def parseString(title=None, raw=None, wikidb=None, revision=None,
lang=None, magicwords=None, expandTemplates=True):
"""parse article with title from raw mediawiki text"""
uniquifier = None
siteinfo = None
assert title is not None, 'no title given'
if raw is None:
page = wikidb.normalize_and_get_page(title, 0)
if page:
raw = page.rawtext
else:
raw = None
assert raw is not None, "cannot get article %r" % (title,)
input_raw = raw
te = None
if wikidb:
if expandTemplates:
te = expander.Expander(raw, pagename=title, wikidb=wikidb)
input_raw = te.expandTemplates(True)
uniquifier = te.uniquifier
if hasattr(wikidb, 'get_siteinfo'):
siteinfo = wikidb.get_siteinfo()
src = None
if hasattr(wikidb, 'getSource'):
src = wikidb.getSource(title, revision=revision)
assert not isinstance(src, dict)
if not src:
src = metabook.source()
if lang is None:
lang = src.language
if magicwords is None:
if siteinfo is not None and 'magicwords' in siteinfo:
magicwords = siteinfo['magicwords']
else:
magicwords = src.get('magicwords')
if siteinfo is None:
nshandler = nshandling.get_nshandler_for_lang(lang)
else:
nshandler = nshandling.nshandler(siteinfo)
a = compat.parse_txt(input_raw, title=title, wikidb=wikidb, nshandler=nshandler, lang=lang, magicwords=magicwords, uniquifier=uniquifier, expander=te)
a.caption = title
if te and te.magic_displaytitle:
a.caption = te.magic_displaytitle
from mwlib.old_uparser import postprocessors
for x in postprocessors:
x(a, title=title, revision=revision, wikidb=wikidb, lang=lang)
return a
def parse_txt(txt, xopts=None, **kwargs):
if xopts is None:
xopts = XBunch(**kwargs)
else:
xopts.__dict__.update(**kwargs)
if xopts.expander is None:
from mwlib.expander import Expander, DictDB
xopts.expander = Expander("", "pagename", wikidb=DictDB())
if xopts.nshandler is None:
xopts.nshandler = nshandling.get_nshandler_for_lang(xopts.lang or 'en')
xopts.imagemod = util.ImageMod(xopts.magicwords)
uniquifier = xopts.uniquifier
if uniquifier is None:
uniquifier = uniq.Uniquifier()
txt = uniquifier.replace_tags(txt)
xopts.uniquifier = uniquifier
tokens = tokenize(txt, uniquifier=uniquifier)
td2 = tagparser()
a = td2.add
a("code", 10)
a("span", 20)
a("li", 25, blocknode=True, nested=False)
a("dl", 28, blocknode=True)
a("dt", 26, blocknode=True, nested=False)
a("dd", 26, blocknode=True, nested=True)
td1 = tagparser()
a = td1.add
a("blockquote", 5)
a("references", 15)
a("p", 30, blocknode=True, nested=False)
a("ul", 35, blocknode=True)
a("ol", 40, blocknode=True)
a("center", 45, blocknode=True)
td_parse_h = tagparser()
for i in range(1, 7):
td_parse_h.add("h%s" % i, i)
parsers = [
fixlitags, mark_style_tags, parse_singlequote, parse_preformatted, td2,
parse_paragraphs, td1, parse_lines, parse_div, parse_links, parse_urls,
parse_inputbox, td_parse_h, parse_sections, remove_table_garbage,
fix_tables, parse_tables, parse_uniq, fix_named_url_double_brackets,
fix_break_between_pre
]
combined_parser(parsers)(tokens, xopts)
return tokens
class nuwiki(object):
def __init__(self, path, allow_pickle=False):
self.path = os.path.abspath(path)
d = os.path.join(self.path, "images", "safe")
if not os.path.exists(d):
try:
os.makedirs(d)
except OSError, exc:
if exc.errno != 17: # file exists
raise
self.excluded = set(x.get("title") for x in self._loadjson("excluded.json", []))
self.revisions = {}
self._read_revisions()
