0ct0pu5/wesnoth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Refactoring step.

Browse source
This commit is contained in:
Eric S. Raymond 2008-11-03 03:04:00 +00:00
parent 3df044e3ff
commit 3863275e23
1 changed files with 99 additions and 95 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

194
data/tools/wmllint
View file

@ -1140,6 +1140,104 @@ def translator(filename, mapxforms, textxform):
else:
return None
def spellcheck(fn, d):
"Spell-check a file using an Enchant dictionary object."
local_spellings = []
# Accept declared spellings for this file
# and for all directories above it.
up = fn
while True:
if not up or up == os.sep:
break
else:
local_spellings += declared_spellings.get(up,[])
up = os.path.dirname(up)
map(d.add_to_session, local_spellings)
for nav in WmlIterator(filename=fn):
# Recognize local spelling exceptions
if not nav.element and "#" in nav.text:
comment = nav.text[nav.text.index("#")-1:]
words = re.search("wmllint: local spellings? (.*)", comment)
if words:
for word in words.group(1).split():
word = word.lower()
d.add_to_session(word)
local_spellings.append(word)
# Spell-check message and story parts
if nav.element in ("message=", "story=", "description="):
(key, prefix, value, comment) = parse_attribute(nav.text)
if "no spellcheck" in comment:
continue
if value.startswith("_"):
value = value[1:].strip()
value = string_strip(value)
value = value.replace("...", " ")
#value = value.replace("_ ", " ")
#value = value.replace(" _", " ")
value = value.replace("female^", " ")
value = value.replace("male^", " ")
if '<' in value:
value = re.sub("<[^>]+>text='([^']*)'<[^>]+>", r"\1", value)
value = re.sub("<[0-9,]+>", "", value)
value = value.lower()
value = re.sub(r'" *\+\s*_ *"', "", value)
for token in value.split():
if d.check(token):
continue
while token and token[0] in " \t(`@*'":
token = token[1:]
while token and token[-1] in "-*').,:;?!& \t":
token = token[:-1]
if token.startswith("$"):
continue
if token.startswith("%"):
continue
if token and token[0].isdigit():
continue
if not token:
continue
token = string_strip(token)
if token and d.check(token):
continue
elif token.endswith("'s") and d.check(token[:-2]):
continue
elif token.endswith("s'") and d.check(token[:-2]):
continue
if "-" in token:
parts = token.split("-")
if filter(lambda w: not w or d.check(w), parts) == parts:
continue
if re.match("[+-][0-9]", token):
continue
if re.match("hm+", token):
continue
if re.match("a+[ur]*g+h*", token):
continue
if re.match("(mu)?ha(ha)*", token):
continue
if re.match("ah+", token):
continue
if re.match("no+", token):
continue
if re.match("no+", token):
continue
if re.match("um+", token):
continue
if re.match("aw+", token):
continue
if re.match("o+h+", token):
continue
print nav.whereami(), 'possible misspelling "%s"' % token
# Take exceptions from name, id, and type fields
if nav.element in ("name=", "id=", "type="):
(key, prefix, value, comment) = parse_attribute(nav.text)
value = string_strip(value)
if value:
d.add_to_session(value)
local_spellings.append(value)
#for word in local_spellings:
# d.remove_from_session(word)
vctypes = (".svn", ".git")
def interesting(fn):
@ -1355,101 +1453,7 @@ if __name__ == '__main__':
for fn in allcfgfiles(dir):
if verbose >= 2:
print fn + ":"
# Accept declared spellings for this file
# and for all directories above it.
local_spellings = []
up = fn
while True:
if not up or up == os.sep:
break
else:
local_spellings += declared_spellings.get(up,[])
up = os.path.dirname(up)
map(d.add_to_session, local_spellings)
for nav in WmlIterator(filename=fn):
# Recognize local spelling exceptions
if not nav.element and "#" in nav.text:
comment = nav.text[nav.text.index("#")-1:]
words = re.search("wmllint: local spellings? (.*)", comment)
if words:
for word in words.group(1).split():
word = word.lower()
d.add_to_session(word)
local_spellings.append(word)
# Spell-check message and story parts
if nav.element in ("message=", "story=", "description="):
(key, prefix, value, comment) = parse_attribute(nav.text)
if "no spellcheck" in comment:
continue
if value.startswith("_"):
value = value[1:].strip()
value = string_strip(value)
value = value.replace("...", " ")
#value = value.replace("_ ", " ")
#value = value.replace(" _", " ")
value = value.replace("female^", " ")
value = value.replace("male^", " ")
if '<' in value:
value = re.sub("<[^>]+>text='([^']*)'<[^>]+>", r"\1", value)
value = re.sub("<[0-9,]+>", "", value)
value = value.lower()
value = re.sub(r'" *\+\s*_ *"', "", value)
for token in value.split():
if d.check(token):
continue
while token and token[0] in " \t(`@*'":
token = token[1:]
while token and token[-1] in "-*').,:;?!& \t":
token = token[:-1]
if token.startswith("$"):
continue
if token.startswith("%"):
continue
if token and token[0].isdigit():
continue
if not token:
continue
token = string_strip(token)
if token and d.check(token):
continue
elif token.endswith("'s") and d.check(token[:-2]):
continue
elif token.endswith("s'") and d.check(token[:-2]):
continue
if "-" in token:
parts = token.split("-")
if filter(lambda w: not w or d.check(w), parts) == parts:
continue
if re.match("[+-][0-9]", token):
continue
if re.match("hm+", token):
continue
if re.match("a+[ur]*g+h*", token):
continue
if re.match("(mu)?ha(ha)*", token):
continue
if re.match("ah+", token):
continue
if re.match("no+", token):
continue
if re.match("no+", token):
continue
if re.match("um+", token):
continue
if re.match("aw+", token):
continue
if re.match("o+h+", token):
continue
print nav.whereami(), 'possible misspelling "%s"' % token
# Take exceptions from name, id, and type fields
if nav.element in ("name=", "id=", "type="):
(key, prefix, value, comment) = parse_attribute(nav.text)
value = string_strip(value)
if value:
d.add_to_session(value)
local_spellings.append(value)
#for word in local_spellings:
# d.remove_from_session(word)
spellcheck(fn, d)
except ImportError:
sys.stderr.write("wmllint: spell check unavailable, install python-enchant to enable\n")
except KeyboardInterrupt: