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

Teach wmllint about some of the new scopes that id= can live inside...

Browse source 
...without being a reference problem.  Refactor spellchecker, we want to
apply it more widely. Add proper line terminations to two files.
This commit is contained in:
Eric S. Raymond 2009-09-04 03:31:28 +00:00
parent 49ccfe0aaf
commit f9b7819412
3 changed files with 100 additions and 85 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
data/campaigns/Northern_Rebirth/utils/utils.cfg
View file

@ -86,4 +86,4 @@
#define STORY_PART_STALRAG SPEECH
{STORY_PART_SPEECH portraits/Stalrag.png (_ "Stalrag") {SPEECH} }
#enddef
#enddef

2
data/core/macros/carryover-utils.cfg
View file

@ -43,4 +43,4 @@ _"No gold carried over to the next scenario."#enddef
<span color='#00ff00'>" + {OBJECTIVE} + "</span> " + {EARLY_FINISH_BONUS_CAPTION}
condition=win
[/objective]
#enddef
#enddef

181
data/tools/wmllint
View file

@ -1021,6 +1021,8 @@ def global_sanity_check(filename, lines):
in_object = False
in_stage = False
in_cfg = False
in_goal = False
in_facet = False
in_sound_source = False
in_remove_sound_source = False
in_message = False
@ -1055,6 +1057,14 @@ def global_sanity_check(filename, lines):
in_cfg = True
elif "[/cfg]" in lines[i]:
in_cfg = False
elif "[goal]" in lines[i]:
in_goal = True
elif "[/goal]" in lines[i]:
in_goal = False
elif "[facet]" in lines[i]:
in_facet = True
elif "[/facet]" in lines[i]:
in_facet = False
elif "[sound_source]" in lines[i]:
in_sound_source = True
elif "[/sound_source]" in lines[i]:
@ -1142,7 +1152,7 @@ def global_sanity_check(filename, lines):
present.append(value)
elif value in ('narrator', 'unit', 'second_unit') or (value and value[0] in ("$", "{")):
continue
elif preamble_seen and not ignore_id and not in_object and not in_cfg and not in_sound_source and not in_remove_sound_source and not in_stage and not value in present:
elif preamble_seen and not ignore_id and not in_object and not in_cfg and not in_facet and not in_sound_source and not in_remove_sound_source and not in_stage and not in_goal and not value in present:
print '"%s", line %d: unknown \'%s\' referred to by id' \
% (filename, i+1, value)
if markcheck and has_tr_mark and not ("wmllint: ignore" in comment or "wmllint: noconvert" in comment):
@ -1730,6 +1740,92 @@ def translator(filename, mapxforms, textxform):
else:
return None
def inner_spellcheck(nav, value, spelldict):
"Spell-check an attribute value or string."
# Strip off translation marks
if value.startswith("_"):
value = value[1:].strip()
# Strip off line continuations, they interfere with string-stripping
value = value.strip()
if value.endswith("+"):
value = value[:-1].rstrip()
# Strip off string quotes
value = string_strip(value)
# Discard extraneous stuff
value = value.replace("...", " ")
value = value.replace("''", "")
value = value.replace("female^", " ")
value = value.replace("male^", " ")
value = value.replace("teamname^", " ")
if '<' in value:
value = re.sub("<[^>]+>text='([^']*)'<[^>]+>", r"\1", value)
value = re.sub("<[0-9,]+>", "", value)
# Fold continued lines
value = re.sub(r'" *\+\s*_? *"', "", value)
# It would be nice to use pyenchant's tokenizer here, but we can't
# because it wants to strip the trailing quotes we need to spot
# the Dwarvish-accent words.
for token in value.split():
# Try it with simple lowercasing first
lowered = token.lower()
if d.check(lowered):
continue
# Strip leading punctuation and grotty Wesnoth highlighters
while lowered and lowered[0] in " \t(`@*'%_":
lowered = lowered[1:]
# Not interested in interpolations or numeric literals
if not lowered or lowered.startswith("$"):
continue
# Suffix handling. Done in two passes because some
# Dwarvish dialect words end in a single quote
while lowered and lowered[-1] in "_-*).,:;?!& \t":
lowered = lowered[:-1]
if lowered and spelldict.check(lowered):
continue;
while lowered and lowered[-1] in "_-*').,:;?!& \t":
lowered = lowered[:-1]
# Not interested in interpolations or numeric literals
if not lowered or lowered.startswith("$") or lowered[0].isdigit():
