def run(self):
"""Run the bot."""
if not all((self.opt.action, self.generator)):
return
catmode = (self.opt.action == 'categories')
for page in self.generator:
try:
self.current_page = page
commons = page.site.image_repository()
commonspage = getattr(pywikibot,
('Page',
'Category')[catmode])(commons,
page.title())
try:
commonspage.get(get_redirect=True)
pagetitle = commonspage.title(with_ns=not catmode)
if page.title() == pagetitle:
old_text = page.get()
text = old_text
# for Commons/Commonscat template
s = self.findTemplate.search(text)
s2 = getattr(
self, 'findTemplate{}'.format(
(2, 3)[catmode]).search(text))
if s or s2:
pywikibot.output('** Already done.')
else:
cats = textlib.getCategoryLinks(text,
site=page.site)
text = textlib.replaceCategoryLinks(
'%s{{commons%s|%s}}' %
(text, ('', 'cat')[catmode], pagetitle),
cats,
site=page.site)
comment = i18n.twtranslate(
page.site,
'commons_link{}-template-added'.format(
('', '-cat')[catmode]))
try:
self.userPut(page,
old_text,
text,
summary=comment)
except EditConflictError:
pywikibot.output(
'Skipping {} because of edit conflict'.
format(page.title()))
except NoPageError:
pywikibot.output('{} does not exist in Commons'.format(
page.__class__.__name__))
except NoPageError:
pywikibot.output('Page {} does not exist'.format(page.title()))
except IsRedirectPageError:
pywikibot.output('Page {} is a redirect; skipping.'.format(
page.title()))
except LockedPageError:
pywikibot.output('Page {} is locked'.format(page.title()))
def standardizePageFooter(self, text):
"""
Standardize page footer.
Makes sure that interwiki links and categories 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. interwiki links
"""
categories = None
interwikiLinks = None
# Pywikibot is no longer allowed to touch categories on the
# German Wikipedia. See
# https://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.code not in ('et', 'it', 'bg', 'ru'):
categories = textlib.getCategoryLinks(text, site=self.site)
if not self.talkpage:
subpage = False
if self.template:
try:
tmpl, loc = moved_links[self.site.code]
del tmpl
except KeyError:
loc = None
if loc is not None and loc in self.title:
subpage = True
interwikiLinks = textlib.getLanguageLinks(text,
insite=self.site,
template_subpage=subpage)
# Removing the interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# Adding categories
if categories:
# TODO: Sorting categories in alphabetic order.
# e.g. using categories.sort()
# TODO: Taking main cats to top
text = textlib.replaceCategoryLinks(text,
categories,
site=self.site)
# Adding the interwiki
if interwikiLinks:
text = textlib.replaceLanguageLinks(text,
interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def apply(self, page, summaries=[], *args):
result = super().apply(page, summaries, *args)
if result:
categories = textlib.getCategoryLinks(page.text, site=self.site)
categories.sort(key=self.sortCategories)
page.text = textlib.replaceCategoryLinks(page.text, categories,
self.site)
return result
def run(self):
"""Run the bot."""
if not all((self.getOption('action'), self.generator)):
return
catmode = (self.getOption('action') == 'categories')
for page in self.generator:
try:
self.current_page = page
commons = page.site.image_repository()
commonspage = getattr(pywikibot,
('Page',
'Category')[catmode])(commons,
page.title())
try:
commonspage.get(get_redirect=True)
pagetitle = commonspage.title(withNamespace=not catmode)
if page.title() == pagetitle:
oldText = page.get()
text = oldText
# for Commons/Commonscat template
s = self.findTemplate.search(text)
s2 = getattr(self, 'findTemplate%d' %
(2, 3)[catmode]).search(text)
if s or s2:
pywikibot.output(u'** Already done.')
else:
cats = textlib.getCategoryLinks(text,
site=page.site)
text = textlib.replaceCategoryLinks(
u'%s{{commons%s|%s}}' %
(text, ('', 'cat')[catmode], pagetitle),
cats,
site=page.site)
comment = i18n.twtranslate(
page.site, 'commons_link%s-template-added' %
('', '-cat')[catmode])
try:
self.userPut(page,
oldText,
text,
summary=comment)
except pywikibot.EditConflict:
pywikibot.output(
u'Skipping %s because of edit conflict' %
page.title())
except pywikibot.NoPage:
pywikibot.output(u'%s does not exist in Commons' %
page.__class__.__name__)
except pywikibot.NoPage:
pywikibot.output(u'Page %s does not exist' % page.title())
except pywikibot.IsRedirectPage:
pywikibot.output(u'Page %s is a redirect; skipping.' %
page.title())
except pywikibot.LockedPage:
pywikibot.output(u'Page %s is locked' % page.title())
def test_adjoining_links(self):
old = self.old.replace(config.LS, '')
cats = textlib.getCategoryLinks(old, site=self.site)
self.assertEqual(self.cats, cats)
sep = config.LS
config.line_separator = '' # use an empty separator temporarily
new = textlib.replaceCategoryLinks(old, cats, site=self.site)
self.assertEqual(old, new)
config.line_separator = sep # restore the default separator
def make_categories(page, list, site=None):
if site is None:
site = pywikibot.Site()
pllist = []
for p in list:
cattitle = "%s:%s" % (site.namespaces.CATEGORY, p)
pllist.append(pywikibot.Page(site, cattitle))
page.put_async(textlib.replaceCategoryLinks(page.get(), pllist,
site=page.site),
summary=i18n.twtranslate(site, 'catall-changing'))
def make_categories(page, list, site=None):
if site is None:
site = pywikibot.Site()
pllist = []
for p in list:
cattitle = "%s:%s" % (site.category_namespace(), p)
pllist.append(pywikibot.Page(site, cattitle))
page.put_async(textlib.replaceCategoryLinks(page.get(), pllist,
site=page.site),
summary=i18n.twtranslate(site, 'catall-changing'))
def make_categories(page, list, site=None):
"""Make categories."""
if site is None:
site = pywikibot.Site()
pllist = []
for p in list:
pllist.append(pywikibot.Page(site, 'Category:' + p))
page.put(textlib.replaceCategoryLinks(page.get(), pllist, site=page.site),
asynchronous=True,
summary=i18n.twtranslate(site, 'catall-changing'))
def test_adjoining_links(self):
cats_std = textlib.getCategoryLinks(self.old, site=self.site)
old = self.old.replace(config.LS, "")
cats = textlib.getCategoryLinks(old, site=self.site)
self.assertEqual(cats_std, cats)
sep = config.LS
config.line_separator = "" # use an empty separator temporarily
new = textlib.replaceCategoryLinks(old, cats, site=self.site)
# Restore the default separator.
config.line_separator = sep
self.assertEqual(old, new)
def run(self):
"""Run the bot."""
