def weblinksIn(text, withoutBracketed = False, onlyBracketed = False):
text = pywikibot.removeDisabledParts(text)
# MediaWiki parses templates before parsing external links. Thus, there
# might be a | or a } directly after a URL which does not belong to
# the URL itself.
# First, remove the curly braces of inner templates:
nestedTemplateR = re.compile(r'{{([^}]*?){{(.*?)}}(.*?)}}')
while nestedTemplateR.search(text):
text = nestedTemplateR.sub(r'{{\1 \2 \3}}', text)
# Then blow up the templates with spaces so that the | and }} will not be regarded as part of the link:.
templateWithParamsR = re.compile(r'{{([^}]*?[^ ])\|([^ ][^}]*?)}}',
re.DOTALL)
while templateWithParamsR.search(text):
text = templateWithParamsR.sub(r'{{ \1 | \2 }}', text)
linkR = pywikibot.compileLinkR(withoutBracketed, onlyBracketed)
# Remove HTML comments in URLs as well as URLs in HTML comments.
# Also remove text inside nowiki links etc.
text = pywikibot.removeDisabledParts(text)
for m in linkR.finditer(text):
yield m.group('url')
def weblinksIn(text, withoutBracketed = False, onlyBracketed = False):
text = pywikibot.removeDisabledParts(text)
# MediaWiki parses templates before parsing external links. Thus, there
# might be a | or a } directly after a URL which does not belong to
# the URL itself.
# First, remove the curly braces of inner templates:
nestedTemplateR = re.compile(r'{{([^}]*?){{(.*?)}}(.*?)}}')
while nestedTemplateR.search(text):
text = nestedTemplateR.sub(r'{{\1 \2 \3}}', text)
# Then blow up the templates with spaces so that the | and }} will not be regarded as part of the link:.
templateWithParamsR = re.compile(r'{{([^}]*?[^ ])\|([^ ][^}]*?)}}',
re.DOTALL)
while templateWithParamsR.search(text):
text = templateWithParamsR.sub(r'{{ \1 | \2 }}', text)
linkR = pywikibot.compileLinkR(withoutBracketed, onlyBracketed)
# Remove HTML comments in URLs as well as URLs in HTML comments.
# Also remove text inside nowiki links etc.
text = pywikibot.removeDisabledParts(text)
for m in linkR.finditer(text):
yield m.group('url')
def __iter__(self):
import xmlreader
dump = xmlreader.XmlDump(self.xmlFilename)
for entry in dump.parse():
text = pywikibot.removeDisabledParts(entry.text)
if self.refR.search(text) and not self.referencesR.search(text):
yield pywikibot.Page(pywikibot.getSite(), entry.title)
def getLinks(wtext):
#adapted from linkedPages() http://svn.wikimedia.org/svnroot/pywikipedia/trunk/pywikipedia/wikipedia.py
links = []
wtext = wikipedia.removeLanguageLinks(wtext, site)
wtext = wikipedia.removeCategoryLinks(wtext, site)
# remove HTML comments, pre, nowiki, and includeonly sections
# from text before processing
wtext = wikipedia.removeDisabledParts(wtext)
# resolve {{ns:-1}} or {{ns:Help}}
wtext = site.resolvemagicwords(wtext)
for match in Rlink.finditer(wtext):
title = match.group('title')
title = title.replace("_", " ").strip(" ")
if title.startswith("#"): # this is an internal section link
continue
if not site.isInterwikiLink(title):
if title.startswith("#"): # [[#intrasection]] same article
continue
title = title.split('#')[
0] # removing sections [[other article#section|blabla]]
title = '%s%s' % (title[:1].upper(), title[1:]) #first up
title = title.strip()
if title.startswith(":") or title.startswith(
"File:") or title.startswith("Image:") or title.startswith(
"Category:"): # files, cats, etc
continue
if title and title not in links:
links.append(title)
return links
def lacksReferences(self, text, verbose = True):
"""
Checks whether or not the page is lacking a references tag.
"""
oldTextCleaned = pywikibot.removeDisabledParts(text)
if self.referencesR.search(oldTextCleaned) or \
self.referencesTagR.search(oldTextCleaned):
if verbose:
pywikibot.output(u'No changes necessary: references tag found.')
return False
elif self.referencesTemplates:
templateR = u'{{(' + u'|'.join(self.referencesTemplates) + ')'
if re.search(templateR, oldTextCleaned, re.IGNORECASE|re.UNICODE):
if verbose:
pywikibot.output(
u'No changes necessary: references template found.')
return False
if not self.refR.search(oldTextCleaned):
if verbose:
pywikibot.output(u'No changes necessary: no ref tags found.')
return False
else:
if verbose:
pywikibot.output(u'Found ref without references.')
return True
def lacksReferences(self, text, verbose = True):
"""
Checks whether or not the page is lacking a references tag.
"""
oldTextCleaned = pywikibot.removeDisabledParts(text)
if self.referencesR.search(oldTextCleaned) or \
self.referencesTagR.search(oldTextCleaned):
if verbose:
pywikibot.output(u'No changes necessary: references tag found.')
return False
elif self.referencesTemplates:
templateR = u'{{(' + u'|'.join(self.referencesTemplates) + ')'
if re.search(templateR, oldTextCleaned, re.IGNORECASE|re.UNICODE):
if verbose:
pywikibot.output(
u'No changes necessary: references template found.')
return False
if not self.refR.search(oldTextCleaned):
if verbose:
pywikibot.output(u'No changes necessary: no ref tags found.')
return False
else:
if verbose:
pywikibot.output(u'Found ref without references.')
