ozbolt/luscenje_struktur
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Removing old logging.debug calls, makes matching stuff much faster :)

Browse Source
This commit is contained in:
Ozbolt Menegatti 2019-06-02 14:03:29 +02:00
parent 2c8a9f0ed0
commit 08c8050f3f
1 changed files with 0 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
wani.py
View File

@ -634,7 +634,6 @@ class Component:
# if more than one match found for particular component # if more than one match found for particular component
elif len(cmatch) > 1: elif len(cmatch) > 1:
logging.debug("MULTIPLE: {}, {}".format(self.idx, cmatch))
# if more than one match in multiple components, NOPE! # if more than one match in multiple components, NOPE!
if len(to_ret) > 1: if len(to_ret) > 1:
logging.warning("Strange multiple match: {}".format( logging.warning("Strange multiple match: {}".format(
@ -652,7 +651,6 @@ class Component:
for tr in to_ret: for tr in to_ret:
tr.update(cmatch[0]) tr.update(cmatch[0])
logging.debug("MA: {}".format(str(to_ret)))
return to_ret return to_ret
def _match_self(self, word): def _match_self(self, word):
@ -667,8 +665,6 @@ class Component:
else: else:
matched = self.restriction.match(word) matched = self.restriction.match(word)
logging.debug("SELF MATCH({}: {} -> {}".format(self.idx, word.text, matched))
# recurse to next # recurse to next
if not matched: if not matched:
return None return None
@ -682,13 +678,11 @@ class Component:
# need to get all links that match # need to get all links that match
for next, link, order in self.next_element: for next, link, order in self.next_element:
next_links = word.get_links(link) next_links = word.get_links(link)
logging.debug("FIND LINKS FOR: {} -> {}: #{}".format(self.idx, next.idx, len(next_links)))
to_ret.append([]) to_ret.append([])
# good flag # good flag
good = next.status != ComponentStatus.Required good = next.status != ComponentStatus.Required
for next_word in next_links: for next_word in next_links:
logging.debug("link: {}: {} -> {}".format(link, word.id, next_word.id))
if not order.match(word, next_word): if not order.match(word, next_word):
continue continue
@ -707,7 +701,6 @@ class Component:
# if none matched, nothing found! # if none matched, nothing found!
if not good: if not good:
logging.debug("BAD")
return None return None
return to_ret return to_ret
@ -1208,7 +1201,6 @@ def match_file(words, structures):
for s in tqdm(structures): for s in tqdm(structures):
for w in words: for w in words:
mhere = s.match(w) mhere = s.match(w)
logging.debug(" GOT: {}".format(len(mhere)))
for match in mhere: for match in mhere:
colocation_id = [(idx, w.lemma) for idx, w in match.items()] colocation_id = [(idx, w.lemma) for idx, w in match.items()]
colocation_id = [s.id] + list(sorted(colocation_id, key=lambda x:x[0])) colocation_id = [s.id] + list(sorted(colocation_id, key=lambda x:x[0]))
@ -1284,9 +1276,6 @@ def main(input_file, structures_file, args):
if args.all: if args.all:
Writer.make_all_writer(args).write_out(structures, colocation_ids) Writer.make_all_writer(args).write_out(structures, colocation_ids)
logging.debug([(k, len(v)) for k, v in matches.items()])
logging.debug(sum(len(v) for _, v in matches.items()))
if __name__ == '__main__': if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Extract structures from a parsed corpus.') parser = argparse.ArgumentParser(description='Extract structures from a parsed corpus.')
parser.add_argument('structures', help='Structures definitions in xml file') parser.add_argument('structures', help='Structures definitions in xml file')