if not all((self.getOption('action'), self.generator)):
return
catmode = (self.getOption('action') == 'categories')
for page in self.generator:
try:
self.current_page = page
commons = page.site.image_repository()
commonspage = getattr(pywikibot,
('Page', 'Category')[catmode]
)(commons, page.title())
try:
commonspage.get(get_redirect=True)
pagetitle = commonspage.title(withNamespace=not catmode)
if page.title() == pagetitle:
oldText = page.get()
text = oldText
# for Commons/Commonscat template
s = self.findTemplate.search(text)
s2 = getattr(self, 'findTemplate%d'
% (2, 3)[catmode]).search(text)
if s or s2:
pywikibot.output(u'** Already done.')
else:
cats = textlib.getCategoryLinks(text,
site=page.site)
text = textlib.replaceCategoryLinks(
u'%s{{commons%s|%s}}'
% (text, ('', 'cat')[catmode], pagetitle),
cats, site=page.site)
comment = i18n.twtranslate(
page.site, 'commons_link%s-template-added'
% ('', '-cat')[catmode])
try:
self.userPut(page, oldText, text,
summary=comment)
except pywikibot.EditConflict:
pywikibot.output(
u'Skipping %s because of edit conflict'
% page.title())
except pywikibot.NoPage:
pywikibot.output(u'%s does not exist in Commons'
% page.__class__.__name__)
except pywikibot.NoPage:
pywikibot.output(u'Page %s does not exist' % page.title())
except pywikibot.IsRedirectPage:
pywikibot.output(u'Page %s is a redirect; skipping.'
% page.title())
except pywikibot.LockedPage:
pywikibot.output(u'Page %s is locked' % page.title())
def make_categories(page, list, site=None):
"""Make categories."""
if site is None:
site = pywikibot.Site()
pllist = []
for p in list:
cattitle = "%s:%s" % (site.namespaces.CATEGORY, p)
pllist.append(pywikibot.Page(site, cattitle))
page.put(textlib.replaceCategoryLinks(page.get(), pllist, site=page.site),
asynchronous=True,
summary=i18n.twtranslate(site, 'catall-changing'))
def test_adjoining_links(self):
"""Test getting and replacing adjacent categories."""
cats_std = textlib.getCategoryLinks(self.old, site=self.site)
old = self.old.replace(config.LS, '')
cats = textlib.getCategoryLinks(old, site=self.site)
self.assertEqual(cats_std, cats)
sep = config.LS
config.line_separator = '' # use an empty separator temporarily
new = textlib.replaceCategoryLinks(old, cats, site=self.site)
# Restore the default separator.
config.line_separator = sep
self.assertEqual(old, new)
def test_adjoining_links(self):
"""Test getting and replacing adjacent categories."""
cats_std = textlib.getCategoryLinks(self.old, site=self.site)
old = self.old.replace(config.LS, '')
cats = textlib.getCategoryLinks(old, site=self.site)
self.assertEqual(cats_std, cats)
sep = config.LS
config.line_separator = '' # use an empty separator temporarily
new = textlib.replaceCategoryLinks(old, cats, site=self.site)
# Restore the default separator.
config.line_separator = sep
self.assertEqual(old, new)
def replace_default_cat_with_new_categories_in_image_text(
old_text, base_category, new_categories):
"""Add new categories to page text and remove any base_category."""
if not new_categories:
# No categories to add. We do not want to remove the base one,
raise NoCategoryToAddException()
# Remove base category
page_text_without_base_category = textlib.replaceCategoryInPlace(
old_text, base_category, None)
final_text = textlib.replaceCategoryLinks(
page_text_without_base_category, new_categories, addOnly=True)
return final_text
def include(self,
pl,
checklinks=True,
realinclude=True,
linkterm=None,
summary=''):
"""Include the current page to the working category."""
global workingcat, parentcats
global checked, tocheck
cl = checklinks
mysite = self.site
if linkterm:
actualworkingcat = pywikibot.Category(mysite,
workingcat.title(),
sort_key=linkterm)
else:
actualworkingcat = workingcat
if realinclude:
try:
text = pl.get()
except pywikibot.NoPage:
pass
except pywikibot.IsRedirectPage:
cl = True
else:
cats = [x for x in pl.categories()]
if workingcat not in cats:
cats = [x for x in pl.categories()]
for c in cats:
if c in parentcats:
if self.removeparent:
pl.change_category(actualworkingcat,
summary=summary)
break
else:
pl.put(textlib.replaceCategoryLinks(text,
cats +
[actualworkingcat],
site=pl.site),
summary=summary)
if cl:
if self.checkforward:
for page2 in pl.linkedPages():
if self.needcheck(page2):
tocheck.append(page2)
checked[page2] = page2
if self.checkbackward:
for ref_page in pl.getReferences():
if self.needcheck(ref_page):
tocheck.append(ref_page)
checked[ref_page] = ref_page
def make_categories(page, list: list, site=None):
"""Make categories.
:param page: The page to update and save
:type page: pywikibot.page.BasePage
:param list: The list which contains categories
"""
if site is None:
site = pywikibot.Site()
pllist = []
for p in list:
pllist.append(pywikibot.Page(site, 'Category:' + p))
page.put(textlib.replaceCategoryLinks(page.get(), pllist, site=page.site),
asynchronous=True,
summary=i18n.twtranslate(site, 'catall-changing'))
def addCategory(site, page, cat):
old_text = page.text
cats = textlib.getCategoryLinks(old_text)
catpl = pywikibot.Category(site, cat)
if catpl not in cats:
print("\t'" + cat + "' not in page categories. Adding")
cats.append(catpl)
text = textlib.replaceCategoryLinks(page.text, cats, site=site)
userPut(page, old_text, text, minor=True, botflag=True)
return True
else:
print("\t'" + cat + "' already in page categories")
return False
def treat(self, page):
"""Process one page."""