return True
def __iter__(self):
import xmlreader
dump = xmlreader.XmlDump(self.xmlFilename)
for entry in dump.parse():
text = pywikibot.removeDisabledParts(entry.text)
if self.refR.search(text) and not self.referencesR.search(text):
yield pywikibot.Page(pywikibot.getSite(), entry.title)
def procesPage(self, page):
"""
Proces a single page
"""
item = pywikibot.DataPage(page)
pywikibot.output('Processing %s' % page)
if not item.exists():
pywikibot.output('%s doesn\'t have a wikidata item :(' % page)
#TODO FIXME: We should provide an option to create the page
else:
pagetext = page.get()
pagetext = pywikibot.removeDisabledParts(pagetext)
templates = pywikibot.extract_templates_and_params(pagetext)
for (template, fielddict) in templates:
# We found the template we were looking for
if template.replace(u'_', u' ') == self.templateTitle:
for field, value in fielddict.items():
# This field contains something useful for us
if field in self.fields:
# Check if the property isn't already set
claim = self.fields[field]
if claim in item.get().get('claims'):
pywikibot.output(
u'A claim for %s already exists. Skipping'
% (claim,))
# TODO FIXME: This is a very crude way of dupe
# checking
else:
# Try to extract a valid page
match = re.search(re.compile(
r'\[\[(?P<title>[^\]|[#<>{}]*)(\|.*?)?\]\]'),
value)
if match:
try:
link = match.group(1)
linkedPage = pywikibot.Page(self.site,
link)
if linkedPage.isRedirectPage():
linkedPage = linkedPage.getRedirectTarget()
linkedItem = pywikibot.DataPage(linkedPage)
pywikibot.output('Adding %s --> %s'
% (claim,
linkedItem.getID()))
if self.setSource(self.site().language()):
item.editclaim(
str(claim),
linkedItem.getID(),
refs={self.setSource(
self.site().language())})
else:
item.editclaim(str(claim),
linkedItem.getID())
except pywikibot.NoPage:
pywikibot.output(
"[[%s]] doesn't exist so I can't link to it"
% linkedItem.title())
def procesPage(self, page):
"""
Proces a single page
"""
item = pywikibot.DataPage(page)
pywikibot.output('Processing %s' % page)
if not item.exists():
pywikibot.output('%s doesn\'t have a wikidata item :(' % page)
#TODO FIXME: We should provide an option to create the page
else:
pagetext = page.get()
pagetext = pywikibot.removeDisabledParts(pagetext)
templates = pywikibot.extract_templates_and_params(pagetext)
for (template, fielddict) in templates:
# We found the template we were looking for
if template.replace(u'_', u' ') == self.templateTitle:
for field, value in fielddict.items():
# This field contains something useful for us
if field in self.fields:
# Check if the property isn't already set
claim = self.fields[field]
if claim in item.get().get('claims'):
pywikibot.output(
u'A claim for %s already exists. Skipping'
% (claim, ))
# TODO FIXME: This is a very crude way of dupe
# checking
else:
# Try to extract a valid page
match = re.search(
re.compile(
r'\[\[(?P<title>[^\]|[#<>{}]*)(\|.*?)?\]\]'
), value)
if match:
try:
link = match.group(1)
linkedPage = pywikibot.Page(
self.site, link)
if linkedPage.isRedirectPage():
linkedPage = linkedPage.getRedirectTarget(
)
linkedItem = pywikibot.DataPage(
linkedPage)
pywikibot.output(
'Adding %s --> %s' %
(claim, linkedItem.getID()))
if self.setSource(
self.site().language()):
item.editclaim(
str(claim),
linkedItem.getID(),
refs={
self.setSource(
self.site().language())
})
else:
item.editclaim(
str(claim), linkedItem.getID())
except pywikibot.NoPage:
pywikibot.output(
"[[%s]] doesn't exist so I can't link to it"
% linkedItem.title())
def standardizePageFooter(self, text):
"""
Makes sure that interwiki links, categories and star templates are
put to the correct position and into the right order.
This combines the old instances standardizeInterwiki and standardizeCategories
The page footer has the following section in that sequence:
1. categories
2. additional information depending on local site policy
3. stars templates for featured and good articles
4. interwiki links
"""
starsList = [
u'bueno',
u'cyswllt[ _]erthygl[ _]ddethol',
u'dolen[ _]ed',
u'destacado',
u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a',
u'link[ _]adq',
u'link[ _]f[lm]',
u'link[ _]km',
u'link[ _]sm',
u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
categories = None
interwikiLinks = None
allstars = []
hasCommentLine = False
# The PyWikipediaBot is no longer allowed to touch categories on the German Wikipedia.
# See http://de.wikipedia.org/wiki/Hilfe_Diskussion:Personendaten/Archiv/bis_2006#Position_der_Personendaten_am_.22Artikelende.22
# ignoring nn-wiki of cause of the comment line above iw section
if not self.template and not '{{Personendaten' in text:
categories = pywikibot.getCategoryLinks(text, site=self.site)
if not self.talkpage: # and pywikibot.calledModuleName() <> 'interwiki':
subpage = False
if self.template:
loc = None
try:
tmpl, loc = moved_links[self.site.lang]
del tmpl
except KeyError:
pass
if loc != None and loc in self.title:
subpage = True
interwikiLinks = pywikibot.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = pywikibot.removeLanguageLinks(text, site=self.site)
# Removing the stars' issue
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile(
'(\{\{(?:template:|)%s\|.*?\}\}[\s]*)' % star, re.I)
found = regex.findall(starstext)
if found != []:
if pywikibot.verbose:
print found
text = regex.sub('', text)
allstars += found
# nn got a message between the categories and the iw's
# and they want to keep it there, first remove it
if self.site.language() == 'nn':
regex = re.compile(
'(<!-- ?interwiki \(no(?:/nb)?, ?sv, ?da first; then other languages alphabetically by name\) ?-->)'
)
found = regex.findall(text)
if found:
if pywikibot.verbose:
print found
hasCommentLine = True
text = regex.sub('', text)
# Adding categories
if categories:
text = pywikibot.replaceCategoryLinks(text,
categories,
site=self.site)
# Put the nn iw message back
if self.site.language() == 'nn' and not self.talkpage and (
interwikiLinks or hasCommentLine):
text = text + '\r\n\r\n' + nn_iw_msg
# Adding stars templates
if allstars:
text = text.strip() + self.site.family.interwiki_text_separator
allstars.sort()
for element in allstars:
text += '%s\r\n' % element.strip()
if pywikibot.verbose:
pywikibot.output(u'%s' % element.strip())
# Adding the interwiki
if interwikiLinks:
text = pywikibot.replaceLanguageLinks(text,
interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def run(self):
"""
Runs the Bot
"""
pywikibot.setAction(pywikibot.translate(self.site, msg))
try:
deadLinks = codecs.open(listof404pages, 'r', 'latin_1').read()
except IOError:
pywikibot.output(
'You need to download http://www.twoevils.org/files/wikipedia/404-links.txt.gz and to ungzip it in the same directory')
raise
socket.setdefaulttimeout(30)
editedpages = 0
for page in self.generator:
try:
# Load the page's text from the wiki
new_text = page.get()
if not page.canBeEdited():
pywikibot.output(u"You can't edit page %s"
% page.title(asLink=True))
continue
except pywikibot.NoPage:
pywikibot.output(u'Page %s not found' % page.title(asLink=True))
continue
except pywikibot.IsRedirectPage:
pywikibot.output(u'Page %s is a redirect'
% page.title(asLink=True))
continue
for match in linksInRef.finditer(pywikibot.removeDisabledParts(page.get())):
#for each link to change
link = match.group(u'url')
#debugging purpose
#print link
if u'jstor.org' in link:
#TODO: Clean URL blacklist
continue
ref = RefLink(link, match.group('name'))
f = None
try:
socket.setdefaulttimeout(20)
try:
f = urllib2.urlopen(ref.url.decode("utf8"))
except UnicodeError:
ref.url = urllib2.quote(ref.url.encode("utf8"),"://")
f = urllib2.urlopen(ref.url)
#Try to get Content-Type from server
headers = f.info()
contentType = headers.getheader('Content-Type')
if contentType and not self.MIME.search(contentType):
if ref.link.lower().endswith('.pdf') and \
not self.ignorepdf:
# If file has a PDF suffix
self.getPDFTitle(ref, f)
else:
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : media : %s '
% ref.link)
if ref.title:
if not re.match(
'(?i) *microsoft (word|excel|visio)',
ref.title):
ref.transform(ispdf=True)
repl = ref.refTitle()
else:
pywikibot.output(
'\03{lightyellow}WARNING\03{default} : PDF title blacklisted : %s '
% ref.title)
repl = ref.refLink()
else:
repl = ref.refLink()
new_text = new_text.replace(match.group(), repl)
