def ReplaceLink(self, text, oldtxt, newtxt):
frmParts = [s.strip(self.stripChars) for s in self.wordBreaker.split(oldtxt)]
toParts = [s.strip(self.stripChars) for s in self.wordBreaker.split(newtxt)]
if len(frmParts) != len(toParts):
raise ValueError("Splitting parts do not match counts")
for i in xrange(0, len(frmParts)):
if len(frmParts[i]) != len(toParts[i]):
raise ValueError("Splitting parts do not match word length")
if len(frmParts[i]) > 0:
text = text.replace(first_lower(frmParts[i]), first_lower(toParts[i]))
text = text.replace(first_upper(frmParts[i]), first_upper(toParts[i]))
return text
def treat_page_and_item(self, page, item):
title = page.properties().get('displaytitle')
if not title:
return
page_title = page.title()
if first_lower(page_title) != title:
return
lang = page.site.lang
label = item.labels.get(lang)
if not label or self.stripped(label) == self.stripped(page_title):
item.labels[lang] = first_lower(label) if label else title
summary = 'importing [%s] label from displaytitle in %s' % (
lang, page.title(as_link=True, insite=item.site))
self.user_edit_entity(item, summary=summary)
def translateAndCapitalizeNamespaces(self, text):
"""Use localized namespace names."""
# arz uses english stylish codes
if self.site.sitename == 'wikipedia:arz':
return text
# wiki links aren't parsed here.
exceptions = ['nowiki', 'comment', 'math', 'pre']
for namespace in self.site.namespaces.values():
if namespace == 0:
# skip main (article) namespace
continue
# a clone is needed. Won't change the namespace dict
namespaces = list(namespace)
if namespace == 6 and self.site.family.name == 'wikipedia':
if self.site.code in ('en', 'fr') and MediaWikiVersion(
self.site.version()) >= MediaWikiVersion('1.14'):
# do not change "Image" on en-wiki and fr-wiki
assert u'Image' in namespaces
namespaces.remove(u'Image')
if self.site.code == 'hu':
# do not change "Kép" on hu-wiki
assert u'Kép' in namespaces
namespaces.remove(u'Kép')
elif self.site.code == 'pt':
# use "Imagem" by default on pt-wiki (per T57242)
assert 'Imagem' in namespaces
namespaces.insert(
0, namespaces.pop(namespaces.index('Imagem')))
# final namespace variant
final_ns = namespaces.pop(0)
if namespace in (2, 3):
# skip localized user namespace, maybe gender is used
namespaces = ['User' if namespace == 2 else 'User talk']
# lowerspaced and underscored namespaces
for i, item in enumerate(namespaces):
item = item.replace(' ', '[ _]')
item = u'[%s%s]' % (item[0], item[0].lower()) + item[1:]
namespaces[i] = item
namespaces.append(first_lower(final_ns))
if final_ns and namespaces:
if self.site.sitename == 'wikipedia:pt' and namespace == 6:
# only change on these file extensions (per T57242)
extensions = ('png', 'gif', 'jpg', 'jpeg', 'svg', 'tiff',
'tif')
text = textlib.replaceExcept(
text,
r'\[\[\s*({}) *:(?P<name>[^\|\]]*?\.({}))'
r'(?P<label>.*?)\]\]'
.format('|'.join(namespaces), '|'.join(extensions)),
r'[[{}:\g<name>\g<label>]]'.format(final_ns),
exceptions)
else:
text = textlib.replaceExcept(
text,
r'\[\[\s*(%s) *:(?P<nameAndLabel>.*?)\]\]'
% '|'.join(namespaces),
r'[[%s:\g<nameAndLabel>]]' % final_ns,
exceptions)
return text
def ReplaceLink(self, text, oldtxt, newtxt):
"""Replace links."""
frmParts = [
s.strip(self.stripChars) for s in self.wordBreaker.split(oldtxt)
]
toParts = [
s.strip(self.stripChars) for s in self.wordBreaker.split(newtxt)
]
if len(frmParts) != len(toParts):
raise ValueError('Splitting parts do not match counts')
for i, part in enumerate(frmParts):
if part != len(toParts[i]):
raise ValueError('Splitting parts do not match word length')
if part:
text = text.replace(first_lower(part), first_lower(toParts[i]))
text = text.replace(first_upper(part), first_upper(toParts[i]))
return text
def correctcap(link, text):
# If text links to a page with title link uncapitalized, uncapitalize link,
# otherwise capitalize it
linkupper = link.title()
linklower = first_lower(linkupper)
if "[[%s]]" % linklower in text or "[[%s|" % linklower in text:
return linklower
else:
return linkupper
def correctcap(link, text):
# If text links to a page with title link uncapitalized, uncapitalize link,
# otherwise capitalize it
linkupper = link.title()
linklower = first_lower(linkupper)
if "[[%s]]" % linklower in text or "[[%s|" % linklower in text:
return linklower
else:
return linkupper
def translateAndCapitalizeNamespaces(self, text):
"""Use localized namespace names."""
# arz uses english stylish codes
if self.site.sitename == 'wikipedia:arz':
return text
# wiki links aren't parsed here.
exceptions = ['nowiki', 'comment', 'math', 'pre']
for namespace in self.site.namespaces.values():
if namespace == 0:
# skip main (article) namespace
continue
# a clone is needed. Won't change the namespace dict
namespaces = list(namespace)
if namespace == 6 and self.site.family.name == 'wikipedia':
if self.site.code in ('en', 'fr') \
and self.site.mw_version >= '1.14':
# do not change "Image" on en-wiki and fr-wiki
assert 'Image' in namespaces
namespaces.remove('Image')
if self.site.code == 'hu':
# do not change "Kép" on hu-wiki
assert 'Kép' in namespaces
namespaces.remove('Kép')
elif self.site.code == 'pt':
# use "Imagem" by default on pt-wiki (per T57242)
assert 'Imagem' in namespaces
namespaces.insert(
0, namespaces.pop(namespaces.index('Imagem')))
# final namespace variant
final_ns = namespaces.pop(0)
if namespace in (2, 3):
# skip localized user namespace, maybe gender is used
namespaces = ['User' if namespace == 2 else 'User talk']
# lowerspaced and underscored namespaces
for i, item in enumerate(namespaces):
item = item.replace(' ', '[ _]')
item = '[%s%s]' % (item[0], item[0].lower()) + item[1:]
namespaces[i] = item
namespaces.append(first_lower(final_ns))
if final_ns and namespaces:
if self.site.sitename == 'wikipedia:pt' and namespace == 6:
# only change on these file extensions (per T57242)
extensions = ('png', 'gif', 'jpg', 'jpeg', 'svg', 'tiff',
'tif')
text = textlib.replaceExcept(
text, r'\[\[\s*({}) *:(?P<name>[^\|\]]*?\.({}))'
r'(?P<label>.*?)\]\]'.format('|'.join(namespaces),
'|'.join(extensions)),
r'[[{}:\g<name>\g<label>]]'.format(final_ns),
exceptions)
else:
text = textlib.replaceExcept(
text, r'\[\[\s*(%s) *:(?P<nameAndLabel>.*?)\]\]' %
'|'.join(namespaces),
r'[[%s:\g<nameAndLabel>]]' % final_ns, exceptions)
return text
def get_missing_labels(self, sitelinks, dont):
labels = {}
for dbname in sitelinks:
# [[d:Topic:Vedxkcb8ek6ss1pc]]
if dbname.startswith('alswiki'):
continue
lang = self.normalize_lang(dbname.rpartition('wik')[0])
if not lang or lang in dont:
continue
# try to defer this as much as possible
link = sitelinks[dbname]
title = link.canonical_title()
# todo: check if this is still needed
if ':' not in title and '/' in title:
continue
# [[d:Topic:Vhs5f72i5obvkr3t]]
if title.startswith('Wikipedia:Artikelwerkstatt/'):
continue
# [[d:Topic:Vw8cayiif34m2eem]]
if dbname.endswith('wikinews') and link.namespace == 14:
continue
# [[d:Topic:Vn16a76j30dblqo7]]
if dbname == 'zh_yuewiki' and title.startswith('Portal:時人時事/'):
continue
# [[d:Topic:Vrel33kwnco2xp55]]
if dbname.endswith('wikisource'):
if link.namespace == link.site.namespaces.lookup_name(
'Author'):
title = title.partition(':')[2]
# [[d:Topic:Uhdjlv9aae6iijuc]]
# todo: create a lib for this
if lang == 'fr' and title.startswith(
('Abbaye ', 'Cathédrale ', 'Chapelle ', 'Cloître ', 'Couvent ',
'Monastère ', 'Église ')):
title = first_lower(title)
label = labels.get(lang)
if label and first_lower(label) != first_lower(title):
labels.pop(lang) # todo: better handling
dont.add(lang)
continue
labels[lang] = title
return labels
def ReplaceLink(self, text, oldtxt, newtxt):
frmParts = [
s.strip(self.stripChars) for s in self.wordBreaker.split(oldtxt)
]
toParts = [
s.strip(self.stripChars) for s in self.wordBreaker.split(newtxt)
]
if len(frmParts) != len(toParts):
raise ValueError(u'Splitting parts do not match counts')
for i in xrange(0, len(frmParts)):
if len(frmParts[i]) != len(toParts[i]):
raise ValueError(u'Splitting parts do not match word length')
if len(frmParts[i]) > 0:
text = text.replace(first_lower(frmParts[i]),
first_lower(toParts[i]))
text = text.replace(first_upper(frmParts[i]),
first_upper(toParts[i]))
return text
def get_missing_labels(self, sitelinks, dont):
labels = {}
for dbname, title in sitelinks.items():
if ':' not in title and '/' in title:
continue
lang = self.normalize_lang(dbname.partition('wik')[0])
if lang and lang not in dont:
# [[d:Topic:Uhdjlv9aae6iijuc]]
# todo: create a lib for this
if lang == 'fr' and title.startswith(
('Abbaye ', 'Cathédrale ', 'Chapelle ', 'Cloître ',
'Couvent ', 'Monastère ', 'Église ')):
title = first_lower(title)
label = labels.get(lang)
if label and first_lower(label) != first_lower(title):
labels.pop(lang) # todo: better handling
dont.add(lang)
continue
labels[lang] = title
return labels
def correctcap(link, text):
"""
Capitalize link.