if page.isRedirectPage():
# if it's a redirect use the redirect target instead
redirTarget = page.getRedirectTarget()
if self.follow_redirects:
self.current_page = redirTarget
else:
pywikibot.warning(
"Page %s is a redirect to %s; skipping." % (page.title(asLink=True), redirTarget.title(asLink=True))
)
# loading it will throw an error if we don't jump out before
return
else:
self.current_page = page
if self.current_page.exists():
# Load the page
text = self.current_page.text
elif self.create:
pywikibot.output("Page %s doesn't exist yet; creating." % (self.current_page.title(asLink=True)))
text = ""
else:
pywikibot.output("Page %s does not exist; skipping." % self.current_page.title(asLink=True))
return
# store old text, so we don't have reload it every time
old_text = text
cats = textlib.getCategoryLinks(text)
pywikibot.output("Current categories:")
for cat in cats:
pywikibot.output("* %s" % cat.title())
catpl = pywikibot.Category(self.current_page.site, self.newcat)
if catpl in cats:
pywikibot.output("%s is already in %s." % (self.current_page.title(), catpl.title()))
else:
if self.sort:
catpl = self.sorted_by_last_name(catpl, self.current_page)
pywikibot.output("Adding %s" % catpl.title(asLink=True))
cats.append(catpl)
text = textlib.replaceCategoryLinks(text, cats, site=self.current_page.site)
comment = self.comment
if not comment:
comment = i18n.twtranslate(
self.current_page.site, "category-adding", {"newcat": catpl.title(withNamespace=False)}
)
try:
self.userPut(self.current_page, old_text, text, summary=comment, minor=True, botflag=True)
except pywikibot.PageSaveRelatedError as error:
pywikibot.output("Page %s not saved: %s" % (self.current_page.title(asLink=True), error))
def remove_categories():
site = pywikibot.Site()
category = pywikibot.Category(site, "Category:Files")
generator = pagegenerators.PreloadingGenerator(
pagegenerators.CategorizedPageGenerator(category=category))
for page in generator:
pywikibot.output("")
pywikibot.output(vars(page))
pywikibot.output(page.text)
page.text = textlib.replaceCategoryLinks(page.text, [], site=page.site)
page.save(summary="Remove categories.",
# asynchronous=True,
)
def set_category_status(site, page, cat, status):
old_text = page.text
cats = textlib.getCategoryLinks(old_text)
catpl = pywikibot.Category(site, cat)
if status:
if catpl not in cats:
cats.append(catpl)
else:
if catpl in cats:
cats.remove(catpl)
text = textlib.replaceCategoryLinks(page.text, cats, site=site)
if old_text != text:
page.text = text
page.save(minor=True, botflag=True)
return True
return False
def set_category_status(site, page, cat, status):
old_text = page.text
cats = textlib.getCategoryLinks(old_text)
catpl = pywikibot.Category(site, cat)
if status:
if catpl not in cats:
cats.append(catpl)
else:
if catpl in cats:
cats.remove(catpl)
text = textlib.replaceCategoryLinks(page.text, cats, site=site)
if old_text != text:
page.text = text
page.save(minor=True, botflag=True)
return True
return False
def apply(self, page, *args):
# remove empty list items
page.text = re.sub(r'^\* *\n', '', page.text, flags=re.M)
# sort categories
categories = textlib.getCategoryLinks(page.text, site=page.site)
category_living = pywikibot.Category(page.site, 'Žijící lidé')
if category_living in categories:
if any(
cat.title(with_ns=False).startswith('Úmrtí ')
for cat in categories):
categories.remove(category_living)
page.text = textlib.replaceCategoryLinks(
page.text, categories, page.site)
page.text = re.sub(
r'(\{\{ *(?:%s)[^}]+\}\}\n)\n(\[\[(?:%s))' %
('|'.join(map(re.escape, self.site.getmagicwords('defaultsort'))),
'|'.join(self.site.namespaces[14])), r'\1\2', page.text)
def harvestSortKey(self, match):
text = match.group()
if self.defaultsortR.search(text):
return text
keys = {}
categories = textlib.getCategoryLinks(text, site=self.site)
if not any(
category.title(with_ns=False) in ('Muži', 'Žijící lidé',
'Ženy')
for category in categories):
return text
for category in categories:
key = category.sortKey
if key:
key = self.tidy_sortkey(key)
if not key.strip():
continue
keys.setdefault(key, 0.0)
keys[key] += 1
if len(keys) > 1:
return text
if not keys:
return text
if sum(keys.values()) < 4:
return text
key = list(keys.keys()).pop()
for category in categories:
if category.sortKey is not None:
if self.tidy_sortkey(category.sortKey) == key:
category.sortKey = None
categories.sort(key=self.sort_category)
text = textlib.removeCategoryLinks(text, self.site)
text += '\n\n{{DEFAULTSORT:%s}}' % key
before, _, after = textlib.replaceCategoryLinks(
text, categories, self.site).rpartition('\n\n') # fixme: safer
return before + '\n' + after
def include(pl, checklinks=True, realinclude=True, linkterm=None):
cl = checklinks
if linkterm:
actualworkingcat = pywikibot.Category(mysite,
workingcat.title(),
sortKey=linkterm)
else:
actualworkingcat = workingcat
if realinclude:
try:
text = pl.get()
except pywikibot.NoPage:
pass
except pywikibot.IsRedirectPage:
cl = True
pass
else:
cats = [x for x in pl.categories()]
if workingcat not in cats:
cats = [x for x in pl.categories()]
for c in cats:
if c in parentcats:
if removeparent:
pl.change_category(actualworkingcat)
break
else:
pl.put(
textlib.replaceCategoryLinks(text,
cats + [actualworkingcat],
site=pl.site))
if cl:
if checkforward:
for page2 in pl.linkedPages():
if needcheck(page2):
tocheck.append(page2)
checked[page2] = page2
if checkbackward:
for refPage in pl.getReferences():
if needcheck(refPage):
tocheck.append(refPage)
checked[refPage] = refPage
def include(self, page, checklinks=True, realinclude=True, linkterm=None):
"""Include the current page to the working category."""
global workingcat, parentcats
global checked
actualworkingcat = workingcat
if linkterm:
actualworkingcat.sortKey = linkterm
if realinclude and page.exists():
if page.isRedirectPage():
checklinks = True
else:
cats = list(page.categories())
if workingcat not in cats \
and not self.change_category(page, cats):
newtext = textlib.replaceCategoryLinks(
page.text, cats + [actualworkingcat],
site=page.site)
page.put(newtext, summary=self.opt.summary)
if checklinks:
self.checklinks(page)
def duplicateSortKey(self, match):
text = match.group()
matches = list(self.defaultsortR.finditer(text))
if not matches:
return text
defaultsort = matches.pop().group(1).strip()
categories = textlib.getCategoryLinks(text, site=self.site)
changed = False
for category in categories:
if self.tidy_sortkey(category.sortKey) == defaultsort:
category.sortKey = None
changed = True
if changed:
categories.sort(key=self.sort_category)
before, _, after = textlib.replaceCategoryLinks(
text, categories, self.site).rpartition('\n\n') # fixme: safer
return before + '\n' + after
else:
return text
def include(pl, checklinks=True, realinclude=True, linkterm=None, summary=''):
cl = checklinks
if linkterm:
actualworkingcat = pywikibot.Category(mysite, workingcat.title(),
sortKey=linkterm)
else:
actualworkingcat = workingcat
if realinclude:
try:
text = pl.get()
except pywikibot.NoPage:
pass
except pywikibot.IsRedirectPage:
cl = True
pass
else:
cats = [x for x in pl.categories()]
if workingcat not in cats:
cats = [x for x in pl.categories()]
for c in cats:
if c in parentcats:
if removeparent:
pl.change_category(actualworkingcat,
summary=summary)
break
else:
pl.put(textlib.replaceCategoryLinks(
text, cats + [actualworkingcat], site=pl.site),
summary=summary)
if cl:
if checkforward:
for page2 in pl.linkedPages():
if needcheck(page2):
tocheck.append(page2)
checked[page2] = page2
if checkbackward:
for refPage in pl.getReferences():
if needcheck(refPage):
tocheck.append(refPage)
checked[refPage] = refPage
def categorize_file_page(site, page, p=0):
status = {"f": 0, "p": 1, "w": 0, "e": 0}
pywikibot.output(f"Page {p + 1}:")
is_file_page = page.is_filepage()
file_page = None
file_info = None
if is_file_page:
file_page = pywikibot.FilePage(page)
try:
file_info = file_page.latest_file_info
except (NoPage, PageRelatedError):
is_file_page = False
if not is_file_page:
pywikibot.error("Page \"" + page.title(as_link=True) +
"\" is not a file page. Skipping page...")