continue
# Get the real url where we end (http redirects !)
redir = f.geturl()
if redir != ref.link and \
domain.findall(redir) == domain.findall(link):
if soft404.search(redir) and \
not soft404.search(ref.link):
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : Redirect 404 : %s '
% ref.link)
continue
if dirIndex.match(redir) and \
not dirIndex.match(ref.link):
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : Redirect to root : %s '
% ref.link)
continue
# uncompress if necessary
if headers.get('Content-Encoding') in ('gzip', 'x-gzip'):
# XXX: small issue here: the whole page is downloaded
# through f.read(). It might fetch big files/pages.
# However, truncating an encoded gzipped stream is not
# an option, for unzipping will fail.
compressed = StringIO.StringIO(f.read())
f = gzip.GzipFile(fileobj=compressed)
# Read the first 1,000,000 bytes (0.95 MB)
linkedpagetext = f.read(1000000)
socket.setdefaulttimeout(None)
except UnicodeError:
#example : http://www.adminet.com/jo/20010615¦/ECOC0100037D.html
# in [[fr:Cyanure]]
pywikibot.output(
u'\03{lightred}Bad link\03{default} : %s in %s'
% (ref.url, page.title(asLink=True)))
continue
except urllib2.HTTPError, e:
pywikibot.output(u'HTTP error (%s) for %s on %s'
% (e.code, ref.url,
page.title(asLink=True)),
toStdout = True)
# 410 Gone, indicates that the resource has been purposely
# removed
if e.code == 410 or \
(e.code == 404 and (u'\t%s\t' % ref.url in deadLinks)):
repl = ref.refDead()
new_text = new_text.replace(match.group(), repl)
continue
except (urllib2.URLError,
socket.error,
IOError,
httplib.error), e:
#except (urllib2.URLError, socket.timeout, ftplib.error, httplib.error, socket.error), e:
pywikibot.output(u'Can\'t retrieve page %s : %s'
% (ref.url, e))
continue
except ValueError:
#Known bug of httplib, google for :
#"httplib raises ValueError reading chunked content"
continue
def add_text(page=None, addText=None, summary=None, regexSkip=None,
regexSkipUrl=None, always=False, up=False, putText=True,
oldTextGiven=None, reorderEnabled=True, create=False):
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
starsList = [
u'bueno',
u'bom interwiki',
u'cyswllt[ _]erthygl[ _]ddethol', u'dolen[ _]ed',
u'destacado', u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a', u'link[ _]adq', u'link[ _]f[lm]', u'link[ _]km',
u'link[ _]sm', u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
errorCount = 0
site = pywikibot.getSite()
pathWiki = site.family.nicepath(site.lang)
site = pywikibot.getSite()
if oldTextGiven is None:
try:
text = page.get()
except pywikibot.NoPage:
if create:
pywikibot.output(u"%s doesn't exist, creating it!"
% page.title())
text = u''
else:
pywikibot.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always)
except pywikibot.IsRedirectPage:
pywikibot.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always)
else:
text = oldTextGiven
# If not up, text put below
if not up:
newtext = text
# Translating the \\n into binary \n
addText = addText.replace('\\n', '\n')
if (reorderEnabled):
# Getting the categories
categoriesInside = pywikibot.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = pywikibot.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = pywikibot.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = pywikibot.removeLanguageLinks(newtext, site)
# Adding the text
newtext += u"\n%s" % addText
# Reputting the categories
newtext = pywikibot.replaceCategoryLinks(newtext,
categoriesInside, site,
True)
# Dealing the stars' issue
allstars = []
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub('', newtext)
allstars += found
if allstars != []:
newtext = newtext.strip() + '\r\n\r\n'
allstars.sort()
for element in allstars:
newtext += '%s\r\n' % element.strip()
# Adding the interwiki
newtext = pywikibot.replaceLanguageLinks(newtext, interwikiInside,
site)
else:
newtext += u"\n%s" % addText
else:
newtext = addText + '\n' + text
if putText and text != newtext:
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<"
% page.title())
#pywikibot.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if always or choice == 'y':
try:
pass
if always:
page.put(newtext, summary,
minorEdit=False)
else:
page.put_async(newtext, summary,
minorEdit=False)
except pywikibot.EditConflict:
pywikibot.output(u'Edit conflict! skip!')
return (False, False, always)
except pywikibot.ServerError:
errorCount += 1
if errorCount < 5:
pywikibot.output(u'Server Error! Wait..')
time.sleep(5)
continue
else:
raise pywikibot.ServerError(u'Fifth Server Error!')
except pywikibot.SpamfilterError, e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s'
% (page.title(), e.url))
return (False, False, always)
except pywikibot.PageNotSaved, error:
pywikibot.output(u'Error putting page: %s' % error.args)
return (False, False, always)
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
def add_text(
page=None,
addText=None,
summary=None,
regexSkip=None,
regexSkipUrl=None,
always=False,
up=False,
putText=True,
oldTextGiven=None,
create=False,
):
if not addText:
raise NoEnoughData("You have to specify what text you want to add!")
if not summary:
summary = wikipedia.translate(wikipedia.getSite(), msg) % addText[:200]
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
starsList = [
u"bueno",
u"cyswllt[ _]erthygl[ _]ddethol",
u"dolen[ _]ed",
u"destacado",
u"destaca[tu]",
u"enllaç[ _]ad",
u"enllaz[ _]ad",
u"leam[ _]vdc",
u"legătură[ _]a[bcf]",
u"liamm[ _]pub",
u"lien[ _]adq",
u"lien[ _]ba",
u"liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt",
u"liên[ _]kết[ _]chọn[ _]lọc",
u"ligam[ _]adq",
u"ligoelstara",
u"ligoleginda",
u"link[ _][afgu]a",
u"link[ _]adq",
u"link[ _]f[lm]",
u"link[ _]km",
u"link[ _]sm",
u"linkfa",
u"na[ _]lotura",
u"nasc[ _]ar",
u"tengill[ _][úg]g",
u"ua",
u"yüm yg",
u"רא",
u"وصلة مقالة جيدة",
u"وصلة مقالة مختارة",
]
errorCount = 0
site = wikipedia.getSite()
# /wiki/ is not always the right path in non-wiki projects
pathWiki = site.family.nicepath(site.lang)
if putText:
wikipedia.output(u"Loading %s..." % page.title())
if oldTextGiven == None:
try:
text = page.get()
except wikipedia.NoPage:
if create:
wikipedia.output(u"%s doesn't exist, creating it!" % page.title())
text = u""
else:
wikipedia.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always) # continue
except wikipedia.IsRedirectPage:
wikipedia.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always) # continue
else:
text = oldTextGiven
# Understand if the bot has to skip the page or not
# In this way you can use both -except and -excepturl
if regexSkipUrl != None:
url = "%s%s" % (pathWiki, page.urlname())
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
wikipedia.output(u"Exception! regex (or word) used with -exceptUrl is in the page. Skip!")