If text links to a page with title link uncapitalized, uncapitalize link,
otherwise capitalize it
"""
linkupper = link.title()
linklower = first_lower(linkupper)
if "[[{0!s}]]".format(linklower) in text or "[[{0!s}|".format(linklower) in text:
return linklower
else:
return linkupper
def get_missing_labels(self, sitelinks, dont):
labels = {}
for dbname, title in sitelinks.items():
if ':' not in title and '/' in title:
continue
# fixme: 'wikidata' -> ('', 'wiki', 'data')
# fixme: 'mediawikiwiki' -> ('media', 'wiki', 'wiki')
lang = self.normalize_lang(dbname.partition('wik')[0])
if lang and lang not in dont:
# [[d:Topic:Uhdjlv9aae6iijuc]]
# todo: create a lib for this
if lang == 'fr' and title.startswith(
('Abbaye ', 'Cathédrale ', 'Chapelle ', 'Cloître ',
'Couvent ', 'Monastère ', 'Église ')):
title = first_lower(title)
label = labels.get(lang)
if label and first_lower(label) != first_lower(title):
labels.pop(lang) # todo: better handling
dont.add(lang)
continue
labels[lang] = title
return labels
def from_cache(self, link):
link = link.replace('_', ' ').strip() # todo: normalize completely
if link not in self.redirects:
return False
if link not in self.cache:
page = pywikibot.Page(self.site, link)
if not page.exists():
pywikibot.warning('%s does not exist' % page.title())
self.redirects.remove(link) # fixme: both cases
return False
if not page.isRedirectPage():
pywikibot.warning('%s is not a redirect' % page.title())
self.redirects.remove(link) # fixme: both cases
return False
target = page.getRedirectTarget()
title = target.title()
if link == first_lower(link):
self.cache[link] = first_lower(title)
else:
self.cache[link] = title
return self.cache[link]
def correctcap(link, text):
"""Return the link capitalized/uncapitalized according to the text.
@param link: link page
@type link: pywikibot.Page
@param text: the wikitext that is supposed to refer to the link
@type text: str
@return: uncapitalized title of the link if the text links to the link
with an uncapitalized title, else capitalized
@rtype: str
"""
linkupper = link.title()
linklower = first_lower(linkupper)
if "[[%s]]" % linklower in text or "[[%s|" % linklower in text:
return linklower
else:
return linkupper
def correctcap(link, text):
"""Return the link capitalized/uncapitalized according to the text.
@param link: link page
@type link: pywikibot.Page
@param text: the wikitext that is supposed to refer to the link
@type text: str
@return: uncapitalized title of the link if the text links to the link
with an uncapitalized title, else capitalized
@rtype: str
"""
linkupper = link.title()
linklower = first_lower(linkupper)
if "[[%s]]" % linklower in text or "[[%s|" % linklower in text:
return linklower
else:
return linkupper
def translateAndCapitalizeNamespaces(self, text):
"""Use localized namespace names."""
# arz uses english stylish codes
if self.site.sitename == 'wikipedia:arz':
return text
family = self.site.family
# wiki links aren't parsed here.
exceptions = ['nowiki', 'comment', 'math', 'pre']
for namespace in self.site.namespaces.values():
if namespace.id in (0, 2, 3):
# skip main (article) namespace
# skip user namespace, maybe gender is used
continue
# a clone is needed. Won't change the namespace dict
namespaces = list(namespace)
thisNs = namespaces.pop(0)
if namespace.id == 6 and family.name == 'wikipedia':
if self.site.code in ('en', 'fr') and MediaWikiVersion(
self.site.version()) >= MediaWikiVersion('1.14'):
# do not change "Image" on en-wiki and fr-wiki
assert u'Image' in namespaces
namespaces.remove(u'Image')
if self.site.code == 'hu':
# do not change "Kép" on hu-wiki
assert u'Kép' in namespaces
namespaces.remove(u'Kép')
elif self.site.code == 'pt':
# TODO: bug T57242
continue
# lowerspaced and underscored namespaces
for i in range(len(namespaces)):
item = namespaces[i].replace(' ', '[ _]')
item = u'[%s%s]' % (item[0], item[0].lower()) + item[1:]
namespaces[i] = item
namespaces.append(first_lower(thisNs))
if thisNs and namespaces:
text = textlib.replaceExcept(
text,
r'\[\[\s*(%s) *:(?P<nameAndLabel>.*?)\]\]'
% '|'.join(namespaces),
r'[[%s:\g<nameAndLabel>]]' % thisNs,
exceptions)
return text
def translateAndCapitalizeNamespaces(self, text):
"""Use localized namespace names."""
# arz uses english stylish codes
if self.site.sitename == 'wikipedia:arz':
return text
family = self.site.family
# wiki links aren't parsed here.
exceptions = ['nowiki', 'comment', 'math', 'pre']
for namespace in self.site.namespaces.values():
if namespace.id in (0, 2, 3):
# skip main (article) namespace
# skip user namespace, maybe gender is used
continue
# a clone is needed. Won't change the namespace dict
namespaces = list(namespace)
thisNs = namespaces.pop(0)
if namespace.id == 6 and family.name == 'wikipedia':
if self.site.code in ('en', 'fr') and \
MediaWikiVersion(self.site.version()) >= MediaWikiVersion('1.14'):
# do not change "Image" on en-wiki and fr-wiki
assert u'Image' in namespaces
namespaces.remove(u'Image')
if self.site.code == 'hu':
# do not change "Kép" on hu-wiki
assert u'Kép' in namespaces
namespaces.remove(u'Kép')
elif self.site.code == 'pt':
# TODO: bug T57242
continue
# lowerspaced and underscored namespaces
for i in range(len(namespaces)):
item = namespaces[i].replace(' ', '[ _]')
item = u'[%s%s]' % (item[0], item[0].lower()) + item[1:]
namespaces[i] = item
namespaces.append(first_lower(thisNs))
if thisNs and namespaces:
text = textlib.replaceExcept(
text,
r'\[\[\s*(%s) *:(?P<nameAndLabel>.*?)\]\]'
% '|'.join(namespaces),
r'[[%s:\g<nameAndLabel>]]' % thisNs,
exceptions)
return text
def translateAndCapitalizeNamespaces(self, text):
"""Use localized namespace names."""