pywikibot.output("")
status["e"] += 1
return status
uri = file_page.full_url()
mime = file_info.mime.lower()
pywikibot.output(" Title: " + file_page.title(as_link=True))
pywikibot.output(" URI: " + uri)
pywikibot.output(" MIME type: " + mime)
# Find a matching category for the file's MIME type or extension.
mime_categories = flattened_categories["mime_categories"]
mime_pattern_category_regexes = flattened_categories[
"mime_pattern_category_regexes"]
found_category = None
if mime in mime_categories:
found_category = mime_categories[mime]
else:
pywikibot.warning(
f"Unrecognized MIME type \"{mime}\". Attempting to search by regex..."
)
status["w"] += 1
for pattern_category_regex in mime_pattern_category_regexes:
regex = pattern_category_regex["regex"]
if regex.search(mime) is not None:
category = pattern_category_regex["category"]
pywikibot.warning(
f"Found category \"{category}\" for unrecognized MIME type \"{mime}\"."
)
status["w"] += 1
found_category = category
break
if found_category is None:
pywikibot.warning(
f"No category found for MIME type \"{mime}\". Attempting to search by file extension..."
)
status["w"] += 1
file_extension = os.path.splitext(uri)[1][1:].strip().lower()
if len(file_extension) > 0:
extension_categories = flattened_categories["extension_categories"]
extension_pattern_category_regexes = flattened_categories[
"extension_pattern_category_regexes"]
if file_extension in extension_categories:
found_category = extension_categories[file_extension]
else:
pywikibot.warning(
f"Unrecognized file extension \"{file_extension}\". Attempting to search by regex..."
)
status["w"] += 1
for pattern_category_regex in extension_pattern_category_regexes:
regex = pattern_category_regex["regex"]
if regex.search(file_extension) is not None:
category = pattern_category_regex["category"]
pywikibot.warning(
f"Found category \"{category}\" for unrecognized file extension \"{file_extension}\"."
)
status["w"] += 1
found_category = category
break
if found_category is None:
pywikibot.error(
f"No category found for MIME type \"{mime}\" or file extension \"{file_extension}\". Skipping file page..."
)
pywikibot.output("")
status["e"] += 1
return status
# Build categories, and add them to the file page.
categories = build_categories(found_category)
pywikibot.output(" Add categories: " + ", ".join(categories))
page_categories = []
category_wikilinks = []
for category in categories:
page_category = pywikibot.Page(site, "Category:" + category)
page_categories.append(page_category)
category_wikilink = "[[Category:{0}|{0}]]".format(category)
category_wikilinks.append(category_wikilink)
file_page.text = textlib.replaceCategoryLinks(file_page.text,
page_categories,
site=file_page.site,
addOnly=True)
summary = "Add the {0} {1}.".format(
"categor" + ("y" if len(category_wikilinks) <= 1 else "ies"),
", ".join(category_wikilinks))
file_page.save(summary=summary, minor=False)
status["f"] += 1
return status
def standardizePageFooter(self, text):
"""
Standardize page footer.
Makes sure that interwiki links and categories are put
into the correct position and into the right order. This
combines the old instances of standardizeInterwiki
and standardizeCategories.
The page footer consists of the following parts
in that sequence:
1. categories
2. additional information depending on the local site policy
3. interwiki
"""
categories = []
interwiki_links = []
# get categories
if not self.template:
categories = textlib.getCategoryLinks(text, site=self.site)
if not self.talkpage:
subpage = False
if self.template:
try:
tmpl, loc = moved_links[self.site.code]
del tmpl
except KeyError:
loc = None
if loc is not None and loc in self.title:
subpage = True
# get interwiki
interwiki_links = textlib.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# remove interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# add categories, main to top
if categories:
# TODO: Sort categories in alphabetic order, e.g. using
# categories.sort()? (T100265)
# TODO: Get main categories from Wikidata?
main = pywikibot.Category(self.site,
'Category:' + self.title,
sort_key=' ')
if main in categories:
categories.pop(categories.index(main))
categories.insert(0, main)
text = textlib.replaceCategoryLinks(text,
categories,
site=self.site)
# add interwiki
if interwiki_links:
text = textlib.replaceLanguageLinks(text,
interwiki_links,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def add_text(page, addText, summary=None, regexSkip=None,
regexSkipUrl=None, always=False, up=False, putText=True,
oldTextGiven=None, reorderEnabled=True, create=False):
"""
Add text to a page.
@rtype: tuple of (text, newtext, always)
"""
site = page.site
if not summary:
summary = i18n.twtranslate(site, '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
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 = page.full_url()
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output(
'Exception! regex (or word) used with -exceptUrl '
'is in the page. Skip!\n'
'Match was: %s' % result)
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output(
'Exception! regex (or word) used with -except '
'is in the page. Skip!\n'
'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 = textlib.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = textlib.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = textlib.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = textlib.removeLanguageLinks(newtext, site)
# Adding the text
newtext += u"%s%s" % (config.line_separator, addText)
# Reputting the categories
newtext = textlib.replaceCategoryLinks(newtext,
categoriesInside, site,
True)
# Dealing the stars' issue
allstars = []
starstext = textlib.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 = textlib.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.input_choice(
u'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('All', 'a'),
('open in Browser', 'b')], 'n', automatic_quit=False)
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
pywikibot.bot.open_webbrowser(page)
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.max_retries:
pywikibot.output(u'Server Error! Wait..')
time.sleep(config.retry_wait)
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.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% page.title())
return (False, False, always)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: %s' % error.args)
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 run(self):
"""Start the bot."""
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
if self.isTitleExcepted(page.title()):
pywikibot.output(
'Skipping {0} because the title is on the exceptions list.'