fn = os.path.join(self.path, 'authors.db')
if not os.path.exists(fn):
self.authors = None
log.warn('no authors present. parsing revision info instead')
else:
self.authors = DumbJsonDB(fn, allow_pickle=allow_pickle)
fn = os.path.join(self.path, 'html.db')
if not os.path.exists(fn):
self.html = self.extractHTML(self._loadjson("parsed_html.json", {}))
log.warn('no html present. parsing revision info instead')
else:
self.html = DumbJsonDB(fn, allow_pickle=allow_pickle)
fn = os.path.join(self.path, 'imageinfo.db')
if not os.path.exists(fn):
self.imageinfo = self._loadjson("imageinfo.json", {})
log.warn('loading imageinfo from pickle')
else:
self.imageinfo = DumbJsonDB(fn, allow_pickle=allow_pickle)
self.redirects = self._loadjson("redirects.json", {})
self.siteinfo = self._loadjson("siteinfo.json", {})
self.nshandler = nshandling.nshandler(self.siteinfo)
self.en_nshandler = nshandling.get_nshandler_for_lang('en')
self.nfo = self._loadjson("nfo.json", {})
self.set_make_print_template()
def __init__(self, path, allow_pickle=False):
self.path = os.path.abspath(path)
d = os.path.join(self.path, "images", "safe")
if not os.path.exists(d):
try:
os.makedirs(d)
except OSError as exc:
if exc.errno != 17: # file exists
raise
self.excluded = set(x.get("title") for x in self._loadjson("excluded.json", []))
self.revisions = {}
self._read_revisions()
fn = os.path.join(self.path, 'authors.db')
if not os.path.exists(fn):
self.authors = None
log.warn('no authors present. parsing revision info instead')
else:
self.authors = DumbJsonDB(fn, allow_pickle=allow_pickle)
fn = os.path.join(self.path, 'html.db')
if not os.path.exists(fn):
self.html = self.extractHTML(self._loadjson("parsed_html.json", {}))
log.warn('no html present. parsing revision info instead')
else:
self.html = DumbJsonDB(fn, allow_pickle=allow_pickle)
fn = os.path.join(self.path, 'imageinfo.db')
if not os.path.exists(fn):
self.imageinfo = self._loadjson("imageinfo.json", {})
log.warn('loading imageinfo from pickle')
else:
self.imageinfo = DumbJsonDB(fn, allow_pickle=allow_pickle)
self.redirects = self._loadjson("redirects.json", {})
self.siteinfo = self._loadjson("siteinfo.json", {})
self.nshandler = nshandling.nshandler(self.siteinfo)
self.en_nshandler = nshandling.get_nshandler_for_lang('en')
self.nfo = self._loadjson("nfo.json", {})
self.set_make_print_template()
def empty():
empty = core.XBunch()
empty.nshandler = nshandling.get_nshandler_for_lang('de')
return empty
def empty():
empty = core.XBunch()
empty.nshandler = nshandling.get_nshandler_for_lang('de')
return empty
def _get_nshandler(self):
if self._nshandler is not None:
return self._nshandler
return nshandling.get_nshandler_for_lang('en') # FIXME
def parseString(title=None,
raw=None,
wikidb=None,
revision=None,
lang=None,
magicwords=None,
expandTemplates=True):
"""parse article with title from raw mediawiki text"""
uniquifier = None
siteinfo = None
assert title is not None, 'no title given'
if raw is None:
page = wikidb.normalize_and_get_page(title, 0)
if page:
raw = page.rawtext
else:
raw = None
assert raw is not None, "cannot get article %r" % (title, )
input = raw
te = None
if wikidb:
if expandTemplates:
te = expander.Expander(raw, pagename=title, wikidb=wikidb)
input = te.expandTemplates(True)
uniquifier = te.uniquifier
if hasattr(wikidb, 'get_siteinfo'):
siteinfo = wikidb.get_siteinfo()
