Skip to content

E
edx-platform
Commit 6f103488 by Steve Strassmann
Options

addressed comments from pull request

parent 91bee1a9
Showing with 132 additions and 305 deletions
cms/djangoapps/contentstore/tests/test_i18n.py
# -*- coding: iso-8859-1 -*-
from unittest import skip
from django.test import TestCase
from django.core.urlresolvers import reverse
from django.contrib.auth.models import User
from django.test.client import Client
from nose.tools import nottest
class InternationalizationTest(TestCase):
from .utils import ModuleStoreTestCase
class InternationalizationTest(ModuleStoreTestCase):
"""
Tests to validate Internationalization.
"""
......@@ -52,6 +52,22 @@ class InternationalizationTest(TestCase):
status_code=200,
html=True)
def test_course_explicit_english(self):
"""Test viewing the index page with no courses"""
# Create a course so there is something to view
self.client = Client()
self.client.login(username=self.uname, password=self.password)
resp = self.client.get(reverse('index'),
{},
HTTP_ACCEPT_LANGUAGE='en'
)
self.assertContains(resp,
'<h1 class="title-1">My Courses</h1>',
status_code=200,
html=True)
# ****
# NOTE:
......@@ -62,7 +78,7 @@ class InternationalizationTest(TestCase):
# actual French at that time.
# Test temporarily disable since it depends on creation of dummy strings
@nottest
@skip
def test_course_with_accents (self):
"""Test viewing the index page with no courses"""
# Create a course so there is something to view
......@@ -75,7 +91,7 @@ class InternationalizationTest(TestCase):
)
TEST_STRING = u'<h1 class="title-1">' \
+ u'My Çöürsés L#' \
+ u'My \xc7\xf6\xfcrs\xe9s L#' \
+ u'</h1>'
self.assertContains(resp,
......
cms/envs/common.py
......@@ -128,6 +128,8 @@ MIDDLEWARE_CLASSES = (
'django.contrib.messages.middleware.MessageMiddleware',
'track.middleware.TrackMiddleware',
'mitxmako.middleware.MakoMiddleware',
# Detects user-requested locale from 'accept-language' header in http request
'django.middleware.locale.LocaleMiddleware',
'django.middleware.transaction.TransactionMiddleware'
......
cms/static/js/base.js
......@@ -826,11 +826,14 @@ function saveSetSectionScheduleDate(e) {
data: JSON.stringify({ 'id': id, 'metadata': {'start': start}})
}).success(function () {
var $thisSection = $('.courseware-section[data-id="' + id + '"]');
var format = gettext('<strong>Will Release:</strong> %(date)s at $(time)s UTC');
var willReleaseAt = interpolate(format, [input_date, input_time], true);
$thisSection.find('.section-published-date').html(
'<span class="published-status"><strong>' + gettext('Will Release:') +
'</strong> ' + input_date + ' at ' + input_time +
' UTC</span><a href="#" class="edit-button" data-date="' + input_date +
'" data-time="' + input_time + '" data-id="' + id + '">' +
'<span class="published-status">' + willReleaseAt + '</span>' +
'<a href="#" class="edit-button" ' +
'" data-date="' + input_date +
'" data-time="' + input_time +
'" data-id="' + id + '">' +
gettext('Edit') + '</a>');
$thisSection.find('.section-published-date').animate({
'background-color': 'rgb(182,37,104)'
......
i18n/converter.py
import re, itertools
# Converter is an abstract class that transforms strings.
# It hides embedded tags (HTML or Python sequences) from transformation
#
# To implement Converter, provide implementation for inner_convert_string()
import re
import itertools
class Converter:
"""Converter is an abstract class that transforms strings.
It hides embedded tags (HTML or Python sequences) from transformation
To implement Converter, provide implementation for inner_convert_string()
# matches tags like these:
# HTML: <B>, </B>, <BR/>, <textformat leading="10">
# Python: %(date)s, %(name)s
#
tag_pattern = re.compile(r'(<[-\w" .:?=/]*>)|({[^}]*})|(%\(.*\)\w)', re.I)
Strategy:
1. extract tags embedded in the string
a. use the index of each extracted tag to re-insert it later
b. replace tags in string with numbers (<0>, <1>, etc.)
c. save extracted tags in a separate list
2. convert string
3. re-insert the extracted tags
"""
# matches tags like these:
# HTML: <B>, </B>, <BR/>, <textformat leading="10">
# Python: %(date)s, %(name)s
tag_pattern = re.compile(r'(<[-\w" .:?=/]*>)|({[^}]*})|(%\([^)]*\)\w)', re.I)
def convert (self, string):
if self.tag_pattern.search(string):
result = self.convert_tagged_string(string)
else:
result = self.inner_convert_string(string)
