"Languages missing: " + u', '.join(missing))
else:
subpage = subpage.replace(u'%missing%', u"Supported languages: all")
else: # Not a translation
h.update('Any-')
subkeys = data.keys()
subkeys.sort()
for k in subkeys:
ordered.append(k + u'=' + data[k])
unordered[k] = data[k]
h.update((k + u'=' + data[k]).encode('utf8'))
if 'allTemplate' in self.dictionaries[currentDict] and (len(unordered) or len(self.dictionaries[currentDict]['allTemplate']['params'])):
allKey = []
keys = unordered.keys()
keys.sort()
for k in keys:
allKey.append(k + u'=' + unordered[k])
insertIndex = 0
if isTranslation and self.defaultLang in data:
insertIndex = 1
ordered.insert(insertIndex, u(self.allKeyName) + u'=' + u(self.dictionaries[currentDict]['allTemplate'].replace(u'%options%', u'|'.join(allKey))))
subpage = subpage.replace(u'%options%', u'|'.join(ordered))
else: # No subkeys
data = u(data)
subpage = self.subpageTemplateID
h.update(u(u'ID-' + data).encode('utf8'))
subpage = subpage.replace(u'%string%', data)
h = u(h.hexdigest())
if keyName in syncData and syncData[keyName] == h:
return # Same hash
syncData[keyName] = h # Update sync data
subpage = subpage.replace(u'%dictionary%', currentDict)
subpage = subpage.replace(u'%dictionaryname%', self.dictionaries[currentDict]['name'])
subpage = subpage.replace(u'%keyname%', keyName)
self.scheduler.schedule(editPage, currentDict + self.subpageSeparator + keyName, subpage, summary=u'Pushed changes from [[:' + currentDict + u']] for string "' + keyName + u'".', minor=True, nocreate=False)
def processComment(self, commentString, currentDict, definedStrings, syncData):
commentContents = []
for extractedStr in self.stringsExtract.finditer(commentString):
comment = u
if extractedStr.group(1):
comment = u'# ' + u(extractedStr.group(1)) + u'\n'
dataString = u(extractedStr.group(3))
if dataString.find(u'\r') == -1 and dataString.find(u'\n') == -1: # Assume no subkeys
data = dataString.strip()
dataWriteback = u' ' + data
else: # There's subkeys; detect whether this is a translation or not
data = {}
isTranslation = True
for translation in self.translationExtract.finditer(dataString.rstrip()):
data[u(translation.group(1))] = u(translation.group(2))
if u(translation.group(1)) not in self.languages:
isTranslation = False
ordered = []
if isTranslation:
for lang in self.languages:
if lang in data:
ordered.append(u' ' + lang + u': ' + data[lang])
else: # Not a translation, so order in alphabetical order
subkeys = data.keys()
subkeys.sort()
for subk in subkeys:
ordered.append(u' ' + subk + u': ' + data[subk])
dataWriteback = u'\n' + u'\n'.join(ordered)
keyNames = u(extractedStr.group(2)).lower().split(u'|')
validKeyNames = []
for keyName in keyNames:
keyName = keyName.replace(u'_', u' ').strip()
if keyName in definedStrings:
continue # Duplicate key
definedStrings.append(keyName)
validKeyNames.append(keyName)
self.generateSubpage(keyName, data, currentDict, syncData)
if len(validKeyNames):
commentContents.append(comment + u' | '.join(validKeyNames) + u':' + dataWriteback)
self.scheduler.execute()
return u'\n\n'.join(commentContents)
def __call__(self, content, **kwargs):
if 'article' not in kwargs:
return content
if u(kwargs['article'].title) not in self.dictionaries:
return content
currentDict = u(kwargs['article'].title)
syncPage = page(self.dictionaries[currentDict]['sync'])
try:
syncDataText = u(syncPage.getWikiText()).split(u'\n')
except: # Page probably doesn't exist
syncDataText = u
syncData = {}
for sync in syncDataText:
sync = u(sync.strip())
if not sync:
continue
sync = sync.split(u':', 2)
if len(sync) == 2:
syncData[sync[0]] = sync[1]
oldSyncData = syncData.copy()
newContent = u
previousIndex = 0
definedStrings = []
for comment in self.commentsExtract.finditer(content):
newContent += content[previousIndex:comment.start()]
previousIndex = comment.end()
# Process current comment
newContent += u
newContent += content[previousIndex:]
# Check if we need to update sync data
needUpdate = False
for k in syncData:
if k not in oldSyncData or oldSyncData[k] != syncData[k]:
needUpdate = True
break
# Check for deleted strings
for k in oldSyncData:
if k not in definedStrings:
try:
deletePage(currentDict + self.subpageSeparator + k, 'Removed deleted string "' + k + u'" from ' + currentDict + u'.')
except:
pass
if k in syncData:
del syncData[k]
needUpdate = True
if needUpdate:
# Build syncdata string representation
syncKeys = syncData.keys()
syncKeys.sort()
syncLines = []
for k in syncKeys:
syncLines.append(k + u':' + syncData[k])
editPage(syncPage, u'\n'.join(syncLines), summary=u'Updated synchronization information for [[:' + currentDict + u']].', minor=True, nocreate=False)
return newContent
def scheduledRun(self):
for d in self.dictionaries:
fixPage(d)
dictUpdater = DictionaryUpdater()
addFilter(dictUpdater)
scheduleTask(dictUpdater.scheduledRun, 3)
File filters
Crush all PNG/JPG images
class imageCrushFilter:
def __init__(self):
self.minRatio = 10 # Compression ratio threshold
self.minByteDiff = 2048 # Byte difference threshold
self.jpgScanMap = u'0: 0 0 0 0 ;1 2: 0 0 0 0 ;0: 1 8 0 2 ;1: 1 8 0 0 ;2: 1 8 0 0 ;0: 9 63 0 2 ;0: 1 63 2 1 ;0: 1 63 1 0 ;1: 9 63 0 0 ;2: 9 63 0 0 ;'.replace(u';', u';\n')
self.filterName = 'Saved crush information'
self.extractHash = compileRegex(r'\{\{(?:png)?crush\s*\|\s*(\w+?)\s*\|\s*(\w+?)\s*}}')
if programExists('pngout'):
self.pngenabled = 'pngout'
elif programExists('pngcrush'):
self.pngenabled = 'pngcrush'
else:
print 'Warning: PNGOut and PNGCrush are not installed or are not in $PATH'
self.pngenabled = None
self.jpgenabled = programExists('jpegtran')
if not self.jpgenabled:
print 'Warning: jpegtran is not installed or not in $PATH'
def __call__(self, content, article, **kwargs):
title = u(article.title).lower()
if title[-4:] == '.png':
isPNG = True
if self.pngenabled is None:
return content
elif title[-5:] == '.jpeg' or title[-4:] == '.jpg':
isPNG = False
if not self.jpgenabled:
return content
else:
return content
try: # This is a high-risk filter, lots of I/O, so wrap it in a big try
filePage = wikitools.wikifile.File(wiki(), article.title)
hashes = [u, u]
hashResult = self.extractHash.search(content)
hashes = None
hashTemplate = None
if hashResult:
hashes = (u(hashResult.group(1)), u(hashResult.group(2)))
hashTemplate = u'{{crush|' + hashes[0] + u'|' + hashes[1] + u'}}'
tempFile = getTempFilename(extension='png')
filePage.download(location=tempFile, urlQuery=u(getRandBits()))
oldHash = getFileHash(tempFile)
if hashes is not None and oldHash in hashes:
return content # Already worked on that one
hashTemplate = u'{{crush|' + oldHash + u'|None}}'
tempOutput = getTempFilename(extension='png')
if isPNG:
if self.pngenabled == 'pngout':
shutil.copyfile(tempFile, tempOutput)
result = subprocess.call(['pngout', '-b256', '-y', tempOutput])
if result == 0:
possibleBlocks = ['0', '64', '128', '192', '256', '512', '1024', '2048']
currentUpperBlock = 4
currentLowerBlock = 4
tryUpper = True
tryLower = True
while tryUpper or tryLower:
if tryUpper:
currentUpperBlock += 1
upperResult = subprocess.call(['pngout', '-b' + possibleBlocks[currentUpperBlock], '-y', tempOutput])
if upperResult != 0 or currentUpperBlock >= len(possibleBlocks) - 1:
tryUpper = False
if tryLower:
currentLowerBlock -= 1
lowerResult = subprocess.call(['pngout', '-b' + possibleBlocks[currentLowerBlock], '-y', tempOutput])
if lowerResult != 0 or currentLowerBlock <= 0:
tryLower = False
elif self.pngenabled == 'pngcrush':
result = subprocess.call(['pngcrush', '-q', '-l', '9', '-reduce', '-rem', 'gAMA', '-rem', 'cHRM', '-rem', 'iCCP', '-rem', 'sRGB'] + [i for l in [('-m', str(i)) for i in range(138)] for i in l] + [tempFile, tempOutput])
else:
result = subprocess.call(['jpegtran', '-o', '-copy', 'none', '-progressive', '-outfile', tempOutput, tempFile])
oldSize = os.path.getsize(tempFile)
newSize = os.path.getsize(tempOutput)
deleteFile(tempFile)
if not result and oldSize > newSize:
# Ready to upload... or are we?
ratio = int(round(100 * (1.0 - float(newSize) / float(oldSize))))
if ratio >= self.minRatio or oldSize - newSize >= self.minByteDiff:
newHash = getFileHash(tempOutput)
if hashes is not None and newHash in hashes:
deleteFile(tempOutput)
return content # Already got that result, no need to reupload
hashTemplate = u'{{crush|' + oldHash + u'|' + newHash + u'}}'
content2 = wikitools.wikifile.File(wiki(), article.title).getWikiText()
hashResult2 = self.extractHash.search(content2)
hashes2 = None
if hashResult2:
hashes2 = (u(hashResult2.group(1)), u(hashResult2.group(2)))
if hashes == hashes2:
uploadFile(tempOutput, u(article.title), u'Crushed version: ' + u(ratio) + u'% reduction / ' + u(oldSize - newSize) + u' bytes saved; from ' + u(oldSize) + u' to ' + u(newSize) + u' bytes.', overwrite=True, reupload=True)
hashes = (oldHash, newHash)
if hashResult:
content = content[:hashResult.start()] + hashTemplate + content[hashResult.end():]
else:
content = content.strip() + u'\n\n' + hashTemplate
deleteFile(tempOutput)
except:
pass # Well, that didn't work
return content
addFileFilter(imageCrushFilter())
Additional tasks
Update {{Blog link}}
def updateTemplateBlogLink():
fConfig = {
'page': 'Template:Blog link',
'begin': u,
'end': u,
'firstElement': u'[http://www.teamfortress.com/post.php?id={{#switch:{{{1|ERROR}}}',
'lastElement': u'#default = ERROR}} {{{2|{{{1|blog post}}}}}}]',
'rss': 'http://www.teamfortress.com/rss.xml',
'linkRegex': compileRegex(r'id=(\d+)'),
'timeoffset': -7
}
bl = page(fConfig['page'])
content = u(bl.getWikiText())
originalContent = content
begin, end = content.find(fConfig['begin']), content.find(fConfig['end'])
if begin == -1 or end == -1:
return # Couldn't find begin/end
try:
feed = feedparser.parse(fConfig['rss'])
except:
return
results = [fConfig['firstElement']]
entries = 0
for i in feed.entries:
res = fConfig['linkRegex'].search(i.link)
date = datetime.datetime.fromtimestamp(time.mktime(i.updated_parsed)) + datetime.timedelta(hours=fConfig['timeoffset'])
if res:
entries += 1
day = date.strftime('%d')
if day[0] == '0':
day = day[1:]
results.append(u(date.strftime('%B ') + day + date.strftime(', %Y') + ' = ' + res.group(1)))
results.append(fConfig['lastElement'])
results = u'\n | '.join(results)
content = content[:begin+len(fConfig['begin'])] + results + content[end:]
if content != originalContent and entries:
editPage(bl, content, summary=u'Updated blog post list.')
# scheduleTask(updateTemplateBlogLink, 10)
Update {{List of item attributes}}
def updateItemAttributes():
try:
attributes = urllib2.urlopen('http://optf2.com/wiki/attributes').read(-1)
except:
return
if not attributes or attributes.find('<!DOCTYPE') != -1:
return
editPage('Template:List of item attributes', '\n' + attributes, summary=u'Updated item attributes', minor=True)
#scheduleTask(updateItemAttributes, 20)
Update {{Item checklist}} on list of subscribers
def itemChecklists():
def updateItemChecklist(checklist, schema, support):
if not checklist.getParam('steamid'):
checklist.setParam('error', 'Unspecified Steam ID.')
