Benutzer:Bot-Schafter/Exzellent
aus Wikipedia, der freien Enzyklopädie
English:
This is the modified code of featured.py, part of the pywikipedia framework. Most likely this is not the lastest version - ask here for it if you want it. You will need also two additinal files to run the script: la.py and lang.py. If you want to copy code from here - click on "Seite bearbeiten" (edit) first and copy the source of the page.
Español:
Este esta el archivo de featured.py - los archivos la.py y lang.py también están necesario. Si tu quieres usar lo - tienes que copiar es texto fuente y no lo que ves! Pero mejor si me preguntas por la ultima versión.
"""
To Do:
struktur vereinfachen
geblockte seiten überspringen
anzeige eindeutiger machen - welche seite wird mit welcher vervollständigt
eine datei für alle sprachen - außer die lang.py
"""
#!/usr/bin/python
# -*- coding: utf-8 -*-
__version__ = '$Id: featured.py,v 1.12 2006/03/15 02:57:06 wikipedian Exp $'
from __future__ import generators
import sys, re
import pagegenerators, wikipedia, catlib, config
from la import fixes
from lang import msg, template, featured_name
#++++++++++++++++++++++++++++
class ReplaceRobot:
"""
A bot that can do text replacements.
"""
def __init__(self, generator, replacements, exceptions = [], acceptall = False):
self.generator = generator
self.replacements = replacements
self.exceptions = exceptions
self.acceptall = acceptall
def checkExceptions(self, text):
"""
If one of the exceptions applies for the given text, returns the
substring which matches the exception. Otherwise it returns None.
"""
for exception in self.exceptions:
hit = exception.search(text)
if hit:
return hit.group(0)
return None
def doReplacements(self, text):
"""
Returns the text which is generated by applying all replacements to the
given text.
"""
#new_text = text
for old, new in self.replacements:
text = wikipedia.replaceExceptMathNowikiAndComments(text, old, new)
return text
def run(self):
global text, new_text, atrans, h
"""
Starts the robot.
"""
h = '0'
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
# try:
# # Load the page's text from the wiki
# original_text = page.get() #page.get()
# wikipedia.output(u'%s' % (original_text))
if not page.canBeEdited():
wikipedia.output(u'Skipping locked page %s' % page.title())
h = '1'
# except wikipedia.NoPage:
# wikipedia.output(u'Page %s not found' % page.title())
# continue
# except wikipedia.IsRedirectPage:
# original_text = page.get(get_redirect=True)
match = self.checkExceptions(text)
# skip all pages that contain certain texts
if match:
print 'enthält irgend eine Exception'
wikipedia.output(u'Skipping %s because it contains %s' % (atrans.title(), match))
else:
new_text = self.doReplacements(text)
if new_text == text:
wikipedia.output('No changes were necessary in %s' % atrans.title())
else:
wikipedia.output(u'\n>>> %s <<<' % atrans.title())
wikipedia.showDiff(text, new_text)
# if not self.acceptall:
# choice = wikipedia.inputChoice(u'Do you want to accept these changes?', ['Yes', 'No', 'All'], ['y', 'N', 'a'], 'N')
# if choice in ['a', 'A']:
# self.acceptall = True
# if self.acceptall or choice in ['y', 'Y']:
# wikipedia.output(u'seitentext gespeichert')
text = new_text
print 'erste Korrektur durchgelaufen'
# try:
# page.put(new_text)
# except wikipedia.EditConflict:
# wikipedia.output(u'Skipping %s because of edit conflict' % (page.title()))
#-------------------------
def mainsort():
global text, rtl
findtemplate = wikipedia.translate(wikipedia.getSite(), template)
wikitext = text
m = 1
i = 0
li1 = ['a']
li2 = ['a']
li3 = ['a']
if re.search('Vorlage:%s\' % findtemplate, wikitext):
