Show More
@@ -1,244 +1,246 | |||
|
1 | 1 | #!/usr/bin/env python |
|
2 | 2 | # |
|
3 | 3 | # check-translation.py - check Mercurial specific translation problems |
|
4 | 4 | |
|
5 | 5 | import polib |
|
6 | 6 | import re |
|
7 | 7 | |
|
8 | 8 | scanners = [] |
|
9 | 9 | checkers = [] |
|
10 | 10 | |
|
11 | 11 | def scanner(): |
|
12 | 12 | def decorator(func): |
|
13 | 13 | scanners.append(func) |
|
14 | 14 | return func |
|
15 | 15 | return decorator |
|
16 | 16 | |
|
17 | 17 | def levelchecker(level, msgidpat): |
|
18 | 18 | def decorator(func): |
|
19 | 19 | if msgidpat: |
|
20 | 20 | match = re.compile(msgidpat).search |
|
21 | 21 | else: |
|
22 | 22 | match = lambda msgid: True |
|
23 | 23 | checkers.append((func, level)) |
|
24 | 24 | func.match = match |
|
25 | 25 | return func |
|
26 | 26 | return decorator |
|
27 | 27 | |
|
28 | 28 | def match(checker, pe): |
|
29 | 29 | """Examine whether POEntry "pe" is target of specified checker or not |
|
30 | 30 | """ |
|
31 | 31 | if not checker.match(pe.msgid): |
|
32 | 32 | return |
|
33 | 33 | # examine suppression by translator comment |
|
34 | 34 | nochecker = 'no-%s-check' % checker.__name__ |
|
35 | 35 | for tc in pe.tcomment.split(): |
|
36 | 36 | if nochecker == tc: |
|
37 | 37 | return |
|
38 | 38 | return True |
|
39 | 39 | |
|
40 | 40 | #################### |
|
41 | 41 | |
|
42 | 42 | def fatalchecker(msgidpat=None): |
|
43 | 43 | return levelchecker('fatal', msgidpat) |
|
44 | 44 | |
|
45 | 45 | @fatalchecker(r'\$\$') |
|
46 | 46 | def promptchoice(pe): |
|
47 | 47 | """Check translation of the string given to "ui.promptchoice()" |
|
48 | 48 | |
|
49 | 49 | >>> pe = polib.POEntry( |
|
50 | 50 | ... msgid ='prompt$$missing &sep$$missing &$$followed by &none', |
|
51 | 51 | ... msgstr='prompt missing &sep$$missing amp$$followed by none&') |
|
52 | 52 | >>> match(promptchoice, pe) |
|
53 | 53 | True |
|
54 | 54 | >>> for e in promptchoice(pe): print e |
|
55 | 55 | number of choices differs between msgid and msgstr |
|
56 | 56 | msgstr has invalid choice missing '&' |
|
57 | 57 | msgstr has invalid '&' followed by none |
|
58 | 58 | """ |
|
59 | 59 | idchoices = [c.rstrip(' ') for c in pe.msgid.split('$$')[1:]] |
|
60 | 60 | strchoices = [c.rstrip(' ') for c in pe.msgstr.split('$$')[1:]] |
|
61 | 61 | |
|
62 | 62 | if len(idchoices) != len(strchoices): |
|
63 | 63 | yield "number of choices differs between msgid and msgstr" |
|
64 | 64 | |
|
65 | 65 | indices = [(c, c.find('&')) for c in strchoices] |
|
66 | 66 | if [c for c, i in indices if i == -1]: |
|
67 | 67 | yield "msgstr has invalid choice missing '&'" |
|
68 | 68 | if [c for c, i in indices if len(c) == i + 1]: |
|
69 | 69 | yield "msgstr has invalid '&' followed by none" |
|
70 | 70 | |
|
71 | 71 | deprecatedpe = None |
|
72 | 72 | @scanner() |
|
73 | 73 | def deprecatedsetup(pofile): |
|
74 |
pes = [p for p in pofile |
|
|
74 | pes = [p for p in pofile | |
|
75 | if ((p.msgid == 'DEPRECATED' or p.msgid == '(DEPRECATED)') and | |
|
76 | p.msgstr)] | |
|
75 | 77 | if len(pes): |
|
76 | 78 | global deprecatedpe |
|
77 | 79 | deprecatedpe = pes[0] |
|
78 | 80 | |
|
79 | 81 | @fatalchecker(r'\(DEPRECATED\)') |
|
80 | 82 | def deprecated(pe): |
|
81 | 83 | """Check for DEPRECATED |
|
82 | 84 | >>> ped = polib.POEntry( |
|
83 | 85 | ... msgid = 'DEPRECATED', |
|
84 | 86 | ... msgstr= 'DETACERPED') |
|
85 | 87 | >>> deprecatedsetup([ped]) |
|
86 | 88 | >>> pe = polib.POEntry( |
|
87 | 89 | ... msgid = 'Something (DEPRECATED)', |
|
