MalafayaBot

#!/usr/bin/env python

# -*- coding: UTF-8 -*-

import sys

import re

import codecs

import query

 

site = None

# Log bot warnings

def log(message):

if page.exists():

text = page.get()

text += u"\r\n* ~~~~~: " + message

page.put(text, "Mensagem de log do bot")

# Gets the base language category for a language code

def getLangCat(langcode):

params = {

}

datas = query.GetData(params, site)

mo = reparsedtext.match(data)

return mo.group(1).strip()

def createFLentry(transl, langcode, pos, title, gloss):

# Build a page

if page.exists():

if page.isRedirectPage():

(langcode, transl))

return True # meaning there is a page there now

if not page.isEmpty():

return True # meaning there is a page there now

# check language section later ...

# Decap gloss (some people insist on capitalizing it, which is wrong) this is almost always right:

gl = gloss.lower()

return False

if gl[1:] != gloss[1:]: gl = gloss # caps in string after first, so probably okay

langcat = getLangCat(langcode)

# Check if base language category exists (we don't want to add words for languages that are not yet cataloged or whose name is not well specified)

if not langCatPage.exists():

log(u"Categoria base para língua '''%s''' com nome 'Categoria:%s' não existe. Entrada '%s' não adicionada" %

(langcode, langcat, transl))

return False

text = u"""={{-%s-}}=

==%s==

else:

text = text % (langcode, pos, transl, title, u' (' + gl + u')', title, langcode, pos, langcat)

try:

page.put(text, comment = u"Criada automaticamente a partir das traduções em [[%s]]" % title, minorEdit = False)

print "Failed to save page"

return False

print "Socket error, maybe not saving page"

return False

if not newCat.exists() or (not newCat.isRedirectPage() and newCat.isEmpty()):

# Create the auto pages category for this language

newCat.put(u"{{catpagautolíngua|%s}}" % langcode, u"Criada automaticamente")

return True

# Converts the Part of Speech to the AO1990

def convertOrtography(pos):

return 'Adjetivo'

return pos

def main():

socket.setdefaulttimeout(30)

pageToProcess = None

for arg in sys.argv[1:]:

if arg.startswith('-'):

print 'Arguments not supported yet'

else: pageToProcess = unicode(arg, 'latin1')

# make sure we are logged in

site.forceLogin()

config.put_throttle = 1

rehead = re.compile(r'={1,4}(.+?)={1,4}')

rehead2 = re.compile(r'={2}(.+?)={2}')

reglosstune = re.compile(r'(.*?)\((.*?)\)')

reglosstune2 = re.compile(r'(.*?):\s?(.*)')

partsOfSpeech = set(['Substantivo', 'Adjetivo', 'Verbo', 'Pronome', 'Locução substantiva', 'Numeral'])

stops = set([])

if (pageToProcess == None):

entry = site.randompage()

pageToProcess = entry.title()

else:

if entry.namespace() != 0:

return

text = entry.get()

if not u'=Português=' in text and not u'={{pt}}=' in text and not u'={{-pt-}}=' in text and not u'= Português =' in text and not u'= {{pt}} =' in text and not u'= {{-pt-}} =' in text:

return

""" if not u'==Tradução==' in text and not u'==Traduções==' in text and not u'=={{tradu}}==' in text:

return"""

lines = text.splitlines()

intrans = False

if mo:

header = mo.group(1).strip()

if header == u"Tradução" or header == u'{{tradução}}' or header == u'Traduções':

if not (rehead3.match(lines[i])):

return

intrans = True

gloss = ''

else: intrans = False

if header in partsOfSpeech and rehead2.match(lines[i]):

pos = convertOrtography(header)

if header in stops and rehead2.match(lines[i]): pos = ''

continue

if not intrans: continue

mo = regloss.match(lines[i])

if mo:

if mo:

gloss = mo.group(2).strip()

continue

# Try all the possible translation variants

transls = []

lang = mo.group(1).strip()

transls += mo.group(2).strip().split('|')

mo = retrans2.match(lines[i])

if mo:

lang = mo.group(2).strip()

transls += mo.group(3).strip().split('|')

mo = retrans3.match(lines[i])

if mo:

lang = mo.group(1).strip()

transls += mo.group(2).strip().split('|')

mo = retrans4.match(lines[i])

if mo:

lang = mo.group(2).strip()

transls += mo.group(3).strip().split('|')

mo = retrans5.match(lines[i])

if mo:

lang = mo.group(2).strip()

transls += mo.group(3).strip().split('|')

if len(transls) == 0: continue

if lang in ignoreLangs:

continue

for transl in transls:

if len(transl) > 0:

createFLentry(transl, lang, pos, entry.title(), gloss)

if __name__ == "__main__":

try:

main()

finally:

</source>