def is_wikisource_author_page(self, title):
"""Patrol a single item."""
if self.site.family.name != 'wikisource':
return False
author_ns = 0
with suppress(AttributeError, KeyError):
author_ns = self.site.family.authornamespaces[self.site.lang][0]
author_ns_prefix = self.site.namespace(author_ns) + ':'
author_page_name = removeprefix(title, author_ns_prefix)
if title != author_page_name:
verbose_output('Found author ' + author_page_name)
return True
return False
def process(self, text):
"""Process the page."""
# keys are ref groups
# values are a dict where :
# keys are ref content
# values are [name, [list of full ref matches],
# quoted, need_to_change]
found_refs = {}
found_ref_names = {}
# Replace key by [value, quoted]
named_repl = {}
for match in self.REFS.finditer(text):
content = match.group('content')
if not content.strip():
continue
params = match.group('params')
group = self.GROUPS.match(params)
if group not in found_refs:
found_refs[group] = {}
groupdict = found_refs[group]
if content in groupdict:
v = groupdict[content]
v[1].append(match.group())
else:
v = [None, [match.group()], False, False]
found = self.NAMES.match(params)
if found:
quoted = found.group('quote') in ['"', "'"]
name = found.group('name')
if v[0]:
if v[0] != name:
named_repl[name] = [v[0], v[2]]
else:
# First name associated with this content
if name == 'population':
pywikibot.output(content)
if name not in found_ref_names:
# first time ever we meet this name
if name == 'population':
pywikibot.output('in')
v[2] = quoted
v[0] = name
else:
# if has_key, means that this name is used
# with another content. We'll need to change it
v[3] = True
found_ref_names[name] = 1
groupdict[content] = v
used_numbers = set()
for name in found_ref_names:
number = removeprefix(name, self.autogen)
with suppress(ValueError):
used_numbers.add(int(number))
# iterator to give the next free number
free_number = iter({str(i) for i in range(1, 1000) # should be enough
if i not in used_numbers})
for (g, d) in found_refs.items():
group = ''
if g:
group = 'group="{}" '.format(group)
for (k, v) in d.items():
if len(v[1]) == 1 and not v[3]:
continue
name = v[0]
if not name:
name = '"{}{}"'.format(self.autogen, next(free_number))
elif v[2]:
name = '"{}"'.format(name)
named = '<ref {}name={}>{}</ref>'.format(group, name, k)
text = text.replace(v[1][0], named, 1)
# make sure that the first (named ref) is not
# removed later :
pos = text.index(named) + len(named)
header = text[:pos]
end = text[pos:]
unnamed = '<ref {}name={} />'.format(group, name)
for ref in v[1][1:]:
# Don't replace inside templates (T266411)
end = replaceExcept(end, re.escape(ref), unnamed,
exceptions=['template'])
text = header + end
for (k, v) in named_repl.items():
# TODO : Support ref groups
name = v[0]
if v[1]:
name = '"{}"'.format(name)
text = re.sub(
r'<ref name\s*=\s*(?P<quote>["\']?)\s*{}\s*(?P=quote)\s*/>'
.format(k),
'<ref name={} />'.format(name), text)
return text
def process(self, text):
"""Process the page."""
# keys are ref groups
# values are a dict where :
# keys are ref content
# values are [name, [list of full ref matches],
# quoted, need_to_change]
found_refs = {}
found_ref_names = set()
# Replace key by [value, quoted]
named_repl = {}
# Parse references
for match in self.REFS.finditer(text):
content = match.group('content')
if not content.strip():
continue
params = match.group('params')
group = self.GROUPS.search(params) or ''
if group not in found_refs:
found_refs[group] = {}
groupdict = found_refs[group]
if content in groupdict:
v = groupdict[content]
v[IX.reflist].append(match.group())
else:
v = [None, [match.group()], False, False]
found = self.NAMES.search(params)
if found:
quoted = found.group('quote') in ['"', "'"]
name = found.group('name')
if not v[IX.name]:
# First name associated with this content
if name not in found_ref_names:
# first time ever we meet this name
v[IX.quoted] = quoted
v[IX.name] = name
else:
# if has_key, means that this name is used
# with another content. We'll need to change it
v[IX.change_needed] = True
elif v[IX.name] != name:
named_repl[name] = [v[IX.name], v[IX.quoted]]
found_ref_names.add(name)
groupdict[content] = v
# Find used autogenerated numbers
used_numbers = set()
for name in found_ref_names:
number = removeprefix(name, self.autogen)
with suppress(ValueError):
used_numbers.add(int(number))
# generator to give the next free number for autogenerating names
free_number = (str(i) for i in itertools.count(start=1)
if i not in used_numbers)
# Fix references
for groupname, references in found_refs.items():
group = 'group="{}" '.format(groupname) if groupname else ''
for ref, v in references.items():
if len(v[IX.reflist]) == 1 and not v[IX.change_needed]:
continue
name = v[IX.name]
if not name:
name = '"{}{}"'.format(self.autogen, next(free_number))
elif v[IX.quoted]:
name = '"{}"'.format(name)
named = '<ref {}name={}>{}</ref>'.format(group, name, ref)
text = text.replace(v[IX.reflist][0], named, 1)
# make sure that the first (named ref) is not removed later
pos = text.index(named) + len(named)
header = text[:pos]
end = text[pos:]
# replace multiple identical references with repeated ref
repeated_ref = '<ref {}name={} />'.format(group, name)
for ref in v[IX.reflist][1:]:
# Don't replace inside templates (T266411)
end = replaceExcept(end, re.escape(ref), repeated_ref,
exceptions=['template'])
text = header + end
# Fix references with different names
for ref, v in named_repl.items():
# TODO : Support ref groups
name = v[IX.name]
if v[IX.reflist]:
name = '"{}"'.format(name)
text = re.sub(
r'<ref name\s*=\s*(?P<quote>["\']?)\s*{}\s*(?P=quote)\s*/>'
.format(ref),
'<ref name={} />'.format(name), text)
return text