continue
# Nuke balanced string quotes if present
lowered = string_strip(lowered)
if lowered and spelldict.check(lowered):
continue
# No match? Strip posessive suffixes and try again.
elif lowered.endswith("'s") and spelldict.check(lowered[:-2]):
continue
# Hyphenated compounds need all their parts good
if "-" in lowered:
parts = lowered.split("-")
if filter(lambda w: not w or spelldict.check(w), parts) == parts:
continue
# Modifier literals aren't interesting
if re.match("[+-][0-9]", lowered):
continue
# Match various onomatopoetic exclamations of variable form
if re.match("hm+", lowered):
continue
if re.match("a+[ur]*g+h*", lowered):
continue
if re.match("(mu)?ha(ha)*", lowered):
continue
if re.match("ah+", lowered):
continue
if re.match("no+", lowered):
continue
if re.match("no+", lowered):
continue
if re.match("um+", lowered):
continue
if re.match("aw+", lowered):
continue
if re.match("o+h+", lowered):
continue
if re.match("s+h+", lowered):
continue
nav.printError('possible misspelling "%s"' % token)
def spellcheck(fn, d):
"Spell-check a file using an Enchant dictionary object."
local_spellings = []
@ -1780,88 +1876,7 @@ def spellcheck(fn, d):
(key, prefix, value, comment) = parse_attribute(nav.text)
if "no spellcheck" in comment:
continue
# Strip off translation marks
if value.startswith("_"):
value = value[1:].strip()
# Strip off line continuations, they interfere with string-stripping
value = value.strip()
if value.endswith("+"):
value = value[:-1].rstrip()
# Strip off string quotes
value = string_strip(value)
# Discard extraneous stuff
value = value.replace("...", " ")
value = value.replace("''", "")
value = value.replace("female^", " ")
value = value.replace("male^", " ")
value = value.replace("teamname^", " ")
if '<' in value:
value = re.sub("<[^>]+>text='([^']*)'<[^>]+>", r"\1", value)
value = re.sub("<[0-9,]+>", "", value)
# Fold continued lines
value = re.sub(r'" *\+\s*_? *"', "", value)
# It would be nice to use pyenchant's tokenizer here, but we can't
# because it wants to strip the trailing quotes we need to spot
# the Dwarvish-accent words.
for token in value.split():
# Try it with simple lowercasing first
lowered = token.lower()
if d.check(lowered):
continue
# Strip leading punctuation and grotty Wesnoth highlighters
while lowered and lowered[0] in " \t(`@*'%_":
lowered = lowered[1:]
# Not interested in interpolations or numeric literals
if not lowered or lowered.startswith("$"):
continue
# Suffix handling. Done in two passes because some
# Dwarvish dialect words end in a single quote
while lowered and lowered[-1] in "_-*).,:;?!& \t":
lowered = lowered[:-1]
if lowered and d.check(lowered):
continue;
while lowered and lowered[-1] in "_-*').,:;?!& \t":
lowered = lowered[:-1]
# Not interested in interpolations or numeric literals
if not lowered or lowered.startswith("$") or lowered[0].isdigit():
continue
# Nuke balanced string quotes if present
lowered = string_strip(lowered)
if lowered and d.check(lowered):
continue
# No match? Strip posessive suffixes and try again.
elif lowered.endswith("'s") and d.check(lowered[:-2]):
continue
# Hyphenated compounds need all their parts good
if "-" in lowered:
parts = lowered.split("-")
if filter(lambda w: not w or d.check(w), parts) == parts:
continue
# Modifier literals aren't interesting
if re.match("[+-][0-9]", lowered):
continue
# Match various onomatopoetic exclamations of variable form
if re.match("hm+", lowered):
continue
if re.match("a+[ur]*g+h*", lowered):
continue
if re.match("(mu)?ha(ha)*", lowered):
continue
if re.match("ah+", lowered):
continue
if re.match("no+", lowered):
continue
if re.match("no+", lowered):
continue
if re.match("um+", lowered):
continue
if re.match("aw+", lowered):
continue
if re.match("o+h+", lowered):
continue
if re.match("s+h+", lowered):
continue
nav.printError('possible misspelling "%s"' % token)
inner_spellcheck(nav, value, d)
# Take exceptions from the id fields
if nav.element == "id=":
(key, prefix, value, comment) = parse_attribute(nav.text)