# arz uses english stylish codes
if self.site.sitename == "wikipedia:arz":
return text
family = self.site.family
# wiki links aren't parsed here.
exceptions = ["nowiki", "comment", "math", "pre"]
for namespace in self.site.namespaces.values():
if namespace.id in (0, 2, 3):
# skip main (article) namespace
# skip user namespace, maybe gender is used
continue
# a clone is needed. Won't change the namespace dict
namespaces = list(namespace)
thisNs = namespaces.pop(0)
if namespace.id == 6 and family.name == "wikipedia":
if self.site.code in ("en", "fr") and MediaWikiVersion(self.site.version()) >= MediaWikiVersion("1.14"):
# do not change "Image" on en-wiki and fr-wiki
assert "Image" in namespaces
namespaces.remove("Image")
if self.site.code == "hu":
# do not change "Kép" on hu-wiki
assert "Kép" in namespaces
namespaces.remove("Kép")
elif self.site.code == "pt":
# TODO: bug T57242
continue
# lowerspaced and underscored namespaces
for i in range(len(namespaces)):
item = namespaces[i].replace(" ", "[ _]")
item = "[%s%s]" % (item[0], item[0].lower()) + item[1:]
namespaces[i] = item
namespaces.append(first_lower(thisNs))
if thisNs and namespaces:
text = textlib.replaceExcept(
text,
r"\[\[\s*(%s) *:(?P<nameAndLabel>.*?)\]\]" % "|".join(namespaces),
r"[[%s:\g<nameAndLabel>]]" % thisNs,
exceptions,
)
return text
def fix_languages(self, data):
ret = False
for lang, norm in self.lang_map.items():
label = data['labels'].get(lang)
if not label:
continue
if norm:
if norm in data['labels']:
aliases = data['aliases'].get(norm, [])
if label not in map(first_lower, aliases):
aliases.append(label)
data['aliases'][norm] = aliases
else:
data['labels'][norm] = label
data['labels'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
description = data['descriptions'].get(lang)
if description:
if norm and norm not in data['descriptions']:
data['descriptions'][norm] = description
data['descriptions'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
old_aliases = data['aliases'].get(lang)
if old_aliases:
if norm:
new_aliases = data['aliases'].get(norm, [])
already = set(map(first_lower, new_aliases))
if norm in data['labels']:
already.add(first_lower(data['labels'][norm]))
for alias in old_aliases:
if alias not in already:
new_aliases.append(alias)
already.add(alias)
data['aliases'][norm] = new_aliases
data['aliases'][lang] = []
ret = True
return ret
def fix_languages(self, data):
ret = False
for lang, norm in self.lang_map.items():
label = data['labels'].get(lang)
if not label:
continue
if norm:
if norm in data['labels']:
aliases = data['aliases'].get(norm, [])
if label not in map(first_lower, aliases):
aliases.append(label)
data['aliases'][norm] = aliases
else:
data['labels'][norm] = label
data['labels'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
description = data['descriptions'].get(lang)
if description:
if norm and norm not in data['descriptions']:
data['descriptions'][norm] = description
data['descriptions'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
old_aliases = data['aliases'].get(lang)
if old_aliases:
if norm:
new_aliases = data['aliases'].get(norm, [])
already = set(map(first_lower, new_aliases))
if norm in data['labels']:
already.add(first_lower(data['labels'][norm]))
for alias in old_aliases:
if alias not in already:
new_aliases.append(alias)
already.add(alias)
data['aliases'][norm] = new_aliases
data['aliases'][lang] = []
ret = True
return ret
def fix_languages(self, data):
ret = False
for lang, norm in self.lang_map.items():
label = data['labels'].get(lang)
if not label:
continue
if norm:
norm_label = data['labels'].get(norm)
if norm_label:
if first_lower(norm_label) != first_lower(label):
aliases = data['aliases'].get(norm, [])
if first_lower(label) not in map(first_lower, aliases):
aliases.append(label)
data['aliases'][norm] = aliases
else:
data['labels'][norm] = label
data['labels'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
description = data['descriptions'].get(lang)
if description:
if norm and norm not in data['descriptions']:
data['descriptions'][norm] = description
data['descriptions'][lang] = ''
ret = True
for lang, norm in self.lang_map.items():
old_aliases = data['aliases'].get(lang)
if not old_aliases:
continue
if norm:
new_aliases = data['aliases'].get(norm, [])
already = set(map(first_lower, new_aliases))
norm_label = data['labels'].get(norm)
if norm_label:
already.add(first_lower(norm_label))
for alias in old_aliases:
if first_lower(alias) not in already:
new_aliases.append(alias)
already.add(first_lower(alias))
# fixme: buggy, raises not-recognized-array
data['aliases'][norm] = new_aliases
data['aliases'][lang] = []
ret = True
return ret
def treat(self, refPage, disambPage):
"""Treat a page.
@param disambPage: the disambiguation page or redirect we don't want
anything to link to
@type disambPage: pywikibot.Page
@param refPage: a page linking to disambPage
@type refPage: pywikibot.Page
@return: False if the user pressed q to completely quit the program,
True otherwise
@rtype: bool
"""
# TODO: break this function up into subroutines!
self.current_page = refPage
include = False
unlink_counter = 0
new_targets = []
try:
text = refPage.get()
ignoreReason = self.checkContents(text)
if ignoreReason:
pywikibot.output('\n\nSkipping %s because it contains %s.\n\n'
% (refPage.title(), ignoreReason))
else:
include = True
except pywikibot.IsRedirectPage:
pywikibot.output(u'%s is a redirect to %s'
% (refPage.title(), disambPage.title()))
if disambPage.isRedirectPage():
target = self.alternatives[0]
if pywikibot.input_yn(u'Do you want to make redirect %s point '
'to %s?' % (refPage.title(), target),
default=False, automatic_quit=False):
redir_text = '#%s [[%s]]' \
% (self.mysite.redirect(), target)
try:
refPage.put_async(redir_text, summary=self.comment)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
else:
choice = pywikibot.input_choice(
u'Do you want to work on pages linking to %s?'
% refPage.title(),
[('yes', 'y'), ('no', 'n'), ('change redirect', 'c')], 'n',
automatic_quit=False)
if choice == 'y':
gen = ReferringPageGeneratorWithIgnore(
refPage, self.primary, main_only=self.main_only
)
preloadingGen = pagegenerators.PreloadingGenerator(gen)
for refPage2 in preloadingGen:
# run until the user selected 'quit'
if not self.treat(refPage2, refPage):
break
elif choice == 'c':
text = refPage.get(get_redirect=True)
include = "redirect"
except pywikibot.NoPage:
pywikibot.output(
u'Page [[%s]] does not seem to exist?! Skipping.'
% refPage.title())
include = False
if include in (True, "redirect"):
# make a backup of the original text so we can show the changes later
original_text = text
n = 0
curpos = 0
dn = False
edited = False
# This loop will run until we have finished the current page
while True:
m = self.linkR.search(text, pos=curpos)
if not m:
if n == 0:
pywikibot.output(u"No changes necessary in %s"
% refPage.title())
return True
else:
# stop loop and save page
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
try:
foundlink = pywikibot.Link(m.group('title'),
disambPage.site)
foundlink.parse()
except pywikibot.Error:
continue
# ignore interwiki links
if foundlink.site != disambPage.site:
continue
# Check whether the link found is to disambPage.
try:
if foundlink.canonical_title() != disambPage.title():
continue
except pywikibot.Error:
# must be a broken link
pywikibot.log(u"Invalid link [[%s]] in page [[%s]]"
% (m.group('title'), refPage.title()))
continue
n += 1
# how many bytes should be displayed around the current link
context = 60
# check if there's a dn-template here already
if (self.dnSkip and self.dn_template_str and
self.dn_template_str[:-2] in text[m.end():m.end() +
len(self.dn_template_str) + 8]):
continue
edit = EditOption('edit page', 'e', text, m.start(), disambPage.title())
context_option = HighlightContextOption(
'more context', 'm', text, 60, start=m.start(), end=m.end())
context_option.before_question = True
options = [ListOption(self.alternatives, ''),
ListOption(self.alternatives, 'r'),
StandardOption('skip link', 's'),
edit,
StandardOption('next page', 'n'),
StandardOption('unlink', 'u')]
if self.dn_template_str:
# '?', '/' for old choice
options += [AliasOption('tag template %s' % self.dn_template_str,
['t', '?', '/'])]
options += [context_option]
if not edited:
options += [ShowPageOption('show disambiguation page', 'd',
m.start(), disambPage)]
options += [
OutputProxyOption('list', 'l',
SequenceOutputter(self.alternatives)),
AddAlternativeOption('add new', 'a',
SequenceOutputter(self.alternatives))]
if edited:
options += [StandardOption('save in this form', 'x')]
# TODO: Output context on each question
answer = pywikibot.input_choice('Option', options,
default=self.always,
force=bool(self.always))
if answer == 'x':
assert edited, 'invalid option before editing'
break
elif answer == 's':
n -= 1 # TODO what's this for?
continue
elif answer == 'e':
text = edit.new_text
edited = True
curpos = 0
continue
elif answer == 'n':
# skip this page
if self.primary:
# If run with the -primary argument, skip this
# occurrence next time.
self.primaryIgnoreManager.ignore(refPage)
return True
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
if answer == 't':
assert self.dn_template_str
# small chunk of text to search
search_text = text[m.end():m.end() + context]
# figure out where the link (and sentance) ends, put note
# there
end_of_word_match = re.search(r'\s', search_text)
if end_of_word_match:
position_split = end_of_word_match.start(0)
else:
position_split = 0
# insert dab needed template
text = (text[:m.end() + position_split] +
self.dn_template_str +
text[m.end() + position_split:])
dn = True
continue
elif answer == 'u':
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
unlink_counter += 1
continue
else:
# Check that no option from above was missed
assert isinstance(answer, tuple), 'only tuple answer left.'