while li1[i]:
li1[i] = re.search('Vorlage:%s\' % findtemplate, wikitext) # li1[i] <- xx
p = re.search('\[\[%s:(.*?)\]\]' % li1[i].group(1), wikitext) # p <- Article
old = re.compile("Vorlage:%s\" % (findtemplate, li1[i].group(1))) # old <- for removing
wikitext = wikipedia.replaceExceptMathNowikiAndComments(wikitext, old, '<--entfFeat-->') # removing and putting temporary placeholder
while re.search(old, wikitext):
wikitext = wikipedia.replaceExceptMathNowikiAndComments(wikitext, old, '<--entfFeat-->')
if p: # check if there is also the article and not only the link fa for xx
p = p.group(1)
p = wikipedia.replaceExceptMathNowikiAndComments(p, '\(', '\\\(')
p = wikipedia.replaceExceptMathNowikiAndComments(p, '\)', '\\\)')
li2[i]=re.compile("\[\[%s:%s\]\]" % (li1[i].group(1), p)) # w <- xx:Article
li3[i] = re.compile("Vorlage:%s" % (findtemplate, li1[i].group(1))) # old <- for adding
i += 1
li1 = li1 + ['a']
li2 = li2 + ['a']
li3 = li3 + ['a']
li1[i] = re.search('Vorlage:%s\' % findtemplate, wikitext) # to check if another run is necessary
else:
print 'Link FA ohne Link'
i = 0
while li1[i]:
print li2[i].pattern
q = li2[i].search(wikitext) # searching for xx:Article
if q:
if (config.mylang in (rtl)) and (fromsite.lang not in (rtl)):
wikitext=(wikitext[:q.start()] # text untill beginning of xx:Article
+ ("%s" % li3[i].pattern) # adding
+ wikitext[q.start():]) # adding xx:Article and the rest
else:
wikitext=(wikitext[:q.end()] # text untill end of xx:Article
+ ("%s" % li3[i].pattern) # adding
+ wikitext[q.end():]) # adding the rest
i += 1
while m:
h = re.search(r'(?i)([\r\n]{1,2}) *?<--entfFeat--> *?', wikitext)
if h:
wikitext = wikipedia.replaceExceptMathNowikiAndComments(wikitext, r'(?i)([\r\n]{1,2}) *?<--entfFeat--> *?', r"") # removing the placehoder including \n
else:
wikitext = wikipedia.replaceExceptMathNowikiAndComments(wikitext, r'<--entfFeat-->', r"")
m = re.search('<--entfFeat-->', wikitext)
text = wikitext
print 'featured articles have to be moved'
else:
sys.exit(1)
else:
print 'no featured article templates to be moved'
def maincorr():# ------------------template corrections
gen = None
# Array which will collect commandline parameters.
# First element is original text, second element is replacement text.
commandline_replacements = []
# Don't edit pages which contain certain texts.
exceptions = []
# Should the elements of 'replacements' and 'exceptions' be interpreted
# as regular expressions?
regex = False
# Predefined fixes from dictionary 'fixes' (see above).
if config.mylang == 'es':
fix = 'es'
print 'sprache ist es'
elif config.mylang =='sv':
fix = 'SV'
else:
fix = 'FA'
print 'sprache ist FA...'
# pages which will be processed when the -page parameter is used
PageTitles = []
# a page whose links will be processed when the -links parameter is used
linkingPageTitle = None
# will become True when the user presses a ('yes to all') or uses the -always
# commandline paramater.
acceptall = False
# Which namespaces should be processed?
# default to [] which means all namespaces will be processed
namespaces = []
# Which page to start
startpage = None
# Load default summary message.
wikipedia.setAction(wikipedia.translate(wikipedia.getSite(), msg))
# Read commandline parameters.
for arg in wikipedia.handleArgs():
PageTitles.append(arg[6:])
source = 'specificPages'
if (len(commandline_replacements) == 2 and fix == None):
replacements.append((commandline_replacements[0], commandline_replacements[1]))
wikipedia.setAction(wikipedia.translate(wikipedia.getSite(), msg ) % ' (-' + commandline_replacements[0] + ' +' + commandline_replacements[1] + ')')
else:
# Perform one of the predefined actions.
try:
fix = fixes[fix]
except KeyError:
wikipedia.output(u'Available predefined fixes are: %s' % fixes.keys())
wikipedia.stopme()
sys.exit()
if fix.has_key('regex'):
regex = fix['regex']
if fix.has_key('msg'):
wikipedia.setAction(wikipedia.translate(wikipedia.getSite(), fix['msg']))
if fix.has_key('exceptions'):
exceptions = fix['exceptions']
replacements = fix['replacements']
# already compile all regular expressions here to save time later
for i in range(len(replacements)):
old, new = replacements[i]
if not regex:
old = re.escape(old)
oldR = re.compile(old, re.UNICODE)
replacements[i] = oldR, new
for i in range(len(exceptions)):
exception = exceptions[i]
if not regex:
exception = re.escape(exception)
exceptionR = re.compile(exception, re.UNICODE)
exceptions[i] = exceptionR
if PageTitles:
pages = [wikipedia.Page(wikipedia.getSite(), PageTitle) for PageTitle in PageTitles]
gen = iter(pages)
if namespaces != []:
gen = pagegenerators.NamespaceFilterPageGenerator(gen, namespaces)
preloadingGen = pagegenerators.PreloadingGenerator(gen, pageNumber = 50)
bot = ReplaceRobot(preloadingGen, replacements, exceptions, acceptall)
bot.run()
#+++++++++++++++++++++++
interactive=0
nocache=0
afterpage=u"!"
try:
import pickle
cache=pickle.load(file("featured/cache","rb"))
except:
cache={}
def featuredArticles(site):
method=featured_name[site.lang][0]
name=featured_name[site.lang][1]
args=featured_name[site.lang][2:]
raw=method(site, name, *args)
arts=[]
for p in raw:
if p.namespace()==0: # Article
arts.append(p)
elif p.namespace()==1: # Article talk (like in English)
arts.append(wikipedia.Page(p.site(), p.titleWithoutNamespace()))
return arts
def findTranslated(page, oursite=None):
if not oursite:
oursite=wikipedia.getSite()
if page.isRedirectPage():
page=wikipedia.Page(page.site(), page.getRedirectTarget())
try:
iw=page.interwiki()
except:
wikipedia.output(u"no interwiki, giving up")
return None
ourpage=None
for p in iw:
if p.site()==oursite:
ourpage=p
break
if not ourpage:
wikipedia.output(u"No corresponding page in "+`oursite`)# % (fromsite.lang)) , a.title()
return None
if not ourpage.exists():
wikipedia.output(u"Our page doesn't exist: "+ourpage.title())
return None
if ourpage.isRedirectPage():
ourpage=wikipedia.Page(ourpage.site(),ourpage.getRedirectTarget())
wikipedia.output(u"Corresponding page is "+ourpage.title())
if ourpage.namespace() != 0:
wikipedia.output(u"...not in the main namespace, skipping")
return None
if ourpage.isRedirectPage():
wikipedia.output(u"double redirect, skipping")
return None
if not ourpage.exists():
wikipedia.output(u"page doesn't exist, skipping")
return None
try:
iw=ourpage.interwiki()
except:
return None
backpage=None
for p in iw:
if p.site()==page.site():
backpage=p
break
if not backpage:
wikipedia.output(u"no back interwiki ref")
return None
if backpage==page:
# everything is ok
return ourpage
if backpage.isRedirectPage():
backpage=wikipedia.Page(backpage.site(),backpage.getRedirectTarget())
if backpage==page:
# everything is ok
return ourpage
wikipedia.output(u"back interwiki ref target is "+backpage.title())
return None
def featuredWithInterwiki(fromsite, tosite):
global text, new_text, atrans, h, rtl
rtl = '\'he\', \'ar\', \'ur\', \'fa\', \'yi\', \'arc\', \'dv\', \'ks\', \'mzn\', \'ps\', \'sd\
if not fromsite.lang in cache:
cache[fromsite.lang]={}
if not tosite.lang in cache[fromsite.lang]:
cache[fromsite.lang][tosite.lang]={}
cc=cache[fromsite.lang][tosite.lang]
if nocache:
cc={}
findtemplate = wikipedia.translate(wikipedia.getSite(), template)
re_Link_FA=re.compile(ur"\{\{%s\|%s\}\}" % (findtemplate, fromsite.lang))
re_this_iw=re.compile(ur"\[\[%s:[^]]+\]\]" % fromsite.lang)
arts=featuredArticles(fromsite)
pairs=[]
for a in arts:
if a.title()<afterpage:
continue
if u"/" in a.title():
wikipedia.output(u"%s is a subpage" % a.title())
continue
if a.title() in cc:
wikipedia.output(u"(cached) %s -> %s"%(a.title(), cc[a.title()]))
continue
if a.isRedirectPage():
a=wikipedia.Page(a.site(),a.getRedirectTarget())
try:
if not a.exists():
wikipedia.output(u"source page doesn't exist: %s" % a.title())
continue
atrans=findTranslated(a,tosite)
if atrans:
text=atrans.get() # ganzen Artikel in text speichern
print 'jetzt müsste er austauschen'
maincorr()
m=re_Link_FA.search(text) # link-FA im text suchen und existens in m speichern
if m: # wert von m (bei FA|xx existiert, dann m==1) prüfen
wikipedia.output(u"(already done)")
else:
# insert just before interwiki ++++++++++++ (schreiben des eintrages) ++++++++++++++++++++++++++++++++++
# print text
mainsort() # --------------------einträge sortieren
if (not interactive or
wikipedia.input(u'Connecting %s -> %s. Proceed? [Y/N]'%(a.title(), atrans.title())) in ['Y','y']
):
m=re_this_iw.search(text) # raussuchen des xx:artikel
if not m:
wikipedia.output(u"no interwiki record, very strange")
continue
if (config.mylang in (rtl)) and (fromsite.lang not in (rtl)):
text=(text[:m.start()] #alles was bis zum interwikilink steht
+ (u"Vorlage:%s" % (findtemplate, fromsite.lang)) # FA einfügen
+ text[m.start():]) #alles was noch danach kommt
else:
text=(text[:m.end()] #alles was bis zum interwikilink steht
+ (u"Vorlage:%s" % (findtemplate, fromsite.lang)) # FA einfügen
+ text[m.end():]) #alles was noch danach kommt
comment = wikipedia.setAction(wikipedia.translate(wikipedia.getSite(), msg) % (fromsite.lang, a.title())) #kommentar für die history
print text
if h == 1:
wikipedia.output(u'Skipping locked page (end)')
if (not interactive or
wikipedia.input(u'Connecting %s -> %s. Proceed? [Y/N]'%(a.title(), atrans.title())) in ['Y','y']
):
atrans.put(text, comment) #abspeichern
else:
print 'seite nicht gesperrt, und trotzdem hat er irgendwas'
cc[a.title()]=atrans.title()
except wikipedia.PageNotSaved, e:
wikipedia.output(u"Page not saved")
if __name__=="__main__":# keine nutzung auf nl
if config.usernames.has_key('wikipedia') and config.usernames['wikipedia'].has_key('nl'):
print "Bot is not to be used at NL Wikipedias."
sys.exit()
# argumente der command line auswerten und in variablen speichern
fromlang=[]
for arg in sys.argv[1:]:
arg = wikipedia.argHandler(arg, 'featured')
if not arg:
pass
elif arg == '-interactive':
interactive=1
elif arg == '-nocache':
nocache=1
elif arg.startswith('-fromlang:'):
fromlang=arg[10:].split(",")
if len(fromlang)==1 and fromlang[0].index("-")>=0:
ll1,ll2=fromlang[0].split("-",1)
if not ll1: ll1=""
if not ll2: ll2="zzzzzzz"
fromlang=[ll for ll in featured_name.keys() if ll>=ll1 and ll<=ll2]
elif arg == '-fromall':
fromlang=featured_name.keys()
elif arg.startswith('-after:'):
afterpage=arg[7:]
# falls keine argument
if not fromlang:
print """usage:
featured [-interactive] [-nocache] [-fromlang:xx,yy|-fromall] [-after:]"""
sys.exit(1)
# wenn alle "-fromall" dann sortiert er hier
fromlang.sort()
try:
for ll in fromlang:
fromsite=wikipedia.Site(ll)
if not fromsite==wikipedia.getSite():
featuredWithInterwiki(fromsite, wikipedia.getSite())
# ende und cache speichern
finally:
wikipedia.stopme()
if not nocache:
import pickle
pickle.dump(cache,file("featured/cache","wb"))
print 'juhuuu - fertig'