return (False, False, always) # continue
if regexSkip != None:
result = re.findall(regexSkip, text)
if result != []:
wikipedia.output(u"Exception! regex (or word) used with -except is in the page. Skip!")
return (False, False, always) # continue
# If not up, text put below
if not up:
newtext = text
# Getting the categories
categoriesInside = wikipedia.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = wikipedia.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = wikipedia.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = wikipedia.removeLanguageLinks(newtext, site)
# nn got a message between the categories and the iw's and they want to keep it there, first remove it
if site.language() == u"nn":
newtext = newtext.replace(nn_iw_msg, "")
# Translating the \\n into binary \n
addText = addText.replace("\\n", "\n")
# Adding the text
newtext += u"\n%s" % addText
# Reputting the categories
newtext = wikipedia.replaceCategoryLinks(newtext, categoriesInside, site, True)
# Put the nn iw message back
if site.language() == u"nn":
newtext = newtext + u"\n" + nn_iw_msg
# Dealing the stars' issue
allstars = []
starstext = wikipedia.removeDisabledParts(text)
for star in starsList:
regex = re.compile("(\{\{(?:template:|)%s\|.*?\}\}[\s]*)" % star, re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub("", newtext)
allstars += found
if allstars != []:
newtext = newtext.strip() + "\r\n\r\n"
allstars.sort()
for element in allstars:
newtext += "%s\r\n" % element.strip()
# Adding the interwiki
newtext = wikipedia.replaceLanguageLinks(newtext, interwikiInside, site)
# If instead the text must be added above...
else:
newtext = addText + "\n" + text
if putText and text != newtext:
wikipedia.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<" % page.title())
wikipedia.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if not always:
choice = wikipedia.inputChoice(
u"Do you want to accept these changes?", ["Yes", "No", "All"], ["y", "N", "a"], "N"
)
if choice == "a":
always = True
elif choice == "n":
return (False, False, always)
if always or choice == "y":
try:
if always:
page.put(newtext, summary)
else:
page.put_async(newtext, summary)
except wikipedia.EditConflict:
wikipedia.output(u"Edit conflict! skip!")
return (False, False, always)
except wikipedia.ServerError:
errorCount += 1
if errorCount < 5:
wikipedia.output(u"Server Error! Wait..")
time.sleep(5)
continue
else:
raise wikipedia.ServerError(u"Fifth Server Error!")
except wikipedia.SpamfilterError, e:
wikipedia.output(u"Cannot change %s because of blacklist entry %s" % (page.title(), e.url))
return (False, False, always)
except wikipedia.PageNotSaved, error:
wikipedia.output(u"Error putting page: %s" % error.args)
return (False, False, always)
except wikipedia.LockedPage:
wikipedia.output(u"Skipping %s (locked page)" % page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
def standardizePageFooter(self, text):
"""
Makes sure that interwiki links, categories and star templates are
put to the correct position and into the right order. This combines the
old instances standardizeInterwiki and standardizeCategories
The page footer has the following section in that sequence:
1. categories
2. ## TODO: template beyond categories ##
3. additional information depending on local site policy
4. stars templates for featured and good articles
5. interwiki links
"""
starsList = [
u'bueno',
u'bom interwiki',
u'cyswllt[ _]erthygl[ _]ddethol', u'dolen[ _]ed',
u'destacado', u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligazón[ _]a[bd]',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a', u'link[ _]adq', u'link[ _]f[lm]', u'link[ _]km',
u'link[ _]sm', u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
categories = None
interwikiLinks = None
allstars = []
# The PyWikipediaBot is no longer allowed to touch categories on the
# German Wikipedia. See
# http://de.wikipedia.org/wiki/Hilfe_Diskussion:Personendaten/Archiv/1#Position_der_Personendaten_am_.22Artikelende.22
# ignoring nn-wiki of cause of the comment line above iw section
if not self.template and '{{Personendaten' not in text and \
'{{SORTIERUNG' not in text and '{{DEFAULTSORT' not in text and \
self.site.lang not in ('et', 'it', 'bg', 'ru'):
try:
categories = pywikibot.getCategoryLinks(text, site=self.site)
# there are categories like [[categoy:Foo {{#time:Y...}}]]
except pywikibot.InvalidTitle:
pass
if not self.talkpage: # and pywikibot.calledModuleName() <> 'interwiki':
subpage = False
if self.template:
loc = None
try:
tmpl, loc = moved_links[self.site.lang]
del tmpl
except KeyError:
pass
if loc is not None and loc in self.title:
subpage = True
interwikiLinks = pywikibot.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = pywikibot.removeLanguageLinks(text, site=self.site)
# Removing the stars' issue
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
text = regex.sub('', text)
allstars += found
# Adding categories
if categories:
##Sorting categories in alphabetic order. beta test only on Persian Wikipedia, TODO fix bug for sorting
#if self.site.language() == 'fa':
# categories.sort()
##Taking main cats to top
# for name in categories:
# if re.search(u"(.+?)\|(.{,1}?)",name.title()) or name.title()==name.title().split(":")[0]+title:
# categories.remove(name)
# categories.insert(0, name)
text = pywikibot.replaceCategoryLinks(text, categories,
site=self.site)
# Adding stars templates
if allstars:
text = text.strip() + self.site.family.interwiki_text_separator
allstars.sort()
for element in allstars:
text += '%s\r\n' % element.strip()
if pywikibot.verbose:
pywikibot.output(u'%s' % element.strip())
# Adding the interwiki
if interwikiLinks:
text = pywikibot.replaceLanguageLinks(text, interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def add_text(
page=None,
addText=None,
summary=None,
regexSkip=None,
regexSkipUrl=None,
always=False,
up=False,
putText=True,
oldTextGiven=None,
reorderEnabled=True,
create=False,
):
if not addText:
raise NoEnoughData("You have to specify what text you want to add!")
if not summary:
summary = i18n.twtranslate(pywikibot.getSite(), "add_text-adding", {"adding": addText[:200]})
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
errorCount = 0
site = pywikibot.getSite()
pathWiki = site.family.nicepath(site.lang)
if putText:
pywikibot.output(u"Loading %s..." % page.title())
if oldTextGiven is None:
try:
text = page.get()
except pywikibot.NoPage:
if create:
pywikibot.output(u"%s doesn't exist, creating it!" % page.title())
text = u""
else:
pywikibot.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always)
except pywikibot.IsRedirectPage:
pywikibot.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always)
else:
text = oldTextGiven
# Understand if the bot has to skip the page or not
# In this way you can use both -except and -excepturl
if regexSkipUrl is not None:
url = "%s%s" % (pathWiki, page.urlname())
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output(
u"""Exception! regex (or word) used with -exceptUrl is in the page. Skip!