return result
# convert_tagged_string(string):
# returns: a converted tagged string
# param: string (contains html tags)
#
# Don't replace characters inside tags
#
# Strategy:
# 1. extract tags embedded in the string
# a. use the index of each extracted tag to re-insert it later
# b. replace tags in string with numbers (<0>, <1>, etc.)
# c. save extracted tags in a separate list
# 2. convert string
# 3. re-insert the extracted tags
#
def convert_tagged_string (self, string):
def convert(self, string):
"""Returns: a converted tagged string
param: string (contains html tags)
Don't replace characters inside tags
"""
(string, tags) = self.detag_string(string)
string = self.inner_convert_string(string)
string = self.retag_string(string, tags)
return string
# extracts tags from string.
#
# returns (string, list) where
# string: string has tags replaced by indices (<BR>... => <0>, <1>, <2>, etc.)
# list: list of the removed tags ("<BR>", "<I>", "</I>")
def detag_string (self, string):
def detag_string(self, string):
"""Extracts tags from string.
returns (string, list) where
string: string has tags replaced by indices (<BR>... => <0>, <1>, <2>, etc.)
list: list of the removed tags ('<BR>', '<I>', '</I>')
"""
counter = itertools.count(0)
count = lambda m: '<%s>' % counter.next()
tags = self.tag_pattern.findall(string)
......@@ -57,9 +49,8 @@ class Converter:
raise Exception('tags dont match:'+string)
return (new, tags)
# substitutes each tag back into string, into occurrences of <0>, <1> etc
#
def retag_string (self, string, tags):
def retag_string(self, string, tags):
"""substitutes each tag back into string, into occurrences of <0>, <1> etc"""
for (i, tag) in enumerate(tags):
p = '<%s>' % i
string = re.sub(p, tag, string, 1)
......@@ -69,6 +60,6 @@ class Converter:
# ------------------------------
# Customize this in subclasses of Converter
def inner_convert_string (self, string):
def inner_convert_string(self, string):
return string # do nothing by default
i18n/dummy.py
# -*- coding: iso-8859-15 -*-
from converter import Converter
# This file converts string resource files.
# Java: file has name like messages_en.properties
# Flex: file has name like locales/en_US/Labels.properties
# Creates new localization properties files in a dummy language (saved as 'vr', Vardebedian)
# Creates new localization properties files in a dummy language
# Each property file is derived from the equivalent en_US file, except
# 1. Every vowel is replaced with an equivalent with extra accent marks
# 2. Every string is padded out to +30% length to simulate verbose languages (e.g. German)
......@@ -18,19 +12,18 @@ from converter import Converter
# Example use:
# >>> from dummy import Dummy
# >>> c = Dummy()
# >>> print c.convert("hello my name is Bond, James Bond")
# hll my nm s Bnd, Jms Bnd Lorem i#
# >>> c.convert("hello my name is Bond, James Bond")
# u'h\xe9ll\xf6 my n\xe4m\xe9 \xefs B\xf6nd, J\xe4m\xe9s B\xf6nd Lorem i#'
#
# >>> print c.convert('don\'t convert <a href="href">tag ids</a>')
# dn't nvrt <a href="href">tg ds</a> Lorem ipsu#
# >>> c.convert('don\'t convert <a href="href">tag ids</a>')
# u'd\xf6n\'t \xe7\xf6nv\xe9rt <a href="href">t\xe4g \xefds</a> Lorem ipsu#'
#
# >>> print c.convert('don\'t convert %(name)s tags on %(date)s')
# dn't nvrt %(name)s tags on %(date)s Lorem ips#
# >>> c.convert('don\'t convert %(name)s tags on %(date)s')
# u"d\xf6n't \xe7\xf6nv\xe9rt %(name)s t\xe4gs \xf6n %(date)s Lorem ips#"
# Substitute plain characters with accented lookalikes.
# http://tlt.its.psu.edu/suggestions/international/web/codehtml.html#accent
# print "print u'\\x%x'" % 207
TABLE = {'A': u'\xC0',
'a': u'\xE4',
'b': u'\xDF',
......@@ -62,23 +55,23 @@ PAD_FACTOR = 1.3
class Dummy (Converter):
'''
"""
A string converter that generates dummy strings with fake accents
and lorem ipsum padding.
'''
"""
def convert (self, string):
def convert(self, string):
result = Converter.convert(self, string)
return self.pad(result)
def inner_convert_string (self, string):
def inner_convert_string(self, string):
for (k,v) in TABLE.items():
string = string.replace(k, v)
return string
def pad (self, string):
'''add some lorem ipsum text to the end of string'''
def pad(self, string):
"""add some lorem ipsum text to the end of string"""
size = len(string)
if size < 7:
target = size*3
......@@ -86,15 +79,15 @@ class Dummy (Converter):
target = int(size*PAD_FACTOR)
return string + self.terminate(LOREM[:(target-size)])
def terminate (self, string):
'''replaces the final char of string with #'''
def terminate(self, string):
"""replaces the final char of string with #"""
return string[:-1]+'#'
def init_msgs (self, msgs):
'''
def init_msgs(self, msgs):
"""
Make sure the first msg in msgs has a plural property.