.format(page.title(as_link=True)))
continue
try:
# Load the page's text from the wiki
original_text = page.get(get_redirect=True)
if not page.canBeEdited():
pywikibot.output("You can't edit page " +
page.title(as_link=True))
continue
except pywikibot.NoPage:
pywikibot.output('Page {0} not found'.format(
page.title(as_link=True)))
continue
applied = set()
new_text = original_text
last_text = None
context = 0
while True:
if self.isTextExcepted(new_text):
pywikibot.output('Skipping {0} because it contains text '
'that is on the exceptions list.'.format(
page.title(as_link=True)))
break
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied,
page)
if not self.recursive:
break
if new_text == original_text:
pywikibot.output('No changes were necessary in ' +
page.title(as_link=True))
break
if hasattr(self, 'addedCat'):
# Fetch only categories in wikitext, otherwise the others
# will be explicitly added.
cats = textlib.getCategoryLinks(new_text, site=page.site)
if self.addedCat not in cats:
cats.append(self.addedCat)
new_text = textlib.replaceCategoryLinks(new_text,
cats,
site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
pywikibot.output(
color_format('\n\n>>> {lightpurple}{0}{default} <<<',
page.title()))
pywikibot.showDiff(original_text, new_text, context=context)
if self.getOption('always'):
break
choice = pywikibot.input_choice(
'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('Edit original', 'e'),
('edit Latest', 'l'), ('open in Browser', 'b'),
('More context', 'm'), ('All', 'a')],
default='N')
if choice == 'm':
context = context * 3 if context else 3
continue
if choice == 'e':
editor = editarticle.TextEditor()
as_edited = editor.edit(original_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
continue
if choice == 'l':
editor = editarticle.TextEditor()
as_edited = editor.edit(new_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
# prevent changes from being applied again
last_text = new_text
continue
if choice == 'b':
pywikibot.bot.open_webbrowser(page)
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output('Page {0} has been deleted.'.format(
page.title()))
break
new_text = original_text
last_text = None
continue
if choice == 'a':
self.options['always'] = True
if choice == 'y':
page.text = new_text
page.save(summary=self.generate_summary(applied),
asynchronous=True,
callback=self._replace_async_callback,
quiet=True)
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output('Page {0}{1} saved'.format(
proc_title, '' if res else ' not'))
# choice must be 'N'
break
if self.getOption('always') and new_text != original_text:
try:
page.text = new_text
page.save(summary=self.generate_summary(applied),
callback=self._replace_sync_callback,
quiet=True)
except pywikibot.EditConflict:
pywikibot.output(
'Skipping {0} because of edit conflict'.format(
page.title(), ))
except pywikibot.SpamfilterError as e:
pywikibot.output(
'Cannot change {0} because of blacklist entry {1}'.
format(page.title(), e.url))
except pywikibot.LockedPage:
pywikibot.output('Skipping {0} (locked page)'.format(
page.title(), ))
except pywikibot.PageNotSaved as error:
pywikibot.output('Error putting page: {0}'.format(
error.args, ))
if self._pending_processed_titles.qsize() > 50:
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output('Page {0}{1} saved'.format(
proc_title, '' if res else ' not'))
def test_standard_links(self):
cats = textlib.getCategoryLinks(self.old, site=self.site)
new = textlib.replaceCategoryLinks(self.old, cats, site=self.site)
self.assertEqual(self.old, new)
def assertRoundtripCategory(self, text, catcount):
cats = textlib.getCategoryLinks(text)
self.assertEqual(len(cats), catcount)
self.assertEqual(text, textlib.replaceCategoryLinks(text,
cats,
site = self.site))
def run(self):
"""Start the bot."""
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
if self.isTitleExcepted(page.title()):
pywikibot.output(
u'Skipping %s because the title is on the exceptions list.'
% page.title(asLink=True))
continue
try:
# Load the page's text from the wiki
original_text = page.get(get_redirect=True)
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
applied = set()
new_text = original_text
while True:
if self.isTextExcepted(new_text):
pywikibot.output(u'Skipping %s because it contains text '
u'that is on the exceptions list.' %
page.title(asLink=True))
break
last_text = None
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied)
if not self.recursive:
break
if new_text == original_text:
pywikibot.output(u'No changes were necessary in %s' %
page.title(asLink=True))
break
if hasattr(self, 'addedCat'):
# Fetch only categories in wikitext, otherwise the others will
# be explicitly added.
cats = textlib.getCategoryLinks(new_text, site=page.site)
if self.addedCat not in cats:
cats.append(self.addedCat)
new_text = textlib.replaceCategoryLinks(new_text,
cats,
site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
pywikibot.output(
u"\n\n>>> \03{lightpurple}%s\03{default} <<<" %
page.title())
pywikibot.showDiff(original_text, new_text)
if self.getOption('always'):
break
choice = pywikibot.input_choice(
u'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('Edit', 'e'),
('open in Browser', 'b'), ('all', 'a')],
default='N')
if choice == 'e':
editor = editarticle.TextEditor()
as_edited = editor.edit(original_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
continue
if choice == 'b':
pywikibot.bot.open_webbrowser(page)
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output(u'Page %s has been deleted.' %
page.title())
break
new_text = original_text
continue
if choice == 'a':
self.options['always'] = True
if choice == 'y':
page.put_async(new_text,
self.generate_summary(applied),
callback=self.count_changes)
# choice must be 'N'
break
if self.getOption('always') and new_text != original_text:
try:
page.put(new_text,
self.generate_summary(applied),
callback=self.count_changes)
except pywikibot.EditConflict:
pywikibot.output(u'Skipping %s because of edit conflict' %
(page.title(), ))
except pywikibot.SpamfilterError as e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s' %
(page.title(), e.url))
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)' %
(page.title(), ))
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: %s' %
(error.args, ))
def standardizePageFooter(self, text):
"""
Standardize page footer.
Makes sure that interwiki links and categories 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. interwiki links
"""
categories = None
interwikiLinks = None
# Pywikibot is no longer allowed to touch categories on the
# German Wikipedia. See
# https://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.code not in ('et', 'it', 'bg', 'ru'):
categories = textlib.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.code]
del tmpl
except KeyError:
pass
if loc is not None and loc in self.title:
subpage = True
interwikiLinks = textlib.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# Adding categories
if categories:
# TODO: Sorting categories in alphabetic order.
# e.g. using categories.sort()
# TODO: 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 = textlib.replaceCategoryLinks(text, categories,
site=self.site)
# Adding the interwiki
if interwikiLinks:
text = textlib.replaceLanguageLinks(text, interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def add_text(page,
addText,
summary=None,
regexSkip=None,
regexSkipUrl=None,
always=False,
up=False,
putText=True,
oldTextGiven=None,
reorderEnabled=True,
create=False):
"""
Add text to a page.
@rtype: tuple of (text, newtext, always)
"""
site = page.site
if not summary:
summary = i18n.twtranslate(site, 'add_text-adding',
{'adding': addText[:200]})
if putText:
pywikibot.output('Loading {}...'.format(page.title()))
text = get_text(page, oldTextGiven, create)
if text is None:
return (False, False, always)
# 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 = page.full_url()
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output('Exception! regex (or word) used with -exceptUrl '
'is in the page. Skip!\n'
'Match was: {}'.format(result))
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output('Exception! regex (or word) used with -except '
'is in the page. Skip!\n'
'Match was: {}'.format(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 = textlib.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = textlib.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = textlib.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = textlib.removeLanguageLinks(newtext, site)
# Adding the text
newtext += '{}{}'.format(config.line_separator, addText)
# Reputting the categories
newtext = textlib.replaceCategoryLinks(newtext, categoriesInside,
site, True)
# Adding the interwiki
newtext = textlib.replaceLanguageLinks(newtext, interwikiInside,
site)
else:
newtext += '{}{}'.format(config.line_separator, addText)
else:
newtext = addText + config.line_separator + text
if putText and text != newtext:
pywikibot.output(
color_format('\n\n>>> {lightpurple}{0}{default} <<<',
page.title()))
pywikibot.showDiff(text, newtext)
# Let's put the changes.
error_count = 0
while True:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
if not putText:
return (text, newtext, always)
if not always:
try:
choice = pywikibot.input_choice(
'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('All', 'a'),
('open in Browser', 'b')], 'n')
except QuitKeyboardInterrupt:
sys.exit('User quit bot run.')
