def __init__(self,
generator,
acceptall=False,
limit=None,
ignorepdf=False):
"""
- generator : Page generator
- acceptall : boolean, is -always on ?
- limit : int, stop after n modified pages
- ignorepdf : boolean
"""
self.generator = generator
self.acceptall = acceptall
self.limit = limit
self.ignorepdf = ignorepdf
self.site = pywikibot.getSite()
# Check
manual = 'mw:Manual:Pywikibot/refLinks'
code = None
for alt in [self.site.code] + i18n._altlang(self.site.code):
if alt in localized_msg:
code = alt
break
if code:
manual += '/%s' % code
self.msg = i18n.twtranslate(self.site, 'reflinks-msg', locals())
self.stopPage = pywikibot.Page(
self.site, pywikibot.translate(self.site, stopPage))
local = pywikibot.translate(self.site, badtitles)
if local:
bad = '(' + globalbadtitles + '|' + local + ')'
else:
bad = globalbadtitles
self.titleBlackList = re.compile(bad, re.I | re.S | re.X)
self.norefbot = noreferences.NoReferencesBot(None)
self.deduplicator = DuplicateReferences()
try:
self.stopPageRevId = self.stopPage.latestRevision()
except pywikibot.NoPage:
pywikibot.output(u'The stop page %s does not exist' %
self.stopPage.title(asLink=True))
raise
# Regex to grasp content-type meta HTML tag in HTML source
self.META_CONTENT = re.compile(ur'(?i)<meta[^>]*content\-type[^>]*>')
# Extract the encoding from a charset property (from content-type !)
self.CHARSET = re.compile(ur'(?i)charset\s*=\s*(?P<enc>[^\'";>/]*)')
# Extract html title from page
self.TITLE = re.compile(ur'(?is)(?<=<title>).*?(?=</title>)')
# Matches content inside <script>/<style>/HTML comments
self.NON_HTML = re.compile(
ur'(?is)<script[^>]*>.*?</script>|<style[^>]*>.*?</style>|<!--.*?-->|<!\[CDATA\[.*?\]\]>'
)
# Authorized mime types for HTML pages
self.MIME = re.compile(
ur'application/(?:xhtml\+xml|xml)|text/(?:ht|x)ml')
elif cols:
text = text.replace(m.group(), '{{reflist|%s}}' % cols.group(2))
else:
text = text.replace(m.group(), '{{reflist}}')
# If more than 30 refs, make sure the reference section is multi column
if text.count('</ref>') > 30:
text = re.sub(
r'(?is)(=\s+(<!--.*?-->)*\s*)(<references />|\{\{reflist\|?3?\}\})',
r'\1{{reflist|colwidth=30em}}', text)
elif text.count('</ref>') < 5:
text = re.sub(r'(?is)(=\s+)\{\{reflist\|(\d+|colwidth=\d+\w+)\}\}',
r'\1{{reflist}}', text)
if noreferences:
norefbot = noreferences.NoReferencesBot(None)
if norefbot.lacksReferences(text, verbose=False):
text = norefbot.addReferences(text)
return text
def test():
tests = ("""
see, <ref /> after,\t<ref > class,
<ref />
sdf her
<ref />
<ref />
, \t but would...
""", )