jonas/master-thesis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fixing tabstops

Browse Source
This commit is contained in:
Jonas Weinz 2019-06-30 15:20:19 +02:00
parent 3fbf5fad68
commit 9f0ef37d80
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
Tagging/conllu_generator.py
View File

@ -25,9 +25,10 @@ CONLLU_ATTRIBUTES = [
def replace_tab(s, tabstop=4): def replace_tab(s, tabstop=4):
result = str() result = str()
s = s.replace("\t", " \t")
for c in s: for c in s:
if c == '\t': if c == '\t':
while (len(result) % tabstop != 0): while (len(result) % (tabstop) != 0):
result += ' ' result += ' '
else: else:
result += c result += c
@ -151,7 +152,7 @@ class ConlluGenerator(object):
else: else:
stemmed_token = self.stemmer.stem(token) stemmed_token = self.stemmer.stem(token)
collnu_doc.add(ConlluElement( collnu_doc.add(ConlluElement(
id=self.id_counter, id=self.id_counter + 1,
form=token, form=token,
lemma=stemmed_token lemma=stemmed_token
)) ))