88 | 90 | ... msgstr= 'something (DEPRECATED)') |
|
89 | 91 | >>> match(deprecated, pe) |
|
90 | 92 | True |
|
91 | 93 | >>> for e in deprecated(pe): print e |
|
92 | 94 | >>> pe = polib.POEntry( |
|
93 | 95 | ... msgid = 'Something (DEPRECATED)', |
|
94 | 96 | ... msgstr= 'something (DETACERPED)') |
|
95 | 97 | >>> match(deprecated, pe) |
|
96 | 98 | True |
|
97 | 99 | >>> for e in deprecated(pe): print e |
|
98 | 100 | >>> pe = polib.POEntry( |
|
99 | 101 | ... msgid = 'Something (DEPRECATED)', |
|
100 | 102 | ... msgstr= 'something') |
|
101 | 103 | >>> match(deprecated, pe) |
|
102 | 104 | True |
|
103 | 105 | >>> for e in deprecated(pe): print e |
|
104 | 106 | msgstr inconsistently translated (DEPRECATED) |
|
105 | 107 | >>> pe = polib.POEntry( |
|
106 | 108 | ... msgid = 'Something (DEPRECATED, foo bar)', |
|
107 | 109 | ... msgstr= 'something (DETACERPED, foo bar)') |
|
108 | 110 | >>> match(deprecated, pe) |
|
109 | 111 | """ |
|
110 | 112 | if not ('(DEPRECATED)' in pe.msgstr or |
|
111 |
(deprecatedpe and |
|
|
113 | (deprecatedpe and | |
|
112 | 114 | deprecatedpe.msgstr in pe.msgstr)): |
|
113 | 115 | yield "msgstr inconsistently translated (DEPRECATED)" |
|
114 | 116 | |
|
115 | 117 | #################### |
|
116 | 118 | |
|
117 | 119 | def warningchecker(msgidpat=None): |
|
118 | 120 | return levelchecker('warning', msgidpat) |
|
119 | 121 | |
|
120 | 122 | @warningchecker() |
|
121 | 123 | def taildoublecolons(pe): |
|
122 | 124 | """Check equality of tail '::'-ness between msgid and msgstr |
|
123 | 125 | |
|
124 | 126 | >>> pe = polib.POEntry( |
|
125 | 127 | ... msgid ='ends with ::', |
|
126 | 128 | ... msgstr='ends with ::') |
|
127 | 129 | >>> for e in taildoublecolons(pe): print e |
|
128 | 130 | >>> pe = polib.POEntry( |
|
129 | 131 | ... msgid ='ends with ::', |
|
130 | 132 | ... msgstr='ends without double-colons') |
|
131 | 133 | >>> for e in taildoublecolons(pe): print e |
|
132 | 134 | tail '::'-ness differs between msgid and msgstr |
|
133 | 135 | >>> pe = polib.POEntry( |
|
134 | 136 | ... msgid ='ends without double-colons', |
|
135 | 137 | ... msgstr='ends with ::') |
|
136 | 138 | >>> for e in taildoublecolons(pe): print e |
|
137 | 139 | tail '::'-ness differs between msgid and msgstr |
|
138 | 140 | """ |
|
139 | 141 | if pe.msgid.endswith('::') != pe.msgstr.endswith('::'): |
|
140 | 142 | yield "tail '::'-ness differs between msgid and msgstr" |
|
141 | 143 | |
|
142 | 144 | @warningchecker() |
|
143 | 145 | def indentation(pe): |
|
144 | 146 | """Check equality of initial indentation between msgid and msgstr |
|
145 | 147 | |
|
146 | 148 | This may report unexpected warning, because this doesn't aware |
|
147 | 149 | the syntax of rst document and the context of msgstr. |
|
148 | 150 | |
|
149 | 151 | >>> pe = polib.POEntry( |
|
150 | 152 | ... msgid =' indented text', |
|
151 | 153 | ... msgstr=' narrowed indentation') |
|
152 | 154 | >>> for e in indentation(pe): print e |
|
153 | 155 | initial indentation width differs betweeen msgid and msgstr |
|
154 | 156 | """ |
|
155 | 157 | idindent = len(pe.msgid) - len(pe.msgid.lstrip()) |
|
156 | 158 | strindent = len(pe.msgstr) - len(pe.msgstr.lstrip()) |
|
157 | 159 | if idindent != strindent: |
|
158 | 160 | yield "initial indentation width differs betweeen msgid and msgstr" |
|
159 | 161 | |
|
160 | 162 | #################### |
|
161 | 163 | |
|
162 | 164 | def check(pofile, fatal=True, warning=False): |
|
163 | 165 | targetlevel = { 'fatal': fatal, 'warning': warning } |