Match was: %s"""
% result
)
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output(
u"""Exception! regex (or word) used with -except is in the page. Skip!
Match was: %s"""
% result
)
return (False, False, always)
# If not up, text put below
if not up:
newtext = text
# Translating the \\n into binary \n
addText = addText.replace("\\n", config.line_separator)
if reorderEnabled:
# Getting the categories
categoriesInside = pywikibot.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = pywikibot.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = pywikibot.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = pywikibot.removeLanguageLinks(newtext, site)
# Adding the text
newtext += u"%s%s" % (config.line_separator, addText)
# Reputting the categories
newtext = pywikibot.replaceCategoryLinks(newtext, categoriesInside, site, True)
# Dealing the stars' issue
allstars = []
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile("(\{\{(?:template:|)%s\|.*?\}\}[\s]*)" % star, re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub("", newtext)
allstars += found
if allstars != []:
newtext = newtext.strip() + config.line_separator * 2
allstars.sort()
for element in allstars:
newtext += "%s%s" % (element.strip(), config.LS)
# Adding the interwiki
newtext = pywikibot.replaceLanguageLinks(newtext, interwikiInside, site)
else:
newtext += u"%s%s" % (config.line_separator, addText)
else:
newtext = addText + config.line_separator + text
if putText and text != newtext:
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<" % page.title())
pywikibot.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if not always:
choice = pywikibot.inputChoice(
u"Do you want to accept these changes?",
["Yes", "No", "All", "open in Browser"],
["y", "n", "a", "b"],
"n",
)
if choice == "a":
always = True
elif choice == "n":
return (False, False, always)
elif choice == "b":
webbrowser.open("http://%s%s" % (page.site.hostname(), page.site.nice_get_address(page.title())))
pywikibot.input("Press Enter when finished in browser.")
if always or choice == "y":
try:
if always:
page.put(newtext, summary, minorEdit=page.namespace() != 3)
else:
page.put_async(newtext, summary, minorEdit=page.namespace() != 3)
except pywikibot.EditConflict:
pywikibot.output(u"Edit conflict! skip!")
return (False, False, always)
except pywikibot.ServerError:
errorCount += 1
if errorCount < config.maxretries:
pywikibot.output(u"Server Error! Wait..")
time.sleep(5)
continue
else:
raise pywikibot.ServerError(u"Fifth Server Error!")
except pywikibot.SpamfilterError as e:
pywikibot.output(u"Cannot change %s because of blacklist entry %s" % (page.title(), e.url))
return (False, False, always)
except pywikibot.PageNotSaved as error:
pywikibot.output(u"Error putting page: %s" % error.args)
return (False, False, always)
except pywikibot.LockedPage:
pywikibot.output(u"Skipping %s (locked page)" % page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
else:
return (text, newtext, always)
def add_text(page=None, addText=None, summary=None, regexSkip=None,
regexSkipUrl=None, always=False, up=False, putText=True,
oldTextGiven=None, reorderEnabled=True, create=False):
if not addText:
raise NoEnoughData('You have to specify what text you want to add!')
if not summary:
summary = i18n.twtranslate(pywikibot.getSite(), 'add_text-adding',
{'adding': addText[:200]})
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
errorCount = 0
site = pywikibot.getSite()
pathWiki = site.family.nicepath(site.lang)
if putText:
pywikibot.output(u'Loading %s...' % page.title())
if oldTextGiven is None:
try:
text = page.get()
except pywikibot.NoPage:
if create:
pywikibot.output(u"%s doesn't exist, creating it!"
% page.title())
text = u''
else:
pywikibot.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always)
except pywikibot.IsRedirectPage:
pywikibot.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always)
else:
text = oldTextGiven
# Understand if the bot has to skip the page or not
# In this way you can use both -except and -excepturl
if regexSkipUrl is not None:
url = '%s%s' % (pathWiki, page.urlname())
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -exceptUrl is in the page. Skip!
Match was: %s''' % result)
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -except is in the page. Skip!
Match was: %s''' % result)
return (False, False, always)
# If not up, text put below
if not up:
newtext = text
# Translating the \\n into binary \n
addText = addText.replace('\\n', config.line_separator)
if (reorderEnabled):
# Getting the categories
categoriesInside = pywikibot.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = pywikibot.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = pywikibot.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = pywikibot.removeLanguageLinks(newtext, site)
# Adding the text
newtext += u"%s%s" % (config.line_separator, addText)
# Reputting the categories
newtext = pywikibot.replaceCategoryLinks(newtext,
categoriesInside, site,
True)
# Dealing the stars' issue
allstars = []
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub('', newtext)
allstars += found
if allstars != []:
newtext = newtext.strip() + config.line_separator * 2
allstars.sort()
for element in allstars:
newtext += '%s%s' % (element.strip(), config.LS)
# Adding the interwiki
newtext = pywikibot.replaceLanguageLinks(newtext, interwikiInside,
site)
else:
newtext += u"%s%s" % (config.line_separator, addText)
else:
newtext = addText + config.line_separator + text
if putText and text != newtext:
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<"
% page.title())
pywikibot.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if not always:
choice = pywikibot.inputChoice(
u'Do you want to accept these changes?',
['Yes', 'No', 'All', 'open in Browser'],
['y', 'n', 'a', 'b'], 'n')
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
webbrowser.open("http://%s%s" % (
page.site.hostname(),
page.site.nice_get_address(page.title())
))
pywikibot.input("Press Enter when finished in browser.")
if always or choice == 'y':
try:
if always:
page.put(newtext, summary,
minorEdit=page.namespace() != 3)
else:
page.put_async(newtext, summary,
minorEdit=page.namespace() != 3)
except pywikibot.EditConflict:
pywikibot.output(u'Edit conflict! skip!')
return (False, False, always)
except pywikibot.ServerError:
errorCount += 1
if errorCount < config.maxretries:
pywikibot.output(u'Server Error! Wait..')
time.sleep(5)
continue
else:
raise pywikibot.ServerError(u'Fifth Server Error!')
except pywikibot.SpamfilterError as e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s'
% (page.title(), e.url))
return (False, False, always)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: %s' % error.args)
return (False, False, always)
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
else:
return (text, newtext, always)
def add_text(page = None, addText = None, summary = None, regexSkip = None,
regexSkipUrl = None, always = False, up = False, putText = True,
oldTextGiven = None, reorderEnabled = True, create=False):
if not addText:
raise NoEnoughData('You have to specify what text you want to add!')
if not summary:
summary = i18n.twtranslate(pywikibot.getSite(), 'add_text-adding',
{'adding': addText[:200]})
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
starsList = [
u'bueno',
u'cyswllt[ _]erthygl[ _]ddethol', u'dolen[ _]ed',
u'destacado', u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a', u'link[ _]adq', u'link[ _]f[lm]', u'link[ _]km',
u'link[ _]sm', u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
errorCount = 0
site = pywikibot.getSite()
# /wiki/ is not always the right path in non-wiki projects
pathWiki = site.family.nicepath(site.lang)
if putText:
pywikibot.output(u'Loading %s...' % page.title())
if oldTextGiven == None:
try:
text = page.get()
except pywikibot.NoPage:
if create:
pywikibot.output(u"%s doesn't exist, creating it!"
