Mercurial > babel > old > babel-test

annotate babel/catalog/pofile.py @ 26:710090104678

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
* Reduce size of locale data pickles by only storing the data provided by each locale itself, and merging inherited data at runtime. * Move locale data loading from `babel.core` into a separate `babel.localedata` module. * Add curency names and symbols to locale data.
author cmlenz
date Sun, 03 Jun 2007 15:27:27 +0000
parents 4fad20ab7cca
children 8521326b0627
rev   line source
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
1 # -*- coding: utf-8 -*-
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
2 #
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
3 # Copyright (C) 2007 Edgewall Software
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
4 # All rights reserved.
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
5 #
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
6 # This software is licensed as described in the file COPYING, which
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
7 # you should have received as part of this distribution. The terms
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
8 # are also available at http://babel.edgewall.org/wiki/License.
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
9 #
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
10 # This software consists of voluntary contributions made by many
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
11 # individuals. For the exact contribution history, see the revision
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
12 # history and logs, available at http://babel.edgewall.org/log/.
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
13
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
14 """Reading and writing of files in the ``gettext`` PO (portable object)
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
15 format.
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
16
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
17 :see: `The Format of PO Files
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
18 <http://www.gnu.org/software/gettext/manual/gettext.html#PO-Files>`_
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
19 """
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
20
5
50ad95bee876 * The creation-date header in generated PO files now includes the timezone offset.
cmlenz
parents: 1
diff changeset
21 from datetime import date, datetime
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
22 import re
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
23 try:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
24 set
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
25 except NameError:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
26 from sets import Set as set
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
27 import textwrap
5
50ad95bee876 * The creation-date header in generated PO files now includes the timezone offset.
cmlenz
parents: 1
diff changeset
28 import time
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
29
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
30 from babel import __version__ as VERSION
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
31
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
32 __all__ = ['escape', 'normalize', 'read_po', 'write_po']
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
33
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
34 def read_po(fileobj):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
35 """Read messages from a ``gettext`` PO (portable object) file from the given
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
36 file-like object.
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
37
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
38 This function yields tuples of the form:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
39
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
40 ``(message, translation, locations, flags)``
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
41
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
42 where:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
43
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
44 * ``message`` is the original (untranslated) message, or a
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
45 ``(singular, plural)`` tuple for pluralizable messages
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
46 * ``translation`` is the translation of the message, or a tuple of
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
47 translations for pluralizable messages
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
48 * ``locations`` is a sequence of ``(filename, lineno)`` tuples
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
49 * ``flags`` is a set of strings (for exampe, "fuzzy")
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
50
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
51 >>> from StringIO import StringIO
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
52 >>> buf = StringIO('''
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
53 ... #: main.py:1
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
54 ... #, fuzzy, python-format
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
55 ... msgid "foo %(name)s"
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
56 ... msgstr ""
21
ddfac856c34f Change pot header's first line, "Translations Template for %%(project)s." instead of "SOME DESCRIPTIVE TITLE.". '''`project`''' and '''`version`''' now default to '''PROJECT''' and '''VERSION''' respectively. Fixed a bug regarding '''Content-Transfer-Encoding''', it shouldn't be the charset, and we're defaulting to `8bit` untill someone complains.
palgarvio
parents: 17
diff changeset
57 ...