return True
supportedItems = {}
for i in support:
supportedItems[i] = 0
try:
steamUser = steam.user.profile(checklist.getParam('steamid'))
backpack = steam.tf2.backpack(steamUser, schema=schema)
except steam.user.ProfileError as e:
checklist.setParam('error', u(e))
return True
except steam.tf2.TF2Error as e:
if u(e) in (u'Bad SteamID64 given', u'Profile set to private'):
checklist.setParam('error', u(e))
return True
return False
for item in backpack:
itemName = u(item.get_name()).lower()
if itemName in supportedItems:
supportedItems[itemName] += 1
for item in supportedItems:
if supportedItems[item] > 1:
checklist.setParam(item, supportedItems[item])
elif supportedItems[item] == 1:
checklist.setParam(item, 'yes')
else:
p = checklist.getParam(item)
if p is not None:
p = p.lower()
if p in (None, 'no', '0'):
checklist.setParam(item, 'no')
elif p not in ('wanted', 'want', 'do not', 'anti', 'do not want'):
checklist.setParam(item, 'had')
return True
try:
schema = steamGetGameSchema(steam.tf2)
allItems = []
for item in schema:
allItems.append(u(item.get_name()).lower())
except:
return # No schema means no fancy
support = []
templateParams = compileRegex(r'\{\{\{\s*([^{}|]+?)\s*\|')
templateCode = page('Template:Item checklist').getWikiText()
res = templateParams.search(templateCode)
while res:
item = u(res.group(1)).lower()
if item not in support and item in allItems:
support.append(item)
templateCode = templateCode[res.end():]
res = templateParams.search(templateCode)
checkPage, checkLinks = linkExtract(page('User:WindBOT/Item_checklists').getWikiText())
linksLeft = checkLinks[:]
for i in range(12):
randLink = random.choice(linksLeft)
linksLeft.remove(randLink)
checklist = page(randLink.getLink())
print 'Updating', checklist
update = False
oldContent = u(checklist.getWikiText())
content, templatelist = templateExtract(oldContent)
for t in templatelist:
if t.getName().lower().find(u'item checklist') != -1:
update = updateItemChecklist(t, schema, support)
break
content = templateRestore(content, templatelist)
if update and oldContent != content:
editPage(checklist, content, summary=u'Updated Item checklist [[:' + u(checklist.title) + u']]', minor=True)
scheduleTask(itemChecklists, 2)
Update {{Dictionary/defindex}} and {{Dictionary/price}}
def updateGameDictionaries():
if steam is None:
return
# Begin configurable section
game = steam.tf2
priceDictionaryPage = u'Template:Dictionary/price'
indexDictionaryPage = u'Template:Dictionary/defindex'
itemEquivalences = {
u'taunt: the high five!': u'high five!',
u'taunt: the director\'s vision': u'director\'s vision',
u'taunt: the schadenfreude': u'schadenfreude',
u'taunt: the meet the medic': u'meet the medic (taunt)'
}
illegalItemCharacters = compileRegex(r'[:%\r\n\t]')
priceDictionaryHeader = u'{{dictionary/header}}\n== /price ==\n\'\'\'This dictionary is automatically updated by WindBOT\'\'\'. Edits made to this page will be overwritten.\n<!--\n'
indexDictionaryHeader = u'{{dictionary/header}}\n== /defindex ==\n\'\'\'This dictionary is automatically updated by WindBOT\'\'\'. Edits made to this page will be overwritten.\n<!--\n'
dictionaryFooter = u'\n-->'
# End configurable section
schema = steamGetGameSchema(game)
assets = steamGetGameAssets(game)
dictionary = {}
defindex = {}
for item in schema:
try:
itemName = u(item.get_name()).lower()
keys = [itemName]
if itemName in itemEquivalences:
itemName = itemEquivalences[itemName]
keys.append(itemName)
defindex[itemName] = u' | '.join([illegalItemCharacters.sub(u, x) for x in keys]) + u': ' + u(item.get_schema_id())
prices = assets.get_price(item.get_schema_id())
if not len(prices):
continue
dictionary[itemName] = itemName + u':'
for currency in sorted(prices):
price = prices[currency]
if price.is_integer():
price = u(int(price))
else:
price = u(round(price, 2))
dot = price.find(u'.')
if dot == len(price) - 2:
price += u'0'
dictionary[itemName] += u'\n ' + u(currency).lower() + u': ' + price
except:
continue
if len(dictionary):
# Is updating time doktor
finalPage = priceDictionaryHeader
for item in sorted(dictionary):
finalPage += u'\n' + dictionary[item] + u'\n'
finalPage += dictionaryFooter
priceDictionaryPage = page(priceDictionaryPage)
editPage(priceDictionaryPage, finalPage, summary=u'Updated item prices from WebAPI.', minor=True, bot=True, nocreate=True)
# Run other filters on the new page
fixPage(priceDictionaryPage)
if len(defindex):
# Is also updating time doktor
finalPage = indexDictionaryHeader
for item in sorted(defindex):
item_nu = item.replace('_', ' ')
finalPage += u'\n' + defindex[item_nu] + u'\n'
finalPage += dictionaryFooter
indexDictionaryPage = page(indexDictionaryPage)
editPage(indexDictionaryPage, finalPage, summary=u'Updated item indexes from WebAPI.', minor=True, bot=True, nocreate=True)
# Run other filters on the new page
fixPage(indexDictionaryPage)
scheduleTask(updateGameDictionaries, 2)