src = None
if hasattr(wikidb, 'getSource'):
src = wikidb.getSource(title, revision=revision)
assert not isinstance(src, dict)
if not src:
src = metabook.source()
if lang is None:
lang = src.language
if magicwords is None:
if siteinfo is not None and 'magicwords' in siteinfo:
magicwords = siteinfo['magicwords']
else:
magicwords = src.get('magicwords')
if siteinfo is None:
nshandler = nshandling.get_nshandler_for_lang(lang)
else:
nshandler = nshandling.nshandler(siteinfo)
a = compat.parse_txt(input,
title=title,
wikidb=wikidb,
nshandler=nshandler,
lang=lang,
magicwords=magicwords,
uniquifier=uniquifier,
expander=te)
a.caption = title
if te and te.magic_displaytitle:
a.caption = te.magic_displaytitle
from mwlib.old_uparser import postprocessors
for x in postprocessors:
x(a, title=title, revision=revision, wikidb=wikidb, lang=lang)
return a
def parse_txt(txt, xopts=None, **kwargs):
if xopts is None:
xopts = XBunch(**kwargs)
else:
xopts.__dict__.update(**kwargs)
if xopts.expander is None:
from mwlib.expander import Expander, DictDB
xopts.expander = Expander("", "pagename", wikidb=DictDB())
if xopts.nshandler is None:
xopts.nshandler = nshandling.get_nshandler_for_lang(xopts.lang or 'en')
xopts.imagemod = util.ImageMod(xopts.magicwords)
uniquifier = xopts.uniquifier
if uniquifier is None:
uniquifier = uniq.Uniquifier()
txt = uniquifier.replace_tags(txt)
xopts.uniquifier = uniquifier
tokens = tokenize(txt, uniquifier=uniquifier)
td2 = tagparser()
a = td2.add
a("code", 10)
a("span", 20)
a("li", 25, blocknode=True, nested=False)
a("dl", 28, blocknode=True)
a("dt", 26, blocknode=True, nested=False)
a("dd", 26, blocknode=True, nested=True)
td1 = tagparser()
a = td1.add
a("blockquote", 5)
a("references", 15)
a("p", 30, blocknode=True, nested=False)
a("ul", 35, blocknode=True)
a("ol", 40, blocknode=True)
a("center", 45, blocknode=True)
td_parse_h = tagparser()
for i in range(1, 7):
td_parse_h.add("h%s" % i, i)
parsers = [fixlitags,
mark_style_tags,
parse_singlequote,
parse_preformatted,
td2,
parse_paragraphs,
td1,
parse_lines,
parse_div,
parse_links,
parse_urls,
parse_inputbox,
td_parse_h,
parse_sections,
remove_table_garbage,
fix_tables,
parse_tables,
parse_uniq,
fix_named_url_double_brackets,
fix_break_between_pre]
combined_parser(parsers)(tokens, xopts)
return tokens
def test_localized_redirect_matcher():
m = nshandling.get_nshandler_for_lang("de").redirect_matcher
assert m("#REDIRECT [[Data structure]]") == "Data structure", "bad redirect"
assert m("#WEITERLEITUNG [[Data structure]]") == "Data structure", "bad redirect"
def test_redirect_matcher():
m = nshandling.get_nshandler_for_lang("en").redirect_matcher
assert m("#REDIRECT [[Data structure#Active data structures]]") == "Data structure", "bad redirect"
def _get_nshandler(self):
if self._nshandler is not None:
return self._nshandler
return nshandling.get_nshandler_for_lang('en') # FIXME
def test_redirect_matcher():
m = nshandling.get_nshandler_for_lang("en").redirect_matcher
assert m("#REDIRECT [[Data structure#Active data structures]]"
) == "Data structure", "bad redirect"
def test_localized_redirect_matcher():
m = nshandling.get_nshandler_for_lang("de").redirect_matcher
assert m(
"#REDIRECT [[Data structure]]") == "Data structure", "bad redirect"
assert m("#WEITERLEITUNG [[Data structure]]"
) == "Data structure", "bad redirect"