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
58 ... #: main.py:3
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
59 ... msgid "bar"
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
60 ... msgid_plural "baz"
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
61 ... msgstr[0] ""
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
62 ... msgstr[1] ""
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
63 ... ''')
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
64 >>> for message, translation, locations, flags in read_po(buf):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
65 ... print (message, translation)
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
66 ... print ' ', (locations, flags)
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
67 (('foo %(name)s',), ('',))
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
68 ((('main.py', 1),), set(['fuzzy', 'python-format']))
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
69 (('bar', 'baz'), ('', ''))
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
70 ((('main.py', 3),), set([]))
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
71
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
72 :param fileobj: the file-like object to read the PO file from
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
73 :return: an iterator over ``(message, translation, location)`` tuples
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
74 :rtype: ``iterator``
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
75 """
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
76 messages = []
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
77 translations = []
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
78 locations = []
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
79 flags = []
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
80 in_msgid = in_msgstr = False
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
81
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
82 def pack():
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
83 translations.sort()
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
84 retval = (tuple(messages), tuple([t[1] for t in translations]),
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
85 tuple(locations), set(flags))
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
86 del messages[:]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
87 del translations[:]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
88 del locations[:]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
89 del flags[:]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
90 return retval
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
91
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
92 for line in fileobj.readlines():
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
93 line = line.strip()
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
94 if line.startswith('#'):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
95 in_msgid = in_msgstr = False
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
96 if messages:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
97 yield pack()
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
98 if line[1:].startswith(':'):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
99 for location in line[2:].lstrip().split():
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
100 filename, lineno = location.split(':', 1)
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
101 locations.append((filename, int(lineno)))
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
102 elif line[1:].startswith(','):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
103 for flag in line[2:].lstrip().split(','):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
104 flags.append(flag.strip())
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
105 elif line:
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
106 if line.startswith('msgid_plural'):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
107 in_msgid = True
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
108 msg = line[12:].lstrip()
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
109 messages.append(msg[1:-1])
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
110 elif line.startswith('msgid'):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
111 in_msgid = True
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
112 if messages:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
113 yield pack()
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
114 msg = line[5:].lstrip()
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
115 messages.append(msg[1:-1])
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
116 elif line.startswith('msgstr'):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
117 in_msgid = False
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
118 in_msgstr = True
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
119 msg = line[6:].lstrip()
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
120 if msg.startswith('['):
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
121 idx, msg = msg[1:].split(']')
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
122 translations.append([int(idx), msg.lstrip()[1:-1]])
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
123 else:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
124 translations.append([0, msg[1:-1]])
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
125 elif line.startswith('"'):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
126 if in_msgid:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
127 messages[-1] += line.rstrip()[1:-1]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
128 elif in_msgstr:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
129 translations[-1][1] += line.rstrip()[1:-1]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
130
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
131 if messages:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
132 yield pack()
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
133
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
134 POT_HEADER = """\
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
135 # Translations Template for %%(project)s.
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
136 # Copyright (C) YEAR ORGANIZATION
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
137 # FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
138 #
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
139 msgid ""
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
140 msgstr ""
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
141 "Project-Id-Version: %%(project)s %%(version)s\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
142 "POT-Creation-Date: %%(creation_date)s\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
143 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
144 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
145 "Language-Team: LANGUAGE <LL@li.org>\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
146 "MIME-Version: 1.0\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
147 "Content-Type: text/plain; charset=%%(charset)s\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
148 "Content-Transfer-Encoding: 8bit\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
149 "Generated-By: Babel %s\\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
150
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
151 """ % VERSION
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
152
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
153 PYTHON_FORMAT = re.compile(r'\%(\([\w]+\))?[diouxXeEfFgGcrs]').search
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
154
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
155 WORD_SEP = re.compile('('
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
156 r'\s+|' # any whitespace
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
157 r'[^\s\w]*\w+[a-zA-Z]-(?=\w+[a-zA-Z])|' # hyphenated words
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
158 r'(?<=[\w\!\"\'\&\.\,\?])-{2,}(?=\w)' # em-dash
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
159 ')')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
160
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
161 def escape(string):
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
162 r"""Escape the given string so that it can be included in double-quoted
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
163 strings in ``PO`` files.
