def main(argv):
from utils import getfp
args = argv[1:] or ['-']
codec = 'utf-8'
for path in args:
print >>sys.stderr, path
(_,fp) = getfp(path)
parser = WikiTextParser()
parser.feed_file(fp, codec=codec)
parser.close()
fp.close()
def f(x, i=0):
if isinstance(x, WikiTree):
print (' '*i+'('+repr(x))
for c in x:
f(c, i+1)
print (' '*i+')')
elif isinstance(x, WikiToken):
print (' '*i+repr(x))
elif isinstance(x, XMLTagToken):
print (' '*i+repr(x))
elif isinstance(x, basestring):
print (' '*i+repr(x))
else:
assert 0, x
f(parser.get_root())
return
def __init__(self, output=None, pathpat=None, codec="utf-8", titleline=False, mode="page"):
assert output is not None or pathpat is not None
self.pathpat = pathpat
self.codec = codec
self.titleline = titleline
self.mode = mode
self._fp = None
if output is not None:
(_, self._fp) = getfp(output, mode="w")
self._pageid = None
self._title = None
self._revid = None
return
def main(argv):
from utils import getfp
class TitleExtractor(MWXMLDumpParser):
def start_revision(self, pageid, title, revid, timestamp):
print (pageid, title)
return
args = argv[1:] or ['-']
for path in args:
print >>sys.stderr, path
(_,fp) = getfp(path)
parser = TitleExtractor()
parser.feed_file(fp)
parser.close()
fp.close()
return 0
def add_data(self, pageid, revid, data):
assert self._pageid == pageid
assert self._title is not None
assert self._revid == revid
if self.pathpat is not None:
if self._fp is not None:
self._fp.close()
name = self._title.encode('utf-8').encode('quopri_codec')
path = (self.pathpat % {'name':name, 'pageid':pageid})
(_,self._fp) = getfp(path, 'w')
assert self._fp is not None
if self.titleline:
title = self._title.encode(self.codec, 'ignore')
self._fp.write(title+'\n')
self._fp.write(data.encode(self.codec, 'ignore'))
self._fp.write('\n\f')
return
def __init__(self,
output=None,
pathpat=None,
codec='utf-8',
titleline=False,
mode='page'):
assert output is not None or pathpat is not None
self.pathpat = pathpat
self.codec = codec
self.titleline = titleline
self.mode = mode
self._fp = None
if output is not None:
(_, self._fp) = getfp(output, mode='w')
self._pageid = None
self._title = None
self._revid = None
return
def add_data(self, pageid, revid, data):
assert self._pageid == pageid
assert self._title is not None
assert self._revid == revid
if self.pathpat is not None:
if self._fp is not None:
self._fp.close()
name = self._title.encode("utf-8").encode("quopri_codec")
path = self.pathpat % {"name": name, "pageid": pageid}
(_, self._fp) = getfp(path, "w")
assert self._fp is not None
if self.mode == "page":
if self.titleline:
title = self._title.encode(self.codec, "ignore")
self._fp.write(title + "\n")
self._fp.write(data.encode(self.codec, "ignore"))
self._fp.write("\n\f")
else:
self._fp.write(self._title.encode(self.codec, "ignore") + "\t")
self._fp.write(data.encode(self.codec, "ignore") + "\n")
return
def add_data(self, pageid, revid, data):
assert self._pageid == pageid
assert self._title is not None
assert self._revid == revid
if self.pathpat is not None:
if self._fp is not None:
self._fp.close()
name = self._title.encode('utf-8').encode('quopri_codec')
path = (self.pathpat % {'name': name, 'pageid': pageid})
(_, self._fp) = getfp(path, 'w')
assert self._fp is not None
if self.mode == 'page':
if self.titleline:
title = self._title.encode(self.codec, 'ignore')
self._fp.write(title + '\n')
self._fp.write(data.encode(self.codec, 'ignore'))
self._fp.write('\n\f')
else:
self._fp.write(self._title.encode(self.codec, 'ignore') + '\t')
self._fp.write(data.encode(self.codec, 'ignore') + '\n')
return
def main(argv):
from utils import getfp
class Tokenizer(WikiTextTokenizer):
def handle_text(self, pos, text):
print(pos, text)
return
def handle_token(self, pos, token):
print(pos, token)
return
args = argv[1:] or ["-"]
codec = "utf-8"
for path in args:
print >>sys.stderr, path
(_, fp) = getfp(path)
tokenizer = Tokenizer()
tokenizer.feed_file(fp, codec=codec)
tokenizer.close()
fp.close()
return
def main(argv):
from utils import getfp
class Tokenizer(WikiTextTokenizer):
def handle_text(self, pos, text):
print(pos, text)
return
def handle_token(self, pos, token):
print(pos, token)
return
args = argv[1:] or ['-']
codec = 'utf-8'
for path in args:
print >> sys.stderr, path
(_, fp) = getfp(path)
tokenizer = Tokenizer()
tokenizer.feed_file(fp, codec=codec)
tokenizer.close()
fp.close()
return
