aboutsummaryrefslogtreecommitdiffstats
path: root/build/lib.linux-x86_64-2.7/dogtail/i18n.py
blob: 8117f8d82c6231934b530f29adb92cdefee4f582 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
# -*- coding: utf-8 -*-
"""
Internationalization facilities

Authors: David Malcolm <dmalcolm@redhat.com>
"""

__author__ = """David Malcolm <dmalcolm@redhat.com>, Zack Cerza <zcerza@redhat.com>"""

import config

import os
import re
import gettext

from logging import debugLogger as logger
from __builtin__ import unicode


def safeDecode(string):
    try:
        string = string.decode('utf-8', 'replace')
    except UnicodeEncodeError:
        string = string.encode('utf-8', 'replace')
    return string


def safeEncode(string):
    pass


"""
Singleton list of TranslationDb instances, to be initialized by the script with
whatever translation databases it wants.
"""
translationDbs = []


class TranslationDb(object):

    """
    Abstract base class representing a database of translations
    """

    def getTranslationsOf(self, srcName):
        """
        Pure virtual method to look up the translation of a string.
        Returns a list of candidate strings (the translation), empty if not found.

        Note that a source string can map to multiple translated strings. For
        example, in the French translation of Evolution, the string "Forward" can
        translate to both
        (i) "Faire suivre" for forwarding an email, and
        (ii) "Suivant" for the next page in a wizard.
        """
        raise NotImplementedError


class GettextTranslationDb(TranslationDb):

    """
    Implementation of TranslationDb which leverages gettext, using a single
    translation mo-file.
    """

    def __init__(self, moFile):
        self.__moFile = moFile
        self.__gnutranslations = gettext.GNUTranslations(open(moFile))

    def getTranslationsOf(self, srcName):
        srcName = safeDecode(srcName)
        # print "searching for translations of %s"%srcName
        # Use a dict to get uniqueness:
        results = {}
        result = self.__gnutranslations.ugettext(srcName)
        if result != srcName:
            results[result] = None

        # Hack alert:
        #
        # Note that typical UI definition in GTK etc contains strings with
        # underscores to denote accelerators.
        # For example, the stock GTK "Add" item has text "_Add" which e.g.
        # translates to "A_jouter" in French
        #
        # Since these underscores have been stripped out before we see these strings,
        # we are looking for a translation of "Add" into "Ajouter" in this case, so
        # we need to fake it, by looking up the string multiple times, with underscores
        # inserted in all possible positions, stripping underscores out of the result.
        # Ugly, but it works.

        for index in range(len(srcName)):
            candidate = srcName[:index] + "_" + srcName[index:]
            result = self.__gnutranslations.ugettext(candidate)
            if result != candidate:
                # Strip out the underscore, and add to the result:
                results[result.replace('_', '')] = True

        return results.keys()


def translate(srcString):
    """
    Look up srcString in the various translation databases (if any), returning
    a list of all matches found (potentially the empty list)
    """
    # Use a dict to get uniqueness:
    results = {}
    # Try to translate the string:
    for translationDb in translationDbs:
        for result in translationDb.getTranslationsOf(srcString):
            result = safeDecode(result)
            results[result] = True

    # No translations found:
    if len(results) == 0:
        if config.config.debugTranslation:
            logger.log('Translation not found for "%s"' % srcString)
    return results.keys()


class TranslatableString(object):

    """
    Class representing a string that we want to match strings against, handling
    translation for us, by looking it up once at construction time.
    """

    def __init__(self, untranslatedString):
        """
        Constructor looks up the string in all of the translation databases, storing
        the various translations it finds.
        """
        untranslatedString = safeDecode(untranslatedString)
        self.untranslatedString = untranslatedString
        self.translatedStrings = translate(untranslatedString)

    def matchedBy(self, string):
        """
        Compare the test string against either the translation of the original
        string (or simply the original string, if no translation was found).
        """
        # print "comparing %s against %s"%(string, self)
        def stringsMatch(inS, outS):
            """
            Compares a regular expression to a string