% page.title())
text = u''
else:
pywikibot.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always) # continue
except pywikibot.IsRedirectPage:
pywikibot.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always) # continue
else:
text = oldTextGiven
# Understand if the bot has to skip the page or not
# In this way you can use both -except and -excepturl
if regexSkipUrl != None:
url = '%s%s' % (pathWiki, page.urlname())
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -exceptUrl is in the page. Skip!
Match was: %s''' % result)
return (False, False, always) # continue
if regexSkip != None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -except is in the page. Skip!
Match was: %s''' % result)
return (False, False, always) # continue
# If not up, text put below
if not up:
newtext = text
# Translating the \\n into binary \n
addText = addText.replace('\\n', '\n')
if (reorderEnabled):
# Getting the categories
categoriesInside = pywikibot.getCategoryLinks(newtext, site)
# Deleting the categories
#newtext = pywikibot.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = pywikibot.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = pywikibot.removeLanguageLinks(newtext, site)
# nn got a message between the categories and the iw's
# and they want to keep it there, first remove it
hasCommentLine = False
if (site.language()==u'nn'):
regex = re.compile('(<!-- ?interwiki \(no(?:/nb)?, ?sv, ?da first; then other languages alphabetically by name\) ?-->)')
found = regex.findall(newtext)
if found:
hasCommentLine = True
newtext = regex.sub('', newtext)
# Adding the text
newtext += u"\n%s" % addText
# Reputting the categories
#newtext = pywikibot.replaceCategoryLinks(newtext,
#categoriesInside, site, True)
#Put the nn iw message back
if site.language()==u'nn' and (interwikiInside or hasCommentLine):
newtext = newtext + u'\r\n\r\n' + nn_iw_msg
# Dealing the stars' issue
allstars = []
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)' % star,
re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub('', newtext)
allstars += found
if allstars != []:
newtext = newtext.strip()+'\r\n\r\n'
allstars.sort()
for element in allstars:
newtext += '%s\r\n' % element.strip()
# Adding the interwiki
newtext = pywikibot.replaceLanguageLinks(newtext, interwikiInside, site)
else:
# Adding the text
newtext += u"\n%s" % addText
# If instead the text must be added above...
else:
newtext = addText + '\n' + text
if putText and text != newtext:
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<"
% page.title())
pywikibot.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if not always:
choice = pywikibot.inputChoice(
u'Do you want to accept these changes?',
['Yes', 'No', 'All'], ['y', 'N', 'a'], 'N')
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
if always or choice == 'y':
try:
if always:
page.put(newtext, summary)
else:
page.put_async(newtext, summary)
except pywikibot.EditConflict:
pywikibot.output(u'Edit conflict! skip!')
return (False, False, always)
except pywikibot.ServerError:
errorCount += 1
if errorCount < 5:
pywikibot.output(u'Server Error! Wait..')
time.sleep(5)
continue
else:
raise pywikibot.ServerError(u'Fifth Server Error!')
except pywikibot.SpamfilterError, e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s'
% (page.title(), e.url))
return (False, False, always)
except pywikibot.PageNotSaved, error:
pywikibot.output(u'Error putting page: %s' % error.args)
return (False, False, always)
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
def run(self):
""" Runs the Bot """
try:
deadLinks = codecs.open(listof404pages, 'r', 'latin_1').read()
except IOError:
pywikibot.output(
'You need to download '
'http://www.twoevils.org/files/wikipedia/404-links.txt.gz '
'and to ungzip it in the same directory')
raise
socket.setdefaulttimeout(30)
editedpages = 0
for page in self.generator:
try:
# Load the page's text from the wiki
new_text = page.get()
if not page.canBeEdited():
pywikibot.output(u"You can't edit page %s" %
page.title(asLink=True))
continue
except pywikibot.NoPage:
pywikibot.output(u'Page %s not found' %
page.title(asLink=True))
continue
except pywikibot.IsRedirectPage:
pywikibot.output(u'Page %s is a redirect' %
page.title(asLink=True))
continue
# for each link to change
for match in linksInRef.finditer(
pywikibot.removeDisabledParts(page.get())):
link = match.group(u'url')
#debugging purpose
#print link
if u'jstor.org' in link:
#TODO: Clean URL blacklist
continue
ref = RefLink(link, match.group('name'))
f = None
try:
socket.setdefaulttimeout(20)
try:
f = urllib2.urlopen(ref.url.decode("utf8"))
except UnicodeError:
ref.url = urllib2.quote(ref.url.encode("utf8"), "://")
f = urllib2.urlopen(ref.url)
#Try to get Content-Type from server
headers = f.info()
contentType = headers.getheader('Content-Type')
if contentType and not self.MIME.search(contentType):
if ref.link.lower().endswith('.pdf') and \
not self.ignorepdf:
# If file has a PDF suffix
self.getPDFTitle(ref, f)
else:
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : '
u'media : %s ' % ref.link)
if ref.title:
if not re.match(
u'(?i) *microsoft (word|excel|visio)',
ref.title):
ref.transform(ispdf=True)
repl = ref.refTitle()
else:
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : '
u'PDF title blacklisted : %s ' % ref.title)
repl = ref.refLink()
else:
repl = ref.refLink()
new_text = new_text.replace(match.group(), repl)
continue
# Get the real url where we end (http redirects !)
redir = f.geturl()
if redir != ref.link and \
domain.findall(redir) == domain.findall(link):
if soft404.search(redir) and \
not soft404.search(ref.link):
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : '
u'Redirect 404 : %s ' % ref.link)
continue
if dirIndex.match(redir) and \
not dirIndex.match(ref.link):
pywikibot.output(
u'\03{lightyellow}WARNING\03{default} : '
u'Redirect to root : %s ' % ref.link)
continue
# uncompress if necessary
if headers.get('Content-Encoding') in ('gzip', 'x-gzip'):
# XXX: small issue here: the whole page is downloaded
# through f.read(). It might fetch big files/pages.
# However, truncating an encoded gzipped stream is not
# an option, for unzipping will fail.
compressed = StringIO.StringIO(f.read())
f = gzip.GzipFile(fileobj=compressed)
# Read the first 1,000,000 bytes (0.95 MB)
linkedpagetext = f.read(1000000)
socket.setdefaulttimeout(None)
except UnicodeError:
# example : http://www.adminet.com/jo/20010615¦/ECOC0100037D.html
# in [[fr:Cyanure]]
pywikibot.output(
u'\03{lightred}Bad link\03{default} : %s in %s' %
(ref.url, page.title(asLink=True)))
continue
except urllib2.HTTPError, e:
pywikibot.output(
u'HTTP error (%s) for %s on %s' %
(e.code, ref.url, page.title(asLink=True)),
toStdout=True)
# 410 Gone, indicates that the resource has been purposely
# removed
if e.code == 410 or \
(e.code == 404 and (u'\t%s\t' % ref.url in deadLinks)):
repl = ref.refDead()
new_text = new_text.replace(match.group(), repl)
continue
except (urllib2.URLError, socket.error, IOError,
httplib.error), e:
pywikibot.output(u'Can\'t retrieve page %s : %s' %
(ref.url, e))
continue
except ValueError:
# Known bug of httplib, google for :
# "httplib raises ValueError reading chunked content"
continue
def standardizePageFooter(self, text):
"""
Makes sure that interwiki links, categories and star templates are
put to the correct position and into the right order.
This combines the old instances standardizeInterwiki and standardizeCategories
The page footer has the following section in that sequence:
1. categories
2. additional information depending on local site policy
3. stars templates for featured and good articles
4. interwiki links
"""
starsList = [
u"bueno",
u"cyswllt[ _]erthygl[ _]ddethol",
u"dolen[ _]ed",
u"destacado",
u"destaca[tu]",
u"enllaç[ _]ad",
u"enllaz[ _]ad",
u"leam[ _]vdc",
u"legătură[ _]a[bcf]",
u"liamm[ _]pub",
u"lien[ _]adq",
u"lien[ _]ba",
u"liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt",
u"liên[ _]kết[ _]chọn[ _]lọc",
u"ligam[ _]adq",
u"ligoelstara",
u"ligoleginda",
u"link[ _][afgu]a",
u"link[ _]adq",
u"link[ _]f[lm]",
u"link[ _]km",
u"link[ _]sm",
u"linkfa",
u"na[ _]lotura",
u"nasc[ _]ar",
u"tengill[ _][úg]g",
u"ua",
u"yüm yg",
u"רא",
u"وصلة مقالة جيدة",
u"وصلة مقالة مختارة",
]
categories = None
interwikiLinks = None
allstars = []
hasCommentLine = False
# The PyWikipediaBot is no longer allowed to touch categories on the German Wikipedia.
# See http://de.wikipedia.org/wiki/Hilfe_Diskussion:Personendaten/Archiv/bis_2006#Position_der_Personendaten_am_.22Artikelende.22
# ignoring nn-wiki of cause of the comment line above iw section
if not self.template and not "{{Personendaten" in text:
categories = pywikibot.getCategoryLinks(text, site=self.site)
if not self.talkpage: # and pywikibot.calledModuleName() <> 'interwiki':
subpage = False
if self.template:
loc = None
try:
tmpl, loc = moved_links[self.site.lang]
del tmpl
except KeyError:
pass
if loc != None and loc in self.title:
subpage = True
interwikiLinks = pywikibot.getLanguageLinks(text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = pywikibot.removeLanguageLinks(text, site=self.site)
# Removing the stars' issue
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile("(\{\{(?:template:|)%s\|.*?\}\}[\s]*)" % star, re.I)
found = regex.findall(starstext)
if found != []:
if pywikibot.verbose:
print found
text = regex.sub("", text)
allstars += found
# nn got a message between the categories and the iw's
# and they want to keep it there, first remove it
if self.site.language() == "nn":
regex = re.compile(
"(<!-- ?interwiki \(no(?:/nb)?, ?sv, ?da first; then other languages alphabetically by name\) ?-->)"
)
found = regex.findall(text)
if found:
if pywikibot.verbose:
print found
hasCommentLine = True
text = regex.sub("", text)
# Adding categories
if categories:
text = pywikibot.replaceCategoryLinks(text, categories, site=self.site)
# Put the nn iw message back
if self.site.language() == "nn" and not self.talkpage and (interwikiLinks or hasCommentLine):
text = text + "\r\n\r\n" + nn_iw_msg
# Adding stars templates
if allstars:
text = text.strip() + self.site.family.interwiki_text_separator
allstars.sort()
for element in allstars:
text += "%s\r\n" % element.strip()
if pywikibot.verbose:
pywikibot.output(u"%s" % element.strip())
# Adding the interwiki
if interwikiLinks:
text = pywikibot.replaceLanguageLinks(
text, interwikiLinks, site=self.site, template=self.template, template_subpage=subpage
)
return text
def add_text(page = None, addText = None, summary = None, regexSkip = None,
regexSkipUrl = None, always = False, up = False, putText = True,
oldTextGiven = None, reorderEnabled = True, create=False):
if not addText:
raise NoEnoughData('You have to specify what text you want to add!')
if not summary:
summary = i18n.twtranslate(pywikibot.getSite(), 'add_text-adding',
{'adding': addText[:200]})
# When a page is tagged as "really well written" it has a star in the
# interwiki links. This is a list of all the templates used (in regex
# format) to make the stars appear.
starsList = [
u'bueno',
u'bom interwiki',
u'cyswllt[ _]erthygl[ _]ddethol', u'dolen[ _]ed',
u'destacado', u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a', u'link[ _]adq', u'link[ _]f[lm]', u'link[ _]km',
u'link[ _]sm', u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
errorCount = 0
site = pywikibot.getSite()
# /wiki/ is not always the right path in non-wiki projects
pathWiki = site.family.nicepath(site.lang)
if putText:
pywikibot.output(u'Loading %s...' % page.title())
if oldTextGiven == None:
try:
text = page.get()
except pywikibot.NoPage:
if create:
pywikibot.output(u"%s doesn't exist, creating it!"
% page.title())
text = u''
else:
pywikibot.output(u"%s doesn't exist, skip!" % page.title())
return (False, False, always) # continue
except pywikibot.IsRedirectPage:
pywikibot.output(u"%s is a redirect, skip!" % page.title())
return (False, False, always) # continue
else:
text = oldTextGiven
# Understand if the bot has to skip the page or not
# In this way you can use both -except and -excepturl
if regexSkipUrl != None:
url = '%s%s' % (pathWiki, page.urlname())
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -exceptUrl is in the page. Skipping!
Match was: %s''' % result)
return (False, False, always) # continue
if regexSkip != None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output(
u'''Exception! regex (or word) used with -except is in the page. Skipping!