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
164
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
165 >>> escape('''Say:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
166 ... "hello, world!"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
167 ... ''')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
168 '"Say:\\n \\"hello, world!\\"\\n"'
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
169
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
170 :param string: the string to escape
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
171 :return: the escaped string
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
172 :rtype: `str` or `unicode`
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
173 """
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
174 return '"%s"' % string.replace('\\', '\\\\') \
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
175 .replace('\t', '\\t') \
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
176 .replace('\r', '\\r') \
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
177 .replace('\n', '\\n') \
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
178 .replace('\"', '\\"')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
179
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
180 def normalize(string, width=76):
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
181 r"""This converts a string into a format that is appropriate for .po files.
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
182
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
183 >>> print normalize('''Say:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
184 ... "hello, world!"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
185 ... ''', width=None)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
186 ""
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
187 "Say:\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
188 " \"hello, world!\"\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
189
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
190 >>> print normalize('''Say:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
191 ... "Lorem ipsum dolor sit amet, consectetur adipisicing elit, "
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
192 ... ''', width=32)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
193 ""
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
194 "Say:\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
195 " \"Lorem ipsum dolor sit "
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
196 "amet, consectetur adipisicing"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
197 " elit, \"\n"
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
198
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
199 :param string: the string to normalize
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
200 :param width: the maximum line width; use `None`, 0, or a negative number
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
201 to completely disable line wrapping
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
202 :return: the normalized string
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
203 :rtype: `unicode`
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
204 """
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
205 if width and width > 0:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
206 lines = []
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
207 for idx, line in enumerate(string.splitlines(True)):
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
208 if len(escape(line)) > width:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
209 chunks = WORD_SEP.split(line)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
210 chunks.reverse()
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
211 while chunks:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
212 buf = []
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
213 size = 2
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
214 while chunks:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
215 l = len(escape(chunks[-1])) - 2
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
216 if size + l < width:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
217 buf.append(chunks.pop())
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
218 size += l
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
219 else:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
220 if not buf:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
221 # handle long chunks by putting them on a
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
222 # separate line
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
223 buf.append(chunks.pop())
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
224 break
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
225 lines.append(u''.join(buf))
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
226 else:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
227 lines.append(line)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
228 else:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
229 lines = string.splitlines(True)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
230
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
231 if len(lines) == 1:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
232 return escape(string)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
233
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
234 # Remove empty trailing line
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
235 if not lines[-1]:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
236 del lines[-1]
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
237 lines[-1] += '\n'
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
238
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
239 return u'""\n' + u'\n'.join([escape(l) for l in lines])
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
240
23
bd45ff425485 Added line-wrap support for `write_po`.
palgarvio
parents: 21
diff changeset
241 def write_po(fileobj, messages, project='PROJECT', version='VERSION', width=76,
21
ddfac856c34f Change pot header's first line, "Translations Template for %%(project)s." instead of "SOME DESCRIPTIVE TITLE.". '''`project`''' and '''`version`''' now default to '''PROJECT''' and '''VERSION''' respectively. Fixed a bug regarding '''Content-Transfer-Encoding''', it shouldn't be the charset, and we're defaulting to `8bit` untill someone complains.
palgarvio
parents: 17
diff changeset
242 charset='utf-8', no_location=False, omit_header=False):
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
243 r"""Write a ``gettext`` PO (portable object) file to the given file-like
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
244 object.