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
pywikibot.bot.open_webbrowser(page)
if always or choice == 'y':
result = put_text(page,
newtext,
summary,
error_count,
asynchronous=not always)
if result is not None:
return (result, result, always)
error_count += 1
def treat(self, page):
"""Work on each page retrieved from generator."""
original_text = page.text
applied = set()
new_text = original_text
last_text = None
context = 0
while True:
if self.isTextExcepted(new_text):
pywikibot.output(
'Skipping {} because it contains text '
'that is on the exceptions list.'.format(page))
return
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied, page)
if not self.opt.recursive:
break
if new_text == original_text:
pywikibot.output('No changes were necessary in ' +
page.title(as_link=True))
return
if self.opt.addcat:
# Fetch only categories in wikitext, otherwise the others
# will be explicitly added.
cats = textlib.getCategoryLinks(new_text, site=page.site)
if self.opt.addcat not in cats:
cats.append(self.opt.addcat)
new_text = textlib.replaceCategoryLinks(new_text,
cats,
site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
self.current_page = page
pywikibot.showDiff(original_text, new_text, context=context)
if self.opt.always:
break
choice = pywikibot.input_choice(
'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('Edit original', 'e'),
('edit Latest', 'l'), ('open in Browser', 'b'),
('More context', 'm'), ('All', 'a')],
default='N')
if choice == 'm':
context = context * 3 if context else 3
continue
if choice in ('e', 'l'):
text_editor = editor.TextEditor()
edit_text = original_text if choice == 'e' else new_text
as_edited = text_editor.edit(edit_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
if choice == 'l':
# prevent changes from being applied again
last_text = new_text
continue
if choice == 'b':
pywikibot.bot.open_webbrowser(page)
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output('Page {0} has been deleted.'.format(
page.title()))
break
new_text = original_text
last_text = None
continue
if choice == 'a':
self.opt.always = True
if choice == 'y':
self.save(page,
original_text,
new_text,
applied,
show_diff=False,
quiet=True,
callback=self._replace_async_callback,
asynchronous=True)
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output('Page {0}{1} saved'.format(
proc_title, '' if res else ' not'))
# choice must be 'N'
break
if self.opt.always and new_text != original_text:
self.save(page,
original_text,
new_text,
applied,
show_diff=False,
asynchronous=False)
def standardizePageFooter(self, text):
"""
Standardize page footer.
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
# https://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.code not in ('et', 'it', 'bg', 'ru'):
categories = textlib.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.code]
del tmpl
except KeyError:
pass
if loc is not None and loc in self.title:
subpage = True
interwikiLinks = textlib.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# Removing the stars' issue
starstext = textlib.removeDisabledParts(text)
for star in starsList:
regex = re.compile(r'(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
text = regex.sub('', text)
allstars += found
# Adding categories
if categories:
# TODO: Sorting categories in alphabetic order.
# e.g. using categories.sort()
# TODO: 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 = textlib.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%s' % (element.strip(), config.line_separator)
pywikibot.log(u'%s' % element.strip())
# Adding the interwiki
if interwikiLinks:
text = textlib.replaceLanguageLinks(text, interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def standardizePageFooter(self, text):
"""
Standardize page footer.
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
# https://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.code not in ('et', 'it', 'bg', 'ru'):
categories = textlib.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.code]
del tmpl
except KeyError:
pass
if loc is not None and loc in self.title:
subpage = True
interwikiLinks = textlib.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# Removing the interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# Removing the stars' issue
starstext = textlib.removeDisabledParts(text)
for star in starsList:
regex = re.compile(r'(\{\{(?:template:|)%s\|.*?\}\}[\s]*)'
% star, re.I)
found = regex.findall(starstext)
if found != []:
text = regex.sub('', text)
allstars += found
# Adding categories
if categories:
# TODO: Sorting categories in alphabetic order.
# e.g. using categories.sort()
# TODO: 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 = textlib.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%s' % (element.strip(), config.line_separator)
pywikibot.log(u'%s' % element.strip())
# Adding the interwiki
if interwikiLinks:
text = textlib.replaceLanguageLinks(text, interwikiLinks,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def standardizePageFooter(self, text):
"""
Standardize page footer.
Makes sure that interwiki links and categories are put
into the correct position and into the right order. This
combines the old instances of standardizeInterwiki
and standardizeCategories.
The page footer consists of the following parts
in that sequence:
1. categories
2. additional information depending on the local site policy
3. interwiki
"""
categories = []
interwiki_links = []
# get categories
if not self.template:
categories = textlib.getCategoryLinks(text, site=self.site)
if not self.talkpage:
subpage = False
if self.template:
try:
tmpl, loc = moved_links[self.site.code]
del tmpl
except KeyError:
loc = None
if loc is not None and loc in self.title:
subpage = True
# get interwiki
interwiki_links = textlib.getLanguageLinks(
text, insite=self.site, template_subpage=subpage)
# remove interwiki
text = textlib.removeLanguageLinks(text, site=self.site)
# add categories, main to top
if categories:
# TODO: Sort categories in alphabetic order, e.g. using
# categories.sort()? (T100265)
# TODO: Get main categories from Wikidata?
main = pywikibot.Category(self.site, 'Category:' + self.title,
sort_key=' ')
if main in categories:
categories.pop(categories.index(main))
categories.insert(0, main)
text = textlib.replaceCategoryLinks(text, categories,
site=self.site)
# add interwiki
if interwiki_links:
text = textlib.replaceLanguageLinks(text, interwiki_links,
site=self.site,
template=self.template,
template_subpage=subpage)
return text
def run(self):
"""Start the bot."""
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
if self.isTitleExcepted(page.title()):
pywikibot.output(
u'Skipping {0!s} because the title is on the exceptions list.'.format(page.title(asLink=True)))
continue
try:
# Load the page's text from the wiki
original_text = page.get(get_redirect=True)
if not page.canBeEdited():
pywikibot.output(u"You can't edit page {0!s}".format(page.title(asLink=True)))
continue
except pywikibot.NoPage:
pywikibot.output(u'Page {0!s} not found'.format(page.title(asLink=True)))
continue
applied = set()
new_text = original_text
while True:
if self.isTextExcepted(new_text):
pywikibot.output(u'Skipping %s because it contains text '
u'that is on the exceptions list.'