Match was: %s''' % result)
return (False, False, always) # continue
# If not up, text put below
if not up:
newtext = text
# Translating the \\n into binary \n
addText = addText.replace('\\n', '\n')
if (reorderEnabled):
# Getting the categories
categoriesInside = pywikibot.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = pywikibot.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = pywikibot.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = pywikibot.removeLanguageLinks(newtext, site)
# nn got a message between the categories and the iw's
# and they want to keep it there, first remove it
hasCommentLine = False
if (site.language()==u'nn'):
regex = re.compile('(<!-- ?interwiki \(no(?:/nb)?, ?sv, ?da first; then other languages alphabetically by name\) ?-->)')
found = regex.findall(newtext)
if found:
hasCommentLine = True
newtext = regex.sub('', newtext)
# Adding the text
newtext += u"\n%s" % addText
# Reputting the categories
newtext = pywikibot.replaceCategoryLinks(newtext,
categoriesInside, site, True)
#Put the nn iw message back
if site.language()==u'nn' and (interwikiInside or hasCommentLine):
newtext = newtext + u'\r\n\r\n' + nn_iw_msg
# Dealing the stars' issue
allstars = []
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)' % star,
re.I)
found = regex.findall(starstext)
if found != []:
newtext = regex.sub('', newtext)
allstars += found
if allstars != []:
newtext = newtext.strip()+'\r\n\r\n'
allstars.sort()
for element in allstars:
newtext += '%s\r\n' % element.strip()
# Adding the interwiki
newtext = pywikibot.replaceLanguageLinks(newtext, interwikiInside, site)
else:
# Adding the text
newtext += u"\n%s" % addText
# If instead the text must be added above...
else:
newtext = addText + '\n' + text
if putText and text != newtext:
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<"
% page.title())
pywikibot.showDiff(text, newtext)
# Let's put the changes.
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if putText:
if not always:
choice = pywikibot.inputChoice(
u'Do you want to accept these changes?',
['Yes', 'No', 'All', 'open in Browser'], ['y', 'N', 'a', 'b'], 'N')
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
webbrowser.open("http://%s%s" % (
page.site().hostname(),
page.site().nice_get_address(page.title())
))
pywikibot.input("Press Enter when finished in browser.")
if always or choice == 'y':
try:
if always:
page.put(newtext, summary)
else:
page.put_async(newtext, summary)
except pywikibot.EditConflict:
pywikibot.output(u'Edit conflict! skip!')
return (False, False, always)
except pywikibot.ServerError:
errorCount += 1
if errorCount < 5:
pywikibot.output(u'Server Error! Wait..')
time.sleep(5)
continue
else:
raise pywikibot.ServerError(u'Fifth Server Error!')
except pywikibot.SpamfilterError, e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s'
% (page.title(), e.url))
return (False, False, always)
except pywikibot.PageNotSaved, error:
pywikibot.output(u'Error putting page: %s' % error.args)
return (False, False, always)
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% page.title())
return (False, False, always)
else:
# Break only if the errors are one after the other...
errorCount = 0
return (True, True, always)
def standardizePageFooter(self, text):
"""
Makes sure that interwiki links, categories and star templates are
put to the correct position and into the right order. This combines the
old instances standardizeInterwiki and standardizeCategories
The page footer has the following section in that sequence:
1. categories
2. additional information depending on local site policy
3. stars templates for featured and good articles
4. interwiki links
"""
starsList = [
u'bueno',
u'bom interwiki',
u'cyswllt[ _]erthygl[ _]ddethol', u'dolen[ _]ed',
u'destacado', u'destaca[tu]',
u'enllaç[ _]ad',
u'enllaz[ _]ad',
u'leam[ _]vdc',
u'legătură[ _]a[bcf]',
u'liamm[ _]pub',
u'lien[ _]adq',
u'lien[ _]ba',
u'liên[ _]kết[ _]bài[ _]chất[ _]lượng[ _]tốt',
u'liên[ _]kết[ _]chọn[ _]lọc',
u'ligam[ _]adq',
u'ligoelstara',
u'ligoleginda',
u'link[ _][afgu]a', u'link[ _]adq', u'link[ _]f[lm]', u'link[ _]km',
u'link[ _]sm', u'linkfa',
u'na[ _]lotura',
u'nasc[ _]ar',
u'tengill[ _][úg]g',
u'ua',
u'yüm yg',
u'רא',
u'وصلة مقالة جيدة',
u'وصلة مقالة مختارة',
]
categories = None
interwikiLinks = None
allstars = []
hasCommentLine = False
# The PyWikipediaBot is no longer allowed to touch categories on the
# German Wikipedia. See
# http://de.wikipedia.org/wiki/Hilfe_Diskussion:Personendaten/Archiv/1#Position_der_Personendaten_am_.22Artikelende.22
# ignoring nn-wiki of cause of the comment line above iw section
if not self.template and not '{{Personendaten' in text:
categories = pywikibot.getCategoryLinks(text, site = self.site)
if not self.talkpage:# and pywikibot.calledModuleName() <> 'interwiki':
subpage = False
if self.template:
loc = None
try:
tmpl, loc = moved_links[self.site.lang]
del tmpl
except KeyError:
pass
if loc != None and loc in self.title:
subpage = True
interwikiLinks = pywikibot.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = pywikibot.removeLanguageLinks(text, site = self.site)
# Removing the stars' issue
starstext = pywikibot.removeDisabledParts(text)
for star in starsList:
regex = re.compile('(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
if pywikibot.verbose:
print found
text = regex.sub('', text)
allstars += found
# nn got a message between the categories and the iw's
# and they want to keep it there, first remove it
if self.site.lang in msg_interwiki:
iw_msg = msg_interwiki[self.site.lang]
if isinstance(iw_msg, tuple):
iw_reg = iw_msg[1]
iw_msg = iw_msg[0]
else:
iw_reg = u'(%s)' % iw_msg
regex = re.compile(iw_reg)
found = regex.findall(text)
if found:
if pywikibot.verbose:
print found
hasCommentLine = True
text = regex.sub('', text)
# Adding categories
if categories:
text = pywikibot.replaceCategoryLinks(text, categories,
site=self.site)
# Put the iw message back
if not self.talkpage and \
((interwikiLinks or hasCommentLine) and
self.site.language() == 'nn' or
(interwikiLinks and hasCommentLine) and
self.site.language() == 'fr'):
text = text + '\r\n\r\n' + iw_msg
# Adding stars templates
if allstars:
text = text.strip()+self.site.family.interwiki_text_separator
allstars.sort()
for element in allstars:
text += '%s\r\n' % element.strip()
if pywikibot.verbose:
pywikibot.output(u'%s' %element.strip())
# Adding the interwiki
if interwikiLinks:
text = pywikibot.replaceLanguageLinks(text, interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
