filters.py
from enchant.tokenize import Filter
def make_EdgecaseFilter(words):
class EdgecaseFilter(Filter):
"""
Some words might be special edgecase words that cannot really be
spellchecked and whose addition to the wordlist doesn't make sense.
Examples are hyphenated words, since the tokenization will split them
apart, which it should do, but which will sometimes lead to incorrect
spellchecks. Therefore, these have to be skipped it manually beforehand.
"""
def _skip(self, word):
if word in words:
return True
elif word.lower() in words:
return True
return EdgecaseFilter
class HTMLFilter(Filter):
"""
We don't want to check HTML entities.
"""
def _skip(self, word):
if "<" in word:
return True
class PythonFormatFilter(Filter):
"""
We need to filter the python-format-words such as %(redeemed)s so that they
don't get spell-checked.
"""
def _skip(self, word):
if word[:2] == "%(":
return True
if "{" in word:
return True
if word[:2] == "#%":
return True
return False