% page.title(asLink=True))
break
last_text = None
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied,
page)
if not self.recursive:
break
if new_text == original_text:
pywikibot.output(u'No changes were necessary in {0!s}'.format(page.title(asLink=True)))
break
if hasattr(self, 'addedCat'):
# Fetch only categories in wikitext, otherwise the others will
# be explicitly added.
cats = textlib.getCategoryLinks(new_text, site=page.site)
if self.addedCat not in cats:
cats.append(self.addedCat)
new_text = textlib.replaceCategoryLinks(new_text,
cats,
site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
pywikibot.output(color_format(
'\n\n>>> {lightpurple}{0}{default} <<<', page.title()))
pywikibot.showDiff(original_text, new_text)
if self.getOption('always'):
break
choice = pywikibot.input_choice(
u'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('Edit', 'e'),
('open in Browser', 'b'), ('all', 'a')],
default='N')
if choice == 'e':
editor = editarticle.TextEditor()
as_edited = editor.edit(original_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
continue
if choice == 'b':
pywikibot.bot.open_webbrowser(page)
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output(u'Page {0!s} has been deleted.'.format(page.title()))
break
new_text = original_text
continue
if choice == 'a':
self.options['always'] = True
if choice == 'y':
page.text = new_text
page.save(summary=self.generate_summary(applied), async=True,
callback=self._count_changes, quiet=True)
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output('Page {0!s}{1!s} saved'.format(proc_title, '' if res else ' not'))
# choice must be 'N'
break
if self.getOption('always') and new_text != original_text:
try:
page.text = new_text
page.save(summary=self.generate_summary(applied),
callback=self._count_changes, quiet=True)
except pywikibot.EditConflict:
pywikibot.output(u'Skipping {0!s} because of edit conflict'.format(page.title()))
except pywikibot.SpamfilterError as e:
pywikibot.output(
u'Cannot change {0!s} because of blacklist entry {1!s}'.format(page.title(), e.url))
except pywikibot.LockedPage:
pywikibot.output(u'Skipping {0!s} (locked page)'.format(page.title()))
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: {0!s}'.format(error.args))
if self._pending_processed_titles.qsize() > 50:
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output('Page {0!s}{1!s} saved'.format(proc_title, '' if res else ' not'))
def run(self):
"""Start the bot."""
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
if self.isTitleExcepted(page.title()):
pywikibot.output(
u'Skipping %s because the title is on the exceptions list.'
% page.title(asLink=True))
continue
try:
# Load the page's text from the wiki
original_text = page.get(get_redirect=True)
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
applied = set()
new_text = original_text
while True:
if self.isTextExcepted(new_text):
pywikibot.output(u'Skipping %s because it contains text '
u'that is on the exceptions list.'
% page.title(asLink=True))
break
last_text = None
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied)
if not self.recursive:
break
if new_text == original_text:
pywikibot.output(u'No changes were necessary in %s'
% page.title(asLink=True))
break
if hasattr(self, "addedCat"):
cats = page.categories(nofollow_redirects=True)
if self.addedCat not in cats:
cats.append(self.addedCat)
new_text = textlib.replaceCategoryLinks(new_text,
cats,
site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
pywikibot.output(u"\n\n>>> \03{lightpurple}%s\03{default} <<<"
% page.title())
pywikibot.showDiff(original_text, new_text)
if self.acceptall:
break
choice = pywikibot.input_choice(
u'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('Edit', 'e'),
('open in Browser', 'b'), ('all', 'a')],
default='N')
if choice == 'e':
editor = editarticle.TextEditor()
as_edited = editor.edit(original_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
continue
if choice == 'b':
webbrowser.open("http://%s%s" % (
page.site.hostname(),
page.site.nice_get_address(page.title(asUrl=True))
))
i18n.input('pywikibot-enter-finished-browser')
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output(u'Page %s has been deleted.'
% page.title())
break
new_text = original_text
continue
if choice == 'a':
self.acceptall = True
if choice == 'y':
page.put_async(new_text, self.generate_summary(applied), callback=self.count_changes)
# choice must be 'N'
break
if self.acceptall and new_text != original_text:
try:
page.put(new_text, self.generate_summary(applied), callback=self.count_changes)
except pywikibot.EditConflict:
pywikibot.output(u'Skipping %s because of edit conflict'
% (page.title(),))
except pywikibot.SpamfilterError as e:
pywikibot.output(
u'Cannot change %s because of blacklist entry %s'
% (page.title(), e.url))
except pywikibot.LockedPage:
pywikibot.output(u'Skipping %s (locked page)'
% (page.title(),))
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: %s'
% (error.args,))
def run(self):
"""Start the bot."""
# Run the generator which will yield Pages which might need to be
# changed.
for page in self.generator:
if self.isTitleExcepted(page.title()):
pywikibot.output("Skipping %s because the title is on the exceptions list." % page.title(asLink=True))
continue
try:
# Load the page's text from the wiki
original_text = page.get(get_redirect=True)
if not page.canBeEdited():
pywikibot.output("You can't edit page %s" % page.title(asLink=True))
continue
except pywikibot.NoPage:
pywikibot.output("Page %s not found" % page.title(asLink=True))
continue
applied = set()
new_text = original_text
while True:
if self.isTextExcepted(new_text):
pywikibot.output(
"Skipping %s because it contains text "
"that is on the exceptions list." % page.title(asLink=True)
)
break
last_text = None
while new_text != last_text:
last_text = new_text
new_text = self.apply_replacements(last_text, applied, page)
if not self.recursive:
break
if new_text == original_text:
pywikibot.output("No changes were necessary in %s" % page.title(asLink=True))
break
if hasattr(self, "addedCat"):
# Fetch only categories in wikitext, otherwise the others will
# be explicitly added.
cats = textlib.getCategoryLinks(new_text, site=page.site)
if self.addedCat not in cats:
cats.append(self.addedCat)
new_text = textlib.replaceCategoryLinks(new_text, cats, site=page.site)
