Fixed #6073 -- Made compilemessages 18n management command reject PO files with BOM.

git-svn-id: http://code.djangoproject.com/svn/django/trunk@14125 bcc190cf-cafb-0310-a4f2-bffc1f526a37

django/core/management/commands/compilemessages.py

@@ -1,9 +1,17 @@
+import codecs
 import os
 import sys
 from optparse import make_option
 from django.core.management.base import BaseCommand, CommandError
 
-def compile_messages(locale=None):
+def has_bom(fn):
+    f = open(fn, 'r')
+    sample = f.read(4)
+    return sample[:3] == '\xef\xbb\xbf' or \
+            sample.startswith(codecs.BOM_UTF16_LE) or \
+            sample.startswith(codecs.BOM_UTF16_BE)
+
+def compile_messages(stderr, locale=None):
     basedirs = [os.path.join('conf', 'locale'), 'locale']
     if os.environ.get('DJANGO_SETTINGS_MODULE'):
         from django.conf import settings
@@ -21,8 +29,11 @@ def compile_messages(locale=None):
         for dirpath, dirnames, filenames in os.walk(basedir):
             for f in filenames:
                 if f.endswith('.po'):
-                    sys.stderr.write('processing file %s in %s\n' % (f, dirpath))
+                    stderr.write('processing file %s in %s\n' % (f, dirpath))
-                    pf = os.path.splitext(os.path.join(dirpath, f))[0]
+                    fn = os.path.join(dirpath, f)
+                    if has_bom(fn):
+                        raise CommandError("The %s file has a BOM (Byte Order Mark). Django only supports .po files encoded in UTF-8 and without any BOM." % fn)
+                    pf = os.path.splitext(fn)[0]
                     # Store the names of the .mo and .po files in an environment
                     # variable, rather than doing a string replacement into the
                     # command, so that we can take advantage of shell quoting, to
@@ -49,4 +60,4 @@ class Command(BaseCommand):
 
     def handle(self, **options):
         locale = options.get('locale')
-        compile_messages(locale)
+        compile_messages(self.stderr, locale=locale)

docs/topics/i18n/localization.txt

@@ -188,6 +188,12 @@ That's it. Your translations are ready for use.
    ``django-admin compilemessages`` works see :ref:`gettext_on_windows` for more
    information.
 
+.. admonition:: .po files: Encoding and BOM usage.
+
+   Django only supports ``.po`` files encoded in UTF-8 and without any BOM
+   (Byte Order Mark) so if your text editor adds such marks to the beginning of
+   files by default then you will need to reconfigure it.
+
 .. _creating-message-files-from-js-code:
 
 Creating message files from JavaScript source code

tests/regressiontests/makemessages/compilation.py

@@ -0,0 +1,35 @@
+import os
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+from django.core.management import CommandError
+from django.core.management.commands.compilemessages import compile_messages
+from django.test import TestCase
+
+LOCALE='es_AR'
+
+
+class MessageCompilationTests(TestCase):
+
+    MO_FILE='locale/%s/LC_MESSAGES/django.mo' % LOCALE
+
+    def setUp(self):
+        self._cwd = os.getcwd()
+        self.test_dir = os.path.abspath(os.path.dirname(__file__))
+
+    def tearDown(self):
+        os.chdir(self._cwd)
+
+
+class PoFileTests(MessageCompilationTests):
+
+    def test_bom_rejection(self):
+        os.chdir(self.test_dir)
+        # We don't use the django.core.management intrastructure (call_command()
+        # et al) because CommandError's cause exit(1) there. We test the
+        # underlying compile_messages function instead
+        out = StringIO()
+        self.assertRaises(CommandError, compile_messages, out, locale=LOCALE)
+        self.failIf(os.path.exists(self.MO_FILE))

tests/regressiontests/makemessages/locale/es_AR/LC_MESSAGES/django.po

@@ -0,0 +1,20 @@
+# Copyright (C) YEAR THE PACKAGE'S COPYRIGHT HOLDER
+# This file is distributed under the same license as the PACKAGE package.
+# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR
+#
+msgid ""
+msgstr ""
+"Project-Id-Version: Django\n"
+"Report-Msgid-Bugs-To: \n"
+"POT-Creation-Date: 2010-01-05 20:25-0300\n"
+"PO-Revision-Date: 2010-02-22 00:14-0300\n"
+"Last-Translator: Translator <translator@example.com>\n"
+"Language-Team: Django-I18N <team@example.com>\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=UTF-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Plural-Forms:  nplurals=2; plural=(n != 1);\n"
+
+#: path/file.py:1
+msgid "This file has a UTF-8 BOM an should be rejected by the Django makemessages command."
+msgstr ""

tests/regressiontests/makemessages/tests.py

@@ -38,3 +38,6 @@ if xgettext_cmd:
         if xversion >= (0, 15):
             from extraction import *
     del p
+
+if find_command('msgfmt'):
+    from compilation import *