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(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_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 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(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 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."""
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 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 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