|
164 | 166 | targetcheckers = [(checker, level) |
|
165 | 167 | for checker, level in checkers |
|
166 | 168 | if targetlevel[level]] |
|
167 | 169 | if not targetcheckers: |
|
168 | 170 | return [] |
|
169 | 171 | |
|
170 | 172 | detected = [] |
|
171 | 173 | for checker in scanners: |
|
172 | 174 | checker(pofile) |
|
173 | 175 | for pe in pofile.translated_entries(): |
|
174 | 176 | errors = [] |
|
175 | 177 | for checker, level in targetcheckers: |
|
176 | 178 | if match(checker, pe): |
|
177 | 179 | errors.extend((level, checker.__name__, error) |
|
178 | 180 | for error in checker(pe)) |
|
179 | 181 | if errors: |
|
180 | 182 | detected.append((pe, errors)) |
|
181 | 183 | return detected |
|
182 | 184 | |
|
183 | 185 | ######################################## |
|
184 | 186 | |
|
185 | 187 | if __name__ == "__main__": |
|
186 | 188 | import sys |
|
187 | 189 | import optparse |
|
188 | 190 | |
|
189 | 191 | optparser = optparse.OptionParser("""%prog [options] pofile ... |
|
190 | 192 | |
|
191 | 193 | This checks Mercurial specific translation problems in specified |
|
192 | 194 | '*.po' files. |
|
193 | 195 | |
|
194 | 196 | Each detected problems are shown in the format below:: |
|
195 | 197 | |
|
196 | 198 | filename:linenum:type(checker): problem detail ..... |
|
197 | 199 | |
|
198 | 200 | "type" is "fatal" or "warning". "checker" is the name of the function |
|
199 | 201 | detecting corresponded error. |
|
200 | 202 | |
|
201 | 203 | Checking by checker "foo" on the specific msgstr can be suppressed by |
|
202 | 204 | the "translator comment" like below. Multiple "no-xxxx-check" should |
|
203 | 205 | be separated by whitespaces:: |
|
204 | 206 | |
|
205 | 207 | # no-foo-check |
|
206 | 208 | msgid = "....." |
|
207 | 209 | msgstr = "....." |
|
208 | 210 | """) |
|
209 | 211 | optparser.add_option("", "--warning", |
|
210 | 212 | help="show also warning level problems", |
|
211 | 213 | action="store_true") |
|
212 | 214 | optparser.add_option("", "--doctest", |
|
213 | 215 | help="run doctest of this tool, instead of check", |
|
214 | 216 | action="store_true") |
|
215 | 217 | (options, args) = optparser.parse_args() |
|
216 | 218 | |
|
217 | 219 | if options.doctest: |
|
218 | 220 | import os |
|
219 | 221 | if 'TERM' in os.environ: |
|
220 | 222 | del os.environ['TERM'] |
|
221 | 223 | import doctest |
|
222 | 224 | failures, tests = doctest.testmod() |
|
223 | 225 | sys.exit(failures and 1 or 0) |
|
224 | 226 | |
|
225 | 227 | # replace polib._POFileParser to show linenum of problematic msgstr |
|
226 | 228 | class ExtPOFileParser(polib._POFileParser): |
|
227 | 229 | def process(self, symbol, linenum): |
|
228 | 230 | super(ExtPOFileParser, self).process(symbol, linenum) |
|
229 | 231 | if symbol == 'MS': # msgstr |
|
230 | 232 | self.current_entry.linenum = linenum |
|
231 | 233 | polib._POFileParser = ExtPOFileParser |
|
232 | 234 | |
|
233 | 235 | detected = [] |
|
234 | 236 | warning = options.warning |
|
235 | 237 | for f in args: |
|
236 | 238 | detected.extend((f, pe, errors) |
|
237 | 239 | for pe, errors in check(polib.pofile(f), |
|
238 | 240 | warning=warning)) |
|
239 | 241 | if detected: |
|
240 | 242 | for f, pe, errors in detected: |
|
241 | 243 | for level, checker, error in errors: |
|
242 | 244 | sys.stderr.write('%s:%d:%s(%s): %s\n' |
|
243 | 245 | % (f, pe.linenum, level, checker, error)) |
|
244 | 246 | sys.exit(1) |
General Comments 0
You need to be logged in to leave comments.
Login now