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
245
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
246 The `messages` parameter is expected to be an iterable object producing
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
247 tuples of the form:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
248
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
249 ``(filename, lineno, funcname, message, flags)``
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
250
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
251 >>> from StringIO import StringIO
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
252 >>> buf = StringIO()
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
253 >>> write_po(buf, [
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
254 ... ('main.py', 1, None, u'foo %(name)s', ('fuzzy',)),
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
255 ... ('main.py', 3, 'ngettext', (u'bar', u'baz'), None)
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
256 ... ], omit_header=True)
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
257
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
258 >>> print buf.getvalue()
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
259 #: main.py:1
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
260 #, fuzzy, python-format
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
261 msgid "foo %(name)s"
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
262 msgstr ""
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
263 <BLANKLINE>
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
264 #: main.py:3
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
265 msgid "bar"
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
266 msgid_plural "baz"
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
267 msgstr[0] ""
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
268 msgstr[1] ""
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
269 <BLANKLINE>
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
270 <BLANKLINE>
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
271
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
272 :param fileobj: the file-like object to write to
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
273 :param messages: an iterable over the messages
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
274 :param project: the project name
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
275 :param version: the project version
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
276 :param width: the maximum line width for the generated output; use `None`,
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
277 0, or a negative number to completely disable line wrapping
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
278 :param charset: the encoding
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
279 :param no_location: do not emit a location comment for every message
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
280 :param omit_header: do not include the ``msgid ""`` entry at the top of the
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
281 output
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
282 """
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
283 def _normalize(key):
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
284 return normalize(key, width=width).encode(charset, 'backslashreplace')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
285
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
286 def _write(text):
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
287 if isinstance(text, unicode):
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
288 text = text.encode(charset)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
289 fileobj.write(text)
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
290
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
291 if not omit_header:
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
292 _write(POT_HEADER % {
5
50ad95bee876 * The creation-date header in generated PO files now includes the timezone offset.
cmlenz
parents: 1
diff changeset
293 'project': project,
50ad95bee876 * The creation-date header in generated PO files now includes the timezone offset.
cmlenz
parents: 1
diff changeset
294 'version': version,
50ad95bee876 * The creation-date header in generated PO files now includes the timezone offset.
cmlenz
parents: 1
diff changeset
295 'creation_date': time.strftime('%Y-%m-%d %H:%M%z'),
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
296 'charset': charset,
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
297 })
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
298
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
299 locations = {}
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
300 msgflags = {}
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
301 msgids = []
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
302
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
303 for filename, lineno, funcname, key, flags in messages:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
304 flags = set(flags or [])
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
305 if key in msgids:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
306 locations[key].append((filename, lineno))
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
307 msgflags[key] |= flags
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
308 else:
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
309 if (isinstance(key, (list, tuple)) and
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
310 filter(None, [PYTHON_FORMAT(k) for k in key])) or \
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
311 (isinstance(key, basestring) and PYTHON_FORMAT(key)):
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
312 flags.add('python-format')
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
313 else:
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
314 flags.discard('python-format')
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
315 locations[key] = [(filename, lineno)]
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
316 msgflags[key] = flags
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
317 msgids.append(key)
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
318
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
319 for msgid in msgids:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
320 if not no_location:
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
321 locs = u' '.join([u'%s:%d' % item for item in locations[msgid]])
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
322 if width and width > 0:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
323 locs = textwrap.wrap(locs, width, break_long_words=False)
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
324 for line in locs:
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
325 _write('#: %s\n' % line.strip())
6
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
326 flags = msgflags[msgid]
1801bc2b60ca Add basic PO file parsing, and change the PO writing procedure to also take flags (such as "python-format" or "fuzzy").
cmlenz
parents: 5
diff changeset
327 if flags:
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
328 _write('#%s\n' % ', '.join([''] + list(flags)))
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
329
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
330 if type(msgid) is tuple:
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
331 assert len(msgid) == 2
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
332 _write('msgid %s\n' % _normalize(msgid[0]))
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
333 _write('msgid_plural %s\n' % _normalize(msgid[1]))
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
334 _write('msgstr[0] ""\n')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
335 _write('msgstr[1] ""\n')
1
f71ca60f2a4a Import of initial code base.
cmlenz
parents:
diff changeset
336 else:
24
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
337 _write('msgid %s\n' % _normalize(msgid))
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
338 _write('msgstr ""\n')
4fad20ab7cca Reimplement line wrapping for PO writing (as the `textwrap` module is too destructive with white space) and move it to the `normalize` function (which was already doing some handling of line breaks).
cmlenz
parents: 23
diff changeset
339 _write('\n')
Copyright (C) 2012-2017 Edgewall Software