            inS: the regular expression (or normal string)
            outS: the normal string to be compared against
            """
            inString = str(inS)
            outString = outS
            if inString == outString:
                return True
            inString = inString + '$'
            inString = safeDecode(inString)
            outString = safeDecode(outString)
            if inString[0] == '*':
                inString = "\\" + inString
            # Escape all parentheses, since grouping will never be needed here
            inString = re.sub('([\(\)])', r'\\\1', inString)
            match = re.match(inString, outString)
            matched = match is not None
            return matched

        matched = False
        # the 'ts' variable keeps track of whether we're working with
        # translated strings. it's only used for debugging purposes.
        #ts = 0
        # print string, str(self)
        for translatedString in self.translatedStrings:
            #ts = ts + 1
            matched = stringsMatch(translatedString, string)
            if not matched:
                matched = translatedString == string
            if matched:
                return matched
        # ts=0
        return stringsMatch(self.untranslatedString, string)

    def __str__(self):
        """
        Provide a meaningful debug version of the string (and the translation in
        use)
        """
        if len(self.translatedStrings) > 0:
            # build an output string, with commas in the correct places
            translations = ""
            for tString in self.translatedStrings:
                translations += u'"%s", ' % safeDecode(tString)
            result = u'"%s" (%s)' % (
                safeDecode(self.untranslatedString), translations)
            return safeDecode(result)
        else:
            return '"%s"' % (self.untranslatedString)


def isMoFile(filename, language=''):
    """
    Does the given filename look like a gettext mo file?

    Optionally: Does the file also contain translations for a certain language,
    for example 'ja'?
    """
    if re.match('(.*)\\.mo$', filename):
        if not language:
            return True
        elif re.match('/usr/share/locale(.*)/%s(.*)/LC_MESSAGES/(.*)\\.mo$' %
                      language, filename):
            return True
        else:
            return False
    else:
        return False


def loadAllTranslationsForLanguage(language):
    import distro
    for moFile in distro.packageDb.getMoFiles(language):
        translationDbs.append(GettextTranslationDb(moFile))


def getMoFilesForPackage(packageName, language='', getDependencies=True):
    """
    Look up the named package and find all gettext mo files within it and its
    dependencies. It is possible to restrict the results to those of a certain
    language, for example 'ja'.
    """
    import distro

    result = []
    for filename in distro.packageDb.getFiles(packageName):
        if isMoFile(filename, language):
            result.append(filename)

    if getDependencies:
        # Recurse:
        for dep in distro.packageDb.getDependencies(packageName):
            # We pass False to the inner call because getDependencies has already
            # walked the full tree
            result.extend(getMoFilesForPackage(dep, language, False))

    return result


def loadTranslationsFromPackageMoFiles(packageName, getDependencies=True):
    """
    Helper function which appends all of the gettext translation mo-files used by
    the package (and its dependencies) to the translation database list.
    """
    # Keep a list of mo-files that are already in use to avoid duplicates.
    moFiles = {}

    def load(packageName, language='', getDependencies=True):
        for moFile in getMoFilesForPackage(packageName, language, getDependencies):
            # Searching the popt mo-files for translations makes gettext bail out,
            # so we ignore them here. This is
            # https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=172155 .
            if not('popt.mo' in moFile or moFile in moFiles):
                try:
                    translationDbs.append(GettextTranslationDb(moFile))
                    moFiles[moFile] = None
                except (AttributeError, IndexError):
                    if config.config.debugTranslation:
                        #import traceback
                        # logger.log(traceback.format_exc())
                        logger.log(
                            "Warning: Failed to load mo-file for translation: " + moFile)

    # Hack alert:
    #
    # The following special-case is necessary for Ubuntu, since their
    # translations are shipped in a single huge package. The downside to
    # this special case, aside from the simple fact that there is one,
    # is that it makes automatic translations much slower.

    import distro
    language = os.environ.get('LANGUAGE', os.environ['LANG'])[0:2]
    if isinstance(distro.distro, distro.Ubuntu):
        load('language-pack-gnome-%s' % language, language)
    load(packageName, language, getDependencies)