assert answer[0] in ['r', ''], 'only valid tuple answers.'
if answer[0] == 'r':
# we want to throw away the original link text
replaceit = link_text == page_title
elif include == "redirect":
replaceit = True
else:
replaceit = False
new_page_title = answer[1]
repPl = pywikibot.Page(pywikibot.Link(new_page_title,
disambPage.site))
if (new_page_title[0].isupper() or
link_text[0].isupper()):
new_page_title = repPl.title()
else:
new_page_title = repPl.title()
new_page_title = first_lower(new_page_title)
if new_page_title not in new_targets:
new_targets.append(new_page_title)
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title,
section,
trailing_chars)
elif replaceit or (new_page_title == link_text and
not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters
# instead of a pipelink
elif (
(len(new_page_title) <= len(link_text)) and
(firstcap(link_text[:len(new_page_title)]) == firstcap(new_page_title)) and
(re.sub(self.trailR, '', link_text[len(new_page_title):]) == '') and
(not section)
):
newlink = "[[%s]]%s" \
% (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" \
% (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
pywikibot.output(text[max(0, m.start() - 30):m.end() + 30])
if text == original_text:
pywikibot.output(u'\nNo changes have been made:\n')
else:
pywikibot.output(u'\nThe following changes have been made:\n')
pywikibot.showDiff(original_text, text)
pywikibot.output(u'')
# save the page
self.setSummaryMessage(disambPage, new_targets, unlink_counter,
dn)
try:
refPage.put_async(text, summary=self.comment)
except pywikibot.LockedPage:
pywikibot.output(u'Page not saved: page is locked')
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
return True
def treat(text, linkedPage, targetPage):
"""Based on the method of the same name in solve_disambiguation.py."""
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
# make a backup of the original text so we can show the changes later
linkR = re.compile(r'\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?'
r'(\|(?P<label>[^\]]*))?\]\](?P<linktrail>' +
linktrail + ')')
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group('title').strip() == '' or \
mysite.isInterwikiLink(m.group('title')):
continue
else:
actualLinkPage = pywikibot.Page(targetPage.site, m.group('title'))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
choice = 'y'
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
if choice in "uU":
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
continue
replaceit = choice in "rR"
# remove preleading ":"
if link_text[0] == ':':
link_text = link_text[1:]
if link_text[0].isupper():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
# remove preleading ":"
if new_page_title[0] == ':':
new_page_title = new_page_title[1:]
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section, trailing_chars)
elif replaceit or (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif len(new_page_title) <= len(link_text) and \
firstcap(link_text[:len(new_page_title)]) == \
firstcap(new_page_title) and \
re.sub(re.compile(linktrail), '', link_text[len(new_page_title):]) == '' and not section:
newlink = "[[%s]]%s" % (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
return text
genFactory = pagegenerators.GeneratorFactory(site=site)
for ns in (0, 14, 100):
if family != 'wikisource' and ns == 100: # fixme: cswikiquote
continue
if family == 'wikisource' and ns == 0:
continue
genFactory.handleArg('-ns:%i' % ns)
genFactory.handleArg('-unconnectedpages')
generator = genFactory.getCombinedGenerator(preload=True)
for page in generator:
if page.namespace() != 14 and page.isDisambig():
continue
for template, fields in textlib.extract_templates_and_params(page.text):
if first_lower(template) not in tp_map[project]:
continue
params = tp_map[project][first_lower(template)]
for key in fields:
if key not in params:
continue
title = fields[key].strip()
if not title:
continue
target_lang = lang
target_family = family
if isinstance(params[key], dict):
if params[key].get('namespaces', []) and page.namespace() not in params[key]['namespaces']:
def treat(self, refPage, disambPage):
"""
Treat a page.
Parameters:
disambPage - The disambiguation page or redirect we don't want
anything to link to
refPage - A page linking to disambPage
Returns False if the user pressed q to completely quit the program.
Otherwise, returns True.
"""
# TODO: break this function up into subroutines!
include = False
unlink_counter = 0
new_targets = []
try:
text = refPage.get()
ignoreReason = self.checkContents(text)
if ignoreReason:
pywikibot.output('\n\nSkipping %s because it contains %s.\n\n'
% (refPage.title(), ignoreReason))
else:
include = True
except pywikibot.IsRedirectPage:
pywikibot.output(u'%s is a redirect to %s'
% (refPage.title(), disambPage.title()))
if disambPage.isRedirectPage():
target = self.alternatives[0]
if pywikibot.input_yn(u'Do you want to make redirect %s point '
'to %s?' % (refPage.title(), target),
default=False, automatic_quit=False):
redir_text = '#%s [[%s]]' \
% (self.mysite.redirect(), target)
try:
refPage.put_async(redir_text, summary=self.comment)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
else:
choice = pywikibot.input_choice(
u'Do you want to work on pages linking to %s?'
% refPage.title(),
[('yes', 'y'), ('no', 'n'), ('change redirect', 'c')], 'n',
automatic_quit=False)
if choice == 'y':
gen = ReferringPageGeneratorWithIgnore(refPage,
self.primary)
preloadingGen = pagegenerators.PreloadingGenerator(gen)
for refPage2 in preloadingGen:
# run until the user selected 'quit'
if not self.treat(refPage2, refPage):
break
elif choice == 'c':
text = refPage.get(get_redirect=True)
include = "redirect"
except pywikibot.NoPage:
pywikibot.output(
u'Page [[%s]] does not seem to exist?! Skipping.'
% refPage.title())
include = False
if include in (True, "redirect"):
# make a backup of the original text so we can show the changes later
original_text = text
n = 0
curpos = 0
dn = False
edited = False
# This loop will run until we have finished the current page
while True:
m = self.linkR.search(text, pos=curpos)
if not m:
if n == 0:
pywikibot.output(u"No changes necessary in %s"
% refPage.title())
return True
else:
# stop loop and save page
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
try:
foundlink = pywikibot.Link(m.group('title'),
disambPage.site)
foundlink.parse()
except pywikibot.Error:
continue
# ignore interwiki links
if foundlink.site != disambPage.site:
continue
# Check whether the link found is to disambPage.
try:
if foundlink.canonical_title() != disambPage.title():
continue
except pywikibot.Error:
# must be a broken link
pywikibot.log(u"Invalid link [[%s]] in page [[%s]]"
% (m.group('title'), refPage.title()))
continue
n += 1
# how many bytes should be displayed around the current link
context = 60
# check if there's a dn-template here already
if (self.dnSkip and self.dn_template_str and
self.dn_template_str[:-2] in text[m.end():m.end() +
len(self.dn_template_str) + 8]):
continue
# This loop will run while the user doesn't choose an option
# that will actually change the page
while True:
self.current_page = refPage
if not self.always:
# at the beginning of the link, start red color.
# at the end of the link, reset the color to default
pywikibot.output(
text[max(0, m.start() - context):m.start()] +
'\03{lightred}' + text[m.start():m.end()] +
'\03{default}' + text[m.end():m.end() + context])
options = ['#', 'r#', '[s]kip link', '[e]dit page',
'[n]ext page', '[u]nlink', '[q]uit']
if self.dn_template_str:
options.append(u'[t]ag template %s' % self.dn_template_str)
options.append('[m]ore context')
if not edited:
options.append('show [d]isambiguation page')
options += ['[l]ist', '[a]dd new']
if edited:
options += ['save in this form [x]']
options = concat_options('Option', 72, options)
choice = pywikibot.input(options)
else:
choice = self.always
if choice in ['a', 'A']:
newAlternative = pywikibot.input(u'New alternative:')
self.alternatives.append(newAlternative)
self.listAlternatives()
elif choice in ['e', 'E']:
editor = editarticle.TextEditor()
newText = editor.edit(text, jumpIndex=m.start(),
highlight=disambPage.title())
# if user didn't press Cancel
if newText and newText != text:
text = newText
break
elif choice in ['d', 'D']:
editor = editarticle.TextEditor()
if disambPage.isRedirectPage():
disambredir = disambPage.getRedirectTarget()
editor.edit(
disambredir.get(),
jumpIndex=m.start(),
highlight=disambredir.title())
else:
editor.edit(
disambPage.get(),
jumpIndex=m.start(),
highlight=disambPage.title())
elif choice in ['l', 'L']:
self.listAlternatives()
elif choice in ['m', 'M']:
# show more text around the link we're working on
context *= 2
else:
break
if choice in ['e', 'E']:
# user has edited the page and then pressed 'OK'
edited = True
curpos = 0
continue
elif choice in ['n', 'N']:
# skip this page
if self.primary:
# If run with the -primary argument, skip this
# occurrence next time.
self.primaryIgnoreManager.ignore(refPage)
return True
elif choice in ['q', 'Q']:
# quit the program
self.quit()
elif choice in ['s', 'S']:
# Next link on this page
n -= 1
continue
elif choice in ['x', 'X'] and edited:
# Save the page as is
break
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
# '?', '/' for old choice
if choice in ['t', 'T', '?', '/'] and self.dn_template_str:
# small chunk of text to search
search_text = text[m.end():m.end() + context]
# figure out where the link (and sentance) ends, put note
# there
end_of_word_match = re.search(r'\s', search_text)
if end_of_word_match:
position_split = end_of_word_match.start(0)
else:
position_split = 0
# insert dab needed template
text = (text[:m.end() + position_split] +
self.dn_template_str +
text[m.end() + position_split:])
dn = True
continue
elif choice in ['u', 'U']:
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
unlink_counter += 1
continue
else:
if len(choice) > 0 and choice[0] == 'r':
# we want to throw away the original link text
replaceit = link_text == page_title
choice = choice[1:]
elif include == "redirect":
replaceit = True
else:
replaceit = False
try:
choice = int(choice)
except ValueError:
pywikibot.output(u"Unknown option")
# step back to ask the user again what to do with the
# current link
curpos -= 1
continue
if choice >= len(self.alternatives) or choice < 0:
pywikibot.output(
u"Choice out of range. Please select a number "
u"between 0 and %i." % (len(self.alternatives) - 1))
# show list of possible choices
self.listAlternatives()
# step back to ask the user again what to do with the
# current link
curpos -= 1
continue
new_page_title = self.alternatives[choice]
repPl = pywikibot.Page(pywikibot.Link(new_page_title,
disambPage.site))
if (new_page_title[0].isupper() or
link_text[0].isupper()):
new_page_title = repPl.title()
else:
new_page_title = repPl.title()
new_page_title = first_lower(new_page_title)
if new_page_title not in new_targets:
new_targets.append(new_page_title)
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title,
section,
trailing_chars)
elif replaceit or (new_page_title == link_text and
not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters
# instead of a pipelink
elif (
(len(new_page_title) <= len(link_text)) and
(firstcap(link_text[:len(new_page_title)]) == firstcap(new_page_title)) and
(re.sub(self.trailR, '', link_text[len(new_page_title):]) == '') and
(not section)
):
newlink = "[[%s]]%s" \
% (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" \
% (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
pywikibot.output(text[max(0, m.start() - 30):m.end() + 30])
if text == original_text:
pywikibot.output(u'\nNo changes have been made:\n')
else:
pywikibot.output(u'\nThe following changes have been made:\n')
pywikibot.showDiff(original_text, text)
pywikibot.output(u'')
# save the page
self.setSummaryMessage(disambPage, new_targets, unlink_counter,
dn)
try:
refPage.put_async(text, summary=self.comment)
except pywikibot.LockedPage:
pywikibot.output(u'Page not saved: page is locked')
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
return True
def treat(self, refPage, disambPage):
"""
Treat a page.
Parameters:
disambPage - The disambiguation page or redirect we don't want
anything to link to
refPage - A page linking to disambPage
Returns False if the user pressed q to completely quit the program.
Otherwise, returns True.
"""
# TODO: break this function up into subroutines!
include = False
unlink_counter = 0
new_targets = []
try:
text = refPage.get()
ignoreReason = self.checkContents(text)
if ignoreReason:
pywikibot.output(
'\n\nSkipping %s because it contains %s.\n\n' %
(refPage.title(), ignoreReason))
else:
include = True
except pywikibot.IsRedirectPage:
pywikibot.output(u'%s is a redirect to %s' %
(refPage.title(), disambPage.title()))
if disambPage.isRedirectPage():
target = self.alternatives[0]
if pywikibot.input_yn(u'Do you want to make redirect %s point '
'to %s?' % (refPage.title(), target),
default=False,
automatic_quit=False):
redir_text = '#%s [[%s]]' \
% (self.mysite.redirect(), target)
try:
refPage.put_async(redir_text, summary=self.comment)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
else:
choice = pywikibot.input_choice(
u'Do you want to work on pages linking to %s?' %
refPage.title(), [('yes', 'y'), ('no', 'n'),
('change redirect', 'c')],
'n',
automatic_quit=False)
if choice == 'y':
gen = ReferringPageGeneratorWithIgnore(
refPage, self.primary)
preloadingGen = pagegenerators.PreloadingGenerator(gen)
for refPage2 in preloadingGen:
# run until the user selected 'quit'
if not self.treat(refPage2, refPage):
break
elif choice == 'c':
text = refPage.get(get_redirect=True)
include = "redirect"
except pywikibot.NoPage:
pywikibot.output(
u'Page [[%s]] does not seem to exist?! Skipping.' %
refPage.title())
include = False
if include in (True, "redirect"):
# make a backup of the original text so we can show the changes later
original_text = text
n = 0
curpos = 0
dn = False
edited = False
# This loop will run until we have finished the current page
while True:
m = self.linkR.search(text, pos=curpos)
if not m:
if n == 0:
pywikibot.output(u"No changes necessary in %s" %
refPage.title())
return True
else:
# stop loop and save page
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
try:
foundlink = pywikibot.Link(m.group('title'),
disambPage.site)
foundlink.parse()
except pywikibot.Error:
continue
# ignore interwiki links
if foundlink.site != disambPage.site:
continue
# Check whether the link found is to disambPage.
try:
if foundlink.canonical_title() != disambPage.title():
continue
except pywikibot.Error:
# must be a broken link
pywikibot.log(u"Invalid link [[%s]] in page [[%s]]" %
(m.group('title'), refPage.title()))
continue
n += 1
# how many bytes should be displayed around the current link
context = 60
# check if there's a dn-template here already
if (self.dnSkip and self.dn_template_str
and self.dn_template_str[:-2]
in text[m.end():m.end() + len(self.dn_template_str) +
8]):
continue
# This loop will run while the user doesn't choose an option
# that will actually change the page
while True:
self.current_page = refPage
if not self.always:
# at the beginning of the link, start red color.
# at the end of the link, reset the color to default
pywikibot.output(text[max(0,
m.start() -
context):m.start()] +
'\03{lightred}' +
text[m.start():m.end()] +
'\03{default}' +
text[m.end():m.end() + context])
options = [
'#', 'r#', '[s]kip link', '[e]dit page',
'[n]ext page', '[u]nlink', '[q]uit'
]
if self.dn_template_str:
options.append(u'[t]ag template %s' %
self.dn_template_str)
options.append('[m]ore context')
if not edited:
options.append('show [d]isambiguation page')
options += ['[l]ist', '[a]dd new']
if edited:
options += ['save in this form [x]']
options = concat_options('Option', 72, options)
choice = pywikibot.input(options)
else:
choice = self.always
if choice in ['a', 'A']:
newAlternative = pywikibot.input(u'New alternative:')
self.alternatives.append(newAlternative)
self.listAlternatives()
elif choice in ['e', 'E']:
editor = editarticle.TextEditor()
newText = editor.edit(text,
jumpIndex=m.start(),
highlight=disambPage.title())
# if user didn't press Cancel
if newText and newText != text:
text = newText
break
elif choice in ['d', 'D']:
editor = editarticle.TextEditor()
if disambPage.isRedirectPage():
disambredir = disambPage.getRedirectTarget()
editor.edit(disambredir.get(),
jumpIndex=m.start(),
highlight=disambredir.title())
else:
editor.edit(disambPage.get(),
jumpIndex=m.start(),
highlight=disambPage.title())
elif choice in ['l', 'L']:
self.listAlternatives()
elif choice in ['m', 'M']:
# show more text around the link we're working on
context *= 2
else:
break
if choice in ['e', 'E']:
# user has edited the page and then pressed 'OK'
edited = True
curpos = 0
continue
elif choice in ['n', 'N']:
# skip this page
if self.primary:
# If run with the -primary argument, skip this
# occurrence next time.
self.primaryIgnoreManager.ignore(refPage)
return True
elif choice in ['q', 'Q']:
# quit the program
self.quit()
elif choice in ['s', 'S']:
# Next link on this page
n -= 1
continue
elif choice in ['x', 'X'] and edited:
# Save the page as is
break
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
# '?', '/' for old choice
if choice in ['t', 'T', '?', '/'] and self.dn_template_str:
# small chunk of text to search
search_text = text[m.end():m.end() + context]
# figure out where the link (and sentance) ends, put note
# there
end_of_word_match = re.search(r'\s', search_text)
if end_of_word_match:
position_split = end_of_word_match.start(0)
else:
position_split = 0
# insert dab needed template
text = (text[:m.end() + position_split] +
self.dn_template_str +
text[m.end() + position_split:])
dn = True
continue
elif choice in ['u', 'U']:
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
unlink_counter += 1
continue
else:
if len(choice) > 0 and choice[0] == 'r':
# we want to throw away the original link text
replaceit = link_text == page_title
choice = choice[1:]
elif include == "redirect":
replaceit = True
else:
replaceit = False
try:
choice = int(choice)
except ValueError:
pywikibot.output(u"Unknown option")
# step back to ask the user again what to do with the
# current link
curpos -= 1
continue
if choice >= len(self.alternatives) or choice < 0:
pywikibot.output(
u"Choice out of range. Please select a number "
u"between 0 and %i." %
(len(self.alternatives) - 1))
# show list of possible choices
self.listAlternatives()
# step back to ask the user again what to do with the
# current link
curpos -= 1
continue
new_page_title = self.alternatives[choice]
repPl = pywikibot.Page(
pywikibot.Link(new_page_title, disambPage.site))
if (new_page_title[0].isupper() or link_text[0].isupper()):
new_page_title = repPl.title()
else:
new_page_title = repPl.title()
new_page_title = first_lower(new_page_title)
if new_page_title not in new_targets:
new_targets.append(new_page_title)
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section,
trailing_chars)
elif replaceit or (new_page_title == link_text
and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters
# instead of a pipelink
elif ((len(new_page_title) <= len(link_text))
and (firstcap(link_text[:len(new_page_title)])
== firstcap(new_page_title))
and (re.sub(self.trailR, '',
link_text[len(new_page_title):]) == '')
and (not section)):
newlink = "[[%s]]%s" \
% (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" \
% (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
pywikibot.output(text[max(0, m.start() - 30):m.end() + 30])
if text == original_text:
pywikibot.output(u'\nNo changes have been made:\n')
else:
pywikibot.output(u'\nThe following changes have been made:\n')
pywikibot.showDiff(original_text, text)
pywikibot.output(u'')
# save the page
self.setSummaryMessage(disambPage, new_targets, unlink_counter,
dn)
try:
refPage.put_async(text, summary=self.comment)
except pywikibot.LockedPage:
pywikibot.output(u'Page not saved: page is locked')
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
return True
def treat(text, linkedPage, targetPage):
"""Based on the method of the same name in solve_disambiguation.py."""
# make a backup of the original text so we can show the changes later
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
linkR = re.compile(
r'\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?(\|(?P<label>[^\]]*))?\]\](?P<linktrail>%s)'
% linktrail)
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group('title') == '' or mysite.isInterwikiLink(m.group('title')):
continue
else:
actualLinkPage = pywikibot.Page(mysite, m.group('title'))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
# how many bytes should be displayed around the current link
context = 30
# at the beginning of the link, start red color.
# at the end of the link, reset the color to default
pywikibot.output(text[max(0,
m.start() - context):m.start()] +
'\03{lightred}' + text[m.start():m.end()] +
'\03{default}' + text[m.end():m.end() + context])
choice = pywikibot.input_choice(
'What should be done with the link?',
(('Do not change', 'n'),
('Change link to \03{lightpurple}%s\03{default}' %
targetPage.title(), 'y'), ('Change and replace text', 'r'),
('Unlink', 'u')),
default='n',
automatic_quit=False)
if choice == 'n':
continue
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
if choice == 'u':
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
continue
if link_text[0].isupper():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
if choice == 'r' and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section, trailing_chars)
elif choice == 'r' or (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif len(new_page_title) <= len(link_text) and \
firstcap(link_text[:len(new_page_title)]) == \
firstcap(new_page_title) and \
re.sub(re.compile(linktrail), '', link_text[len(new_page_title):]) == '' and not section:
newlink = "[[%s]]%s" % (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
return text
def replace_links(self, text, linkedPage, targetPage):
"""Replace all source links by target."""
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
# make a backup of the original text so we can show the changes later
linkR = re.compile(r'\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?'
r'(\|(?P<label>[^\]]*))?\]\](?P<linktrail>' + linktrail + ')')
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group('title').strip() == '' or \
mysite.isInterwikiLink(m.group('title')):
continue
else:
actualLinkPage = pywikibot.Page(targetPage.site, m.group('title'))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
if section and targetPage.section():
pywikibot.warning(
'Source section {0} and target section {1} found. '
'Skipping.'.format(section, targetPage))
continue
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
# remove preleading ":"
if link_text[0] == ':':
link_text = link_text[1:]
if link_text[0].isupper() or link_text[0].isdigit():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
# remove preleading ":"
if new_page_title[0] == ':':
new_page_title = new_page_title[1:]
if (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif (len(new_page_title) <= len(link_text) and
firstcap(link_text[:len(new_page_title)]) ==
firstcap(new_page_title) and
re.sub(re.compile(linktrail), '',
link_text[len(new_page_title):]) == '' and
not section):
newlink = "[[%s]]%s" % (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
return text
def handleOneLink(match):
titleWithSection = match.group('titleWithSection')
label = match.group('label')
trailingChars = match.group('linktrail')
newline = match.group('newline')
try:
is_interwiki = self.site.isInterwikiLink(titleWithSection)
except ValueError: # T111513
is_interwiki = True
if not is_interwiki:
# The link looks like this:
# [[page_title|link_text]]trailing_chars
# We only work on namespace 0 because pipes and linktrails work
# differently for images and categories.
page = pywikibot.Page(
pywikibot.Link(titleWithSection, self.site))
try:
namespace = page.namespace()
except pywikibot.InvalidTitle:
return match.group()
if namespace == 0:
# Replace underlines by spaces, also multiple underlines
titleWithSection = re.sub('_+', ' ', titleWithSection)
# Remove double spaces
titleWithSection = re.sub(' +', ' ', titleWithSection)
# Remove unnecessary leading spaces from title,
# but remember if we did this because we eventually want
# to re-add it outside of the link later.
titleLength = len(titleWithSection)
titleWithSection = titleWithSection.lstrip()
hadLeadingSpaces = (len(titleWithSection) != titleLength)
hadTrailingSpaces = False
# Remove unnecessary trailing spaces from title,
# but remember if we did this because it may affect
# the linktrail and because we eventually want to
# re-add it outside of the link later.
if not trailingChars:
titleLength = len(titleWithSection)
titleWithSection = titleWithSection.rstrip()
hadTrailingSpaces = (len(titleWithSection) !=
titleLength)
# Convert URL-encoded characters to unicode
from pywikibot.page import url2unicode
titleWithSection = url2unicode(titleWithSection,
encodings=self.site)
if titleWithSection == '':
# just skip empty links.
return match.group()
# Remove unnecessary initial and final spaces from label.
# Please note that some editors prefer spaces around pipes.
# (See [[en:Wikipedia:Semi-bots]]). We remove them anyway.
if label is not None:
# Remove unnecessary leading spaces from label,
# but remember if we did this because we want
# to re-add it outside of the link later.
labelLength = len(label)
label = label.lstrip()
hadLeadingSpaces = (len(label) != labelLength)
# Remove unnecessary trailing spaces from label,
# but remember if we did this because it affects
# the linktrail.
if not trailingChars:
labelLength = len(label)
label = label.rstrip()
hadTrailingSpaces = (len(label) != labelLength)
else:
label = titleWithSection
if trailingChars:
label += trailingChars
if self.site.siteinfo['case'] == 'first-letter':
firstcase_title = first_lower(titleWithSection)
firstcase_label = first_lower(label)
else:
firstcase_title = titleWithSection
firstcase_label = label
if firstcase_label == firstcase_title:
newLink = '[[%s]]' % label
# Check if we can create a link with trailing characters
# instead of a pipelink
elif (firstcase_label.startswith(firstcase_title) and
trailR.sub('', label[len(titleWithSection):]) == ''):
newLink = '[[%s]]%s' % (label[:len(titleWithSection)],
label[len(titleWithSection):])
else:
# Try to capitalize the first letter of the title.
# Not useful for languages that don't capitalize nouns.
# TODO: Add a configuration variable for each site,
# which determines if the link target is written in
# uppercase
if self.site.sitename == 'wikipedia:de':
titleWithSection = first_upper(titleWithSection)
newLink = "[[%s|%s]]" % (titleWithSection, label)