# Show the title of the page we're working on.
# Highlight the title in purple.
pywikibot.output(color_format("\n\n>>> {lightpurple}{0}{default} <<<", page.title()))
pywikibot.showDiff(original_text, new_text)
if self.getOption("always"):
break
choice = pywikibot.input_choice(
"Do you want to accept these changes?",
[("Yes", "y"), ("No", "n"), ("Edit", "e"), ("open in Browser", "b"), ("all", "a")],
default="N",
)
if choice == "e":
editor = editarticle.TextEditor()
as_edited = editor.edit(original_text)
# if user didn't press Cancel
if as_edited and as_edited != new_text:
new_text = as_edited
continue
if choice == "b":
pywikibot.bot.open_webbrowser(page)
try:
original_text = page.get(get_redirect=True, force=True)
except pywikibot.NoPage:
pywikibot.output("Page %s has been deleted." % page.title())
break
new_text = original_text
continue
if choice == "a":
self.options["always"] = True
if choice == "y":
page.text = new_text
page.save(
summary=self.generate_summary(applied), async=True, callback=self._count_changes, quiet=True
)
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output("Page %s%s saved" % (proc_title, "" if res else " not"))
# choice must be 'N'
break
if self.getOption("always") and new_text != original_text:
try:
page.text = new_text
page.save(summary=self.generate_summary(applied), callback=self._count_changes, quiet=True)
except pywikibot.EditConflict:
pywikibot.output("Skipping %s because of edit conflict" % (page.title(),))
except pywikibot.SpamfilterError as e:
pywikibot.output("Cannot change %s because of blacklist entry %s" % (page.title(), e.url))
except pywikibot.LockedPage:
pywikibot.output("Skipping %s (locked page)" % (page.title(),))
except pywikibot.PageNotSaved as error:
pywikibot.output("Error putting page: %s" % (error.args,))
if self._pending_processed_titles.qsize() > 50:
while not self._pending_processed_titles.empty():
proc_title, res = self._pending_processed_titles.get()
pywikibot.output("Page %s%s saved" % (proc_title, "" if res else " not"))
def test_standard_links(self):
"""Test getting and replacing categories."""
cats = textlib.getCategoryLinks(self.old, site=self.site)
new = textlib.replaceCategoryLinks(self.old, cats, site=self.site)
self.assertEqual(self.old, new)
def add_text(
page,
addText: str,
summary: Optional[str] = None,
regexSkip: Optional[str] = None,
regexSkipUrl: Optional[str] = None,
always: bool = False,
up: bool = False,
putText: bool = True,
oldTextGiven: Optional[str] = None,
reorderEnabled: bool = True,
create: bool = False
) -> Union[Tuple[bool, bool, bool], Tuple[str, str, bool]]:
"""
Add text to a page.
@param page: The page to add text to
@type page: pywikibot.page.BasePage
@param addText: Text to add
@param summary: Summary of changes. If None, beginning of addText is used.
@param regexSkip: Abort if text on page matches
@param regexSkipUrl: Abort if full url matches
@param always: Always add text without user confirmation
@param up: If True, add text to top of page, else add at bottom.
@param putText: If True, save changes to the page, else return
(_, newtext, _)
@param oldTextGiven: If None fetch page text, else use this text
@param reorderEnabled: If True place text above categories and
interwiki, else place at page bottom. No effect if up = False.
@param create: Create page if it does not exist
@return: If putText=True: (success, success, always)
else: (_, newtext, _)
"""
site = page.site
if not summary:
summary = i18n.twtranslate(site, 'add_text-adding',
{'adding': addText[:200]})
if putText:
pywikibot.output('Loading {}...'.format(page.title()))
text = get_text(page, oldTextGiven, create)
if text is None:
return (False, False, always)
# 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 = page.full_url()
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output('Exception! regex (or word) used with -exceptUrl '
'is in the page. Skip!\n'
'Match was: {}'.format(result))
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output('Exception! regex (or word) used with -except '
'is in the page. Skip!\n'
'Match was: {}'.format(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', '\n')
if reorderEnabled:
# Getting the categories
categoriesInside = textlib.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = textlib.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = textlib.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = textlib.removeLanguageLinks(newtext, site)
# Adding the text
newtext += '\n' + addText
# Reputting the categories
newtext = textlib.replaceCategoryLinks(newtext, categoriesInside,
site, True)
# Adding the interwiki
newtext = textlib.replaceLanguageLinks(newtext, interwikiInside,
site)
else:
newtext += '\n' + addText
else:
newtext = addText + '\n' + text
if not putText:
# If someone load it as module, maybe it's not so useful to put the
# text in the page
return (text, newtext, always)
if text != newtext:
pywikibot.output(
color_format('\n\n>>> {lightpurple}{0}{default} <<<',
page.title()))
pywikibot.showDiff(text, newtext)
# Let's put the changes.
error_count = 0
while True:
if not always:
try:
choice = pywikibot.input_choice(
'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('All', 'a'),
('open in Browser', 'b')], 'n')
except QuitKeyboardInterrupt:
sys.exit('User quit bot run.')
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
pywikibot.bot.open_webbrowser(page)
continue
# either always or choice == 'y' is selected
result = put_text(page,
newtext,
summary,
error_count,
asynchronous=not always)
if result is not None:
return (result, result, always)
error_count += 1
def apply(self, text, page):
categories = textlib.getCategoryLinks(text)
if len(categories) > len(set(categories)):
deduplicate(categories)
text = textlib.replaceCategoryLinks(text, categories, page.site)
return text
def add_text(page,
addText,
summary=None,
regexSkip=None,
regexSkipUrl=None,
always=False,
up=False,
putText=True,
oldTextGiven=None,
reorderEnabled=True,
create=False):
"""
Add text to a page.
@rtype: tuple of (text, newtext, always)
"""
site = page.site
if not summary:
summary = i18n.twtranslate(site, '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
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 = page.full_url()
result = re.findall(regexSkipUrl, site.getUrl(url))
if result != []:
pywikibot.output('Exception! regex (or word) used with -exceptUrl '
'is in the page. Skip!\n'
'Match was: %s' % result)
return (False, False, always)
if regexSkip is not None:
result = re.findall(regexSkip, text)
if result != []:
pywikibot.output('Exception! regex (or word) used with -except '
'is in the page. Skip!\n'
'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 = textlib.getCategoryLinks(newtext, site)
# Deleting the categories
newtext = textlib.removeCategoryLinks(newtext, site)
# Getting the interwiki
interwikiInside = textlib.getLanguageLinks(newtext, site)
# Removing the interwiki
newtext = textlib.removeLanguageLinks(newtext, site)
# Adding the text
newtext += u"%s%s" % (config.line_separator, addText)
# Reputting the categories
newtext = textlib.replaceCategoryLinks(newtext, categoriesInside,
site, True)
# Dealing the stars' issue
allstars = []
starstext = textlib.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 = textlib.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.input_choice(
u'Do you want to accept these changes?',
[('Yes', 'y'), ('No', 'n'), ('All', 'a'),
('open in Browser', 'b')],
'n',
automatic_quit=False)
if choice == 'a':
always = True
elif choice == 'n':
return (False, False, always)
elif choice == 'b':
pywikibot.bot.open_webbrowser(page)
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.max_retries:
pywikibot.output(u'Server Error! Wait..')
time.sleep(config.retry_wait)
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.LockedPage:
pywikibot.output(u'Skipping %s (locked page)' %
page.title())
return (False, False, always)
except pywikibot.PageNotSaved as error:
pywikibot.output(u'Error putting page: %s' % error.args)
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 test_standard_links(self):
cats = textlib.getCategoryLinks(self.old, site=self.site)
new = textlib.replaceCategoryLinks(self.old, cats, site=self.site)
self.assertEqual(self.old, new)