mirror of
https://github.com/simon987/hexlib.git
synced 2025-04-17 17:16:41 +00:00
normalize dashes in preprocess
This commit is contained in:
parent
8d8f9e8751
commit
e1537297d7
@ -59,6 +59,9 @@ def _transform_trigram(ngram_seq, ngrams):
|
|||||||
SINGLE_QUOTES = ("’", "`")
|
SINGLE_QUOTES = ("’", "`")
|
||||||
SINGLE_QUOTE_TRANS = str.maketrans("".join(SINGLE_QUOTES), "".join(repeat("'", len(SINGLE_QUOTES))))
|
SINGLE_QUOTE_TRANS = str.maketrans("".join(SINGLE_QUOTES), "".join(repeat("'", len(SINGLE_QUOTES))))
|
||||||
|
|
||||||
|
DASHES = ("–", "⸺", "–")
|
||||||
|
DASHES_TRANS = str.maketrans("".join(DASHES), "".join(repeat("-", len(DASHES))))
|
||||||
|
|
||||||
PUNCTUATION = ".,;:\"!?/()|*=>"
|
PUNCTUATION = ".,;:\"!?/()|*=>"
|
||||||
PUNCTUATION_TRANS = str.maketrans(PUNCTUATION, " " * len(PUNCTUATION))
|
PUNCTUATION_TRANS = str.maketrans(PUNCTUATION, " " * len(PUNCTUATION))
|
||||||
|
|
||||||
@ -72,6 +75,8 @@ def preprocess(text, lowercase=False, clean_html=False, remove_punctuation=False
|
|||||||
if fix_single_quotes:
|
if fix_single_quotes:
|
||||||
text = text.translate(SINGLE_QUOTE_TRANS)
|
text = text.translate(SINGLE_QUOTE_TRANS)
|
||||||
|
|
||||||
|
text = text.translate(DASHES_TRANS)
|
||||||
|
|
||||||
if remove_urls:
|
if remove_urls:
|
||||||
text = LINK_RE.sub(" ", text)
|
text = LINK_RE.sub(" ", text)
|
||||||
|
|
||||||
|
2
setup.py
2
setup.py
@ -2,7 +2,7 @@ from setuptools import setup
|
|||||||
|
|
||||||
setup(
|
setup(
|
||||||
name="hexlib",
|
name="hexlib",
|
||||||
version="1.66",
|
version="1.67",
|
||||||
description="Misc utility methods",
|
description="Misc utility methods",
|
||||||
author="simon987",
|
author="simon987",
|
||||||
author_email="me@simon987.net",
|
author_email="me@simon987.net",
|
||||||
|
Loading…
x
Reference in New Issue
Block a user