# re-add spaces that were pulled out of the link.
# Examples:
# text[[ title ]]text -> text [[title]] text
# text[[ title | name ]]text -> text [[title|name]] text
# text[[ title |name]]text -> text[[title|name]]text
# text[[title| name]]text -> text [[title|name]]text
if hadLeadingSpaces and not newline:
newLink = ' ' + newLink
if hadTrailingSpaces:
newLink = newLink + ' '
if newline:
newLink = newline + newLink
return newLink
# don't change anything
return match.group()
def handleOneLink(match):
titleWithSection = match.group('titleWithSection')
label = match.group('label')
trailingChars = match.group('linktrail')
newline = match.group('newline')
if not self.site.isInterwikiLink(titleWithSection):
# The link looks like this:
# [[page_title|link_text]]trailing_chars
# We only work on namespace 0 because pipes and linktrails work
# differently for images and categories.
page = pywikibot.Page(pywikibot.Link(titleWithSection,
self.site))
try:
namespace = page.namespace()
except pywikibot.InvalidTitle:
return match.group()
if namespace == 0:
# Replace underlines by spaces, also multiple underlines
titleWithSection = re.sub('_+', ' ', titleWithSection)
# Remove double spaces
titleWithSection = re.sub(' +', ' ', titleWithSection)
# Remove unnecessary leading spaces from title,
# but remember if we did this because we eventually want
# to re-add it outside of the link later.
titleLength = len(titleWithSection)
titleWithSection = titleWithSection.lstrip()
hadLeadingSpaces = (len(titleWithSection) != titleLength)
hadTrailingSpaces = False
# Remove unnecessary trailing spaces from title,
# but remember if we did this because it may affect
# the linktrail and because we eventually want to
# re-add it outside of the link later.
if not trailingChars:
titleLength = len(titleWithSection)
titleWithSection = titleWithSection.rstrip()
hadTrailingSpaces = (len(titleWithSection) !=
titleLength)
# Convert URL-encoded characters to unicode
from pywikibot.page import url2unicode
titleWithSection = url2unicode(titleWithSection,
encodings=self.site)
if titleWithSection == '':
# just skip empty links.
return match.group()
# Remove unnecessary initial and final spaces from label.
# Please note that some editors prefer spaces around pipes.
# (See [[en:Wikipedia:Semi-bots]]). We remove them anyway.
if label is not None:
# Remove unnecessary leading spaces from label,
# but remember if we did this because we want
# to re-add it outside of the link later.
labelLength = len(label)
label = label.lstrip()
hadLeadingSpaces = (len(label) != labelLength)
# Remove unnecessary trailing spaces from label,
# but remember if we did this because it affects
# the linktrail.
if not trailingChars:
labelLength = len(label)
label = label.rstrip()
hadTrailingSpaces = (len(label) != labelLength)
else:
label = titleWithSection
if trailingChars:
label += trailingChars
if titleWithSection == label or \
first_lower(titleWithSection) == label:
newLink = "[[%s]]" % label
# Check if we can create a link with trailing characters
# instead of a pipelink
elif (len(titleWithSection) <= len(label) and
label[:len(titleWithSection)] == titleWithSection and
re.sub(trailR, '',
label[len(titleWithSection):]) == ''):
newLink = "[[%s]]%s" % (label[:len(titleWithSection)],
label[len(titleWithSection):])
else:
# Try to capitalize the first letter of the title.
# Not useful for languages that don't capitalize nouns.
# TODO: Add a configuration variable for each site,
# which determines if the link target is written in
# uppercase
if self.site.sitename == 'wikipedia:de':
titleWithSection = first_upper(titleWithSection)
newLink = "[[%s|%s]]" % (titleWithSection, label)
# re-add spaces that were pulled out of the link.
# Examples:
# text[[ title ]]text -> text [[title]] text
# text[[ title | name ]]text -> text [[title|name]] text
# text[[ title |name]]text -> text[[title|name]]text
# text[[title| name]]text -> text [[title|name]]text
if hadLeadingSpaces and not newline:
newLink = ' ' + newLink
if hadTrailingSpaces:
newLink = newLink + ' '
if newline:
newLink = newline + newLink
return newLink
# don't change anything
return match.group()
genFactory = pagegenerators.GeneratorFactory(site=site)
for ns in (0, 14, 100):
if family != 'wikisource' and ns == 100: # fixme: cswikiquote
continue
if family == 'wikisource' and ns == 0:
continue
genFactory.handleArg('-ns:%i' % ns)
genFactory.handleArg('-unconnectedpages')
generator = genFactory.getCombinedGenerator(preload=True)
for page in generator:
if page.namespace() != 14 and page.isDisambig():
continue
for template, fields in textlib.extract_templates_and_params(page.text):
if first_lower(template) not in tp_map[project].keys():
continue
params = tp_map[project][first_lower(template)]
for key in fields.keys():
if key not in params.keys():
continue
title = fields[key].strip()
if not title:
continue
target_lang = lang
target_family = family
if isinstance(params[key], dict):
if params[key].get('namespaces', []) and page.namespace() not in params[key]['namespaces']:
def treat_disamb_only(self, refPage, disambPage):
"""Resolve the links to disambPage but don't look for its redirects.
@param disambPage: the disambiguation page or redirect we don't want
anything to link to
@type disambPage: pywikibot.Page
@param refPage: a page linking to disambPage
@type refPage: pywikibot.Page
@return: "nextpage" if the user enters "n" to skip this page,
"nochange" if the page needs no change, and
"done" if the page is processed successfully
@rtype: str
"""
# TODO: break this function up into subroutines!
self.current_page = refPage
include = False
unlink_counter = 0
new_targets = []
try:
text = refPage.get()
ignoreReason = self.checkContents(text)
if ignoreReason:
pywikibot.output(
'\n\nSkipping %s because it contains %s.\n\n' %
(refPage.title(), ignoreReason))
else:
include = True
except pywikibot.IsRedirectPage:
pywikibot.output(u'%s is a redirect to %s' %
(refPage.title(), disambPage.title()))
if disambPage.isRedirectPage():
target = self.alternatives[0]
if pywikibot.input_yn(u'Do you want to make redirect %s point '
'to %s?' % (refPage.title(), target),
default=False,
automatic_quit=False):
redir_text = '#%s [[%s]]' \
% (self.mysite.redirect(), target)
try:
refPage.put(redir_text,
summary=self.comment,
asynchronous=True)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
else:
choice = pywikibot.input_choice(
u'Do you want to work on pages linking to %s?' %
refPage.title(), [('yes', 'y'), ('no', 'n'),
('change redirect', 'c')],
'n',
automatic_quit=False)
if choice == 'y':
gen = ReferringPageGeneratorWithIgnore(
refPage, self.primary, main_only=self.main_only)
preloadingGen = pagegenerators.PreloadingGenerator(gen)
for refPage2 in preloadingGen:
# run until the user selected 'quit'
self.treat(refPage2, refPage)
elif choice == 'c':
text = refPage.get(get_redirect=True)
include = "redirect"
except pywikibot.NoPage:
pywikibot.output(
u'Page [[%s]] does not seem to exist?! Skipping.' %
refPage.title())
include = False
if include in (True, "redirect"):
# save the original text so we can show the changes later
original_text = text
n = 0
curpos = 0
dn = False
edited = False
# This loop will run until we have finished the current page
while True:
m = self.linkR.search(text, pos=curpos)
if not m:
if n == 0:
# No changes necessary for this disambiguation title.
return 'nochange'
else:
# stop loop and save page
break
# Ensure that next time around we will not find this same hit.
curpos = m.start() + 1
try:
foundlink = pywikibot.Link(m.group('title'),
disambPage.site)
foundlink.parse()
except pywikibot.Error:
continue
# ignore interwiki links
if foundlink.site != disambPage.site:
continue
# Check whether the link found is to disambPage.
try:
if foundlink.canonical_title() != disambPage.title():
continue
except pywikibot.Error:
# must be a broken link
pywikibot.log(u"Invalid link [[%s]] in page [[%s]]" %
(m.group('title'), refPage.title()))
continue
n += 1
# how many bytes should be displayed around the current link
context = 60
# check if there's a dn-template here already
if (self.dnSkip and self.dn_template_str
and self.dn_template_str[:-2]
in text[m.end():m.end() + len(self.dn_template_str) +
8]):
continue
edit = EditOption('edit page', 'e', text, m.start(),
disambPage.title())
context_option = HighlightContextOption('more context',
'm',
text,
60,
start=m.start(),
end=m.end())
context_option.before_question = True
options = [
ListOption(self.alternatives, ''),
ListOption(self.alternatives, 'r'),
StandardOption('skip link', 's'), edit,
StandardOption('next page', 'n'),
StandardOption('unlink', 'u')
]
if self.dn_template_str:
# '?', '/' for old choice
options += [
AliasOption('tag template %s' % self.dn_template_str,
['t', '?', '/'])
]
options += [context_option]
if not edited:
options += [
ShowPageOption('show disambiguation page', 'd',
m.start(), disambPage)
]
options += [
OutputProxyOption('list', 'l',
SequenceOutputter(self.alternatives)),
AddAlternativeOption('add new', 'a',
SequenceOutputter(self.alternatives))
]
if edited:
options += [StandardOption('save in this form', 'x')]
# TODO: Output context on each question
answer = pywikibot.input_choice('Option',
options,
default=self.always,
force=bool(self.always))
if answer == 'x':
assert edited, 'invalid option before editing'
break
elif answer == 's':
n -= 1 # TODO what's this for?
continue
elif answer == 'e':
text = edit.new_text
edited = True
curpos = 0
continue
elif answer == 'n':
# skip this page
if self.primary:
# If run with the -primary argument, skip this
# occurrence next time.
self.primaryIgnoreManager.ignore(refPage)
return 'nextpage'
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
if answer == 't':
assert self.dn_template_str
# small chunk of text to search
search_text = text[m.end():m.end() + context]
# figure out where the link (and sentance) ends, put note
# there
end_of_word_match = re.search(r'\s', search_text)
if end_of_word_match:
position_split = end_of_word_match.start(0)
else:
position_split = 0
# insert dab needed template
text = (text[:m.end() + position_split] +
self.dn_template_str +
text[m.end() + position_split:])
dn = True
continue
elif answer == 'u':
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
unlink_counter += 1
continue
else:
# Check that no option from above was missed
assert isinstance(answer, tuple), 'only tuple answer left.'
assert answer[0] in ['r', ''], 'only valid tuple answers.'
if answer[0] == 'r':
# we want to throw away the original link text
replaceit = link_text == page_title
elif include == "redirect":
replaceit = True
else:
replaceit = False
new_page_title = answer[1]
repPl = pywikibot.Page(
pywikibot.Link(new_page_title, disambPage.site))
if (new_page_title[0].isupper() or link_text[0].isupper()):
new_page_title = repPl.title()
else:
new_page_title = repPl.title()
new_page_title = first_lower(new_page_title)
if new_page_title not in new_targets:
new_targets.append(new_page_title)
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section,
trailing_chars)
elif replaceit or (new_page_title == link_text
and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters
# instead of a pipelink
elif ((len(new_page_title) <= len(link_text))
and (firstcap(link_text[:len(new_page_title)])
== firstcap(new_page_title))
and (re.sub(self.trailR, '',
link_text[len(new_page_title):]) == '')
and (not section)):
newlink = "[[%s]]%s" \
% (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" \
% (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
# Todo: This line is unreachable (T155337)
pywikibot.output(text[max(0, m.start() - 30):m.end() + 30])
if text == original_text:
pywikibot.output(u'\nNo changes have been made:\n')
else:
pywikibot.output(u'\nThe following changes have been made:\n')
pywikibot.showDiff(original_text, text)
pywikibot.output(u'')
# save the page
self.setSummaryMessage(disambPage, new_targets, unlink_counter,
dn)
try:
refPage.put(text, summary=self.comment, asynchronous=True)
except pywikibot.LockedPage:
pywikibot.output(u'Page not saved: page is locked')
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Page not saved: %s' % error.args)
return 'done'
def treat(text, linkedPage, targetPage):
"""Based on the method of the same name in solve_disambiguation.py."""
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
# make a backup of the original text so we can show the changes later
linkR = re.compile(
r"\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?" r"(\|(?P<label>[^\]]*))?\]\](?P<linktrail>" + linktrail + ")"
)
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group("title").strip() == "" or mysite.isInterwikiLink(m.group("title")):
continue
else:
actualLinkPage = pywikibot.Page(targetPage.site, m.group("title"))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
choice = "y"
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group("title")
link_text = m.group("label")
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group("section") is None:
section = ""
else:
section = m.group("section")
trailing_chars = m.group("linktrail")
if trailing_chars:
link_text += trailing_chars
if choice in "uU":
# unlink - we remove the section if there's any
text = text[: m.start()] + link_text + text[m.end() :]
continue
replaceit = choice in "rR"
# remove preleading ":"
if link_text[0] == ":":
link_text = link_text[1:]
if link_text[0].isupper():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
# remove preleading ":"
if new_page_title[0] == ":":
new_page_title = new_page_title[1:]
if replaceit and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section, trailing_chars)
elif replaceit or (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif (
len(new_page_title) <= len(link_text)
and firstcap(link_text[: len(new_page_title)]) == firstcap(new_page_title)
and re.sub(re.compile(linktrail), "", link_text[len(new_page_title) :]) == ""
and not section
):
newlink = "[[%s]]%s" % (link_text[: len(new_page_title)], link_text[len(new_page_title) :])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[: m.start()] + newlink + text[m.end() :]
continue
return text
def replace_links(self, text, linkedPage, targetPage):
"""Replace all source links by target."""
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
# make a backup of the original text so we can show the changes later
linkR = re.compile(r'\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?'
r'(\|(?P<label>[^\]]*))?\]\](?P<linktrail>' +
linktrail + ')')
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group('title').strip() == '' or \
mysite.isInterwikiLink(m.group('title')):
continue
else:
actualLinkPage = pywikibot.Page(targetPage.site,
m.group('title'))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
if section and targetPage.section():
pywikibot.warning(
'Source section {0} and target section {1} found. '
'Skipping.'.format(section, targetPage))
continue
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
# remove preleading ":"
if link_text[0] == ':':
link_text = link_text[1:]
if link_text[0].isupper() or link_text[0].isdigit():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
# remove preleading ":"
if new_page_title[0] == ':':
new_page_title = new_page_title[1:]
if (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif (len(new_page_title) <= len(link_text)
and firstcap(link_text[:len(new_page_title)])
== firstcap(new_page_title)
and re.sub(re.compile(linktrail), '',
link_text[len(new_page_title):]) == ''
and not section):
newlink = "[[%s]]%s" % (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
return text
def treat(text, linkedPage, targetPage):
"""Based on the method of the same name in solve_disambiguation.py."""
# make a backup of the original text so we can show the changes later
mysite = pywikibot.Site()
linktrail = mysite.linktrail()
linkR = re.compile(
r'\[\[(?P<title>[^\]\|#]*)(?P<section>#[^\]\|]*)?(\|(?P<label>[^\]]*))?\]\](?P<linktrail>%s)'
% linktrail)
curpos = 0
# This loop will run until we have finished the current page
while True:
m = linkR.search(text, pos=curpos)
if not m:
break
# Make sure that next time around we will not find this same hit.
curpos = m.start() + 1
# ignore interwiki links and links to sections of the same page
if m.group('title') == '' or mysite.isInterwikiLink(m.group('title')):
continue
else:
actualLinkPage = pywikibot.Page(mysite, m.group('title'))
# Check whether the link found is to page.
if actualLinkPage != linkedPage:
continue
# how many bytes should be displayed around the current link
context = 30
# at the beginning of the link, start red color.
# at the end of the link, reset the color to default
pywikibot.output(text[max(0, m.start() - context): m.start()] +
'\03{lightred}' + text[m.start(): m.end()] +
'\03{default}' + text[m.end(): m.end() + context])
choice = pywikibot.input_choice(
'What should be done with the link?',
(('Do not change', 'n'),
('Change link to \03{lightpurple}%s\03{default}'
% targetPage.title(), 'y'),
('Change and replace text', 'r'), ('Unlink', 'u')),
default='n', automatic_quit=False)
if choice == 'n':
continue
# The link looks like this:
# [[page_title|link_text]]trailing_chars
page_title = m.group('title')
link_text = m.group('label')
if not link_text:
# or like this: [[page_title]]trailing_chars
link_text = page_title
if m.group('section') is None:
section = ''
else:
section = m.group('section')
trailing_chars = m.group('linktrail')
if trailing_chars:
link_text += trailing_chars
if choice == 'u':
# unlink - we remove the section if there's any
text = text[:m.start()] + link_text + text[m.end():]
continue
if link_text[0].isupper():
new_page_title = targetPage.title()
else:
new_page_title = first_lower(targetPage.title())
if choice == 'r' and trailing_chars:
newlink = "[[%s%s]]%s" % (new_page_title, section, trailing_chars)
elif choice == 'r' or (new_page_title == link_text and not section):
newlink = "[[%s]]" % new_page_title
# check if we can create a link with trailing characters instead of a
# pipelink
elif len(new_page_title) <= len(link_text) and \
firstcap(link_text[:len(new_page_title)]) == \
firstcap(new_page_title) and \
re.sub(re.compile(linktrail), '', link_text[len(new_page_title):]) == '' and not section:
newlink = "[[%s]]%s" % (link_text[:len(new_page_title)],
link_text[len(new_page_title):])
else:
newlink = "[[%s%s|%s]]" % (new_page_title, section, link_text)
text = text[:m.start()] + newlink + text[m.end():]
continue
return text