msgs is list of instances of pofile.Msg
'''
"""
if len(msgs)==0:
return
headers = msgs[0].get_property('msgstr')
......@@ -105,82 +98,35 @@ class Dummy (Converter):
headers.append(plural)
def convert_msg (self, msg):
'''
def convert_msg(self, msg):
"""
Takes one Msg object and converts it (adds a dummy translation to it)
msg is an instance of pofile.Msg
'''
source = msg.get_property('msgid')
if len(source)==1 and len(source[0])==0:
"""
source = msg.msgid
if len(source)==0:
# don't translate empty string
return
plural = msg.get_property('msgid_plural')
plural = msg.msgid_plural
if len(plural)>0:
# translate singular and plural
foreign_single = self.convert(merge(source))
foreign_plural = self.convert(merge(plural))
msg.set_property('msgstr[0]', split(foreign_single))
msg.set_property('msgstr[1]', split(foreign_plural))
foreign_single = self.convert(source)
foreign_plural = self.convert(plural)
plural = {'0': self.final_newline(source, foreign_single),
'1': self.final_newline(plural, foreign_plural)}
msg.msgstr_plural = plural
return
else:
src_merged = merge(source)
foreign = self.convert(src_merged)
if len(source)>1:
# If last char is a newline, make sure translation
# has a newline too.
if src_merged[-2:]=='\\n':
foreign += '\\n'
msg.set_property('msgstr', split(foreign))
# ----------------------------------
# String splitting utility functions
SPLIT_SIZE = 70
def merge (string_list):
'''returns a single string: concatenates string_list'''
return ''.join(string_list)
# .po file format requires long strings to be broken
# up into several shorter (<80 char) strings.
# The first string is empty (""), which indicates
# that more are to be read on following lines.
def split (string):
'''
Returns string split into fragments of a given size.
If there are multiple fragments, insert "" as the first fragment.
'''
result = [chunk for chunk in chunks(string, SPLIT_SIZE)]
if len(result)>1:
result = [''] + result
return result
def chunks(string, size):
'''
Generate fragments of a given size from string. Avoid breaking
the string in the middle of an escape sequence (e.g. "\n")
'''
strlen=len(string)-1
esc = False
last = 0
for i,char in enumerate(string):
if not esc and char == '\\':
esc = True
continue
if esc:
esc = False
if i>=last+size-1 or i==strlen:
chunk = string[last:i+1]
last = i+1
yield chunk
# testing
# >>> a = "abcd\\efghijklmnopqrstuvwxyz"
# >>> SPLIT_SIZE = 5
# >>> split(a)
# ['abcd\\e', 'fghij', 'klmno', 'pqrst', 'uvwxy', 'z']
# >>> merge(split(a))
# 'abcd\\efghijklmnopqrstuvwxyz'
foreign = self.convert(source)
msg.msgstr = self.final_newline(source, foreign)
def final_newline(self, original, translated):
""" Returns a new translated string.
If last char of original is a newline, make sure translation
has a newline too.
"""
if len(original)>1:
if original[-1]=='\n' and translated[-1]!='\n':
return translated + '\n'
return translated
i18n/make_dummy.py
......@@ -16,7 +16,7 @@
# mitx/conf/locale/vr/LC_MESSAGES/django.po
import os, sys
from pofile import PoFile
import polib
from dummy import Dummy
# Dummy language
......@@ -28,23 +28,26 @@ from dummy import Dummy
OUT_LANG = 'fr'
def main (file):
'''
def main(file):
"""
Takes a source po file, reads it, and writes out a new po file
containing a dummy translation.
'''
pofile = PoFile(file)
"""
if not os.path.exists(file):
raise IOError('File does not exist: %s' % file)
pofile = polib.pofile(file)
converter = Dummy()
converter.init_msgs(pofile.msgs)
for msg in pofile.msgs:
converter.init_msgs(pofile.translated_entries())
for msg in pofile:
converter.convert_msg(msg)
new_file = new_filename(file, OUT_LANG)
create_dir_if_necessary(new_file)
pofile.write(new_file)
pofile.save(new_file)
def new_filename (original_filename, new_lang):
'''Returns a filename derived from original_filename, using new_lang as the locale'''
def new_filename(original_filename, new_lang):
"""Returns a filename derived from original_filename, using new_lang as the locale"""
orig_dir = os.path.dirname(original_filename)
msgs_dir = os.path.basename(orig_dir)
orig_file = os.path.basename(original_filename)
......
i18n/pofile.py deleted 100644 → 0
import re, codecs
from operator import itemgetter
# Django stores externalized strings in .po and .mo files.
# po files are human readable and contain metadata about the strings.
# mo files are machine readable and optimized for runtime performance.
# See https://docs.djangoproject.com/en/1.3/topics/i18n/internationalization/
# See http://www.gnu.org/software/gettext/manual/html_node/PO-Files.html
# Usage:
# >>> pofile = PoFile('/path/to/file')
class PoFile:
# Django requires po files to be in UTF8 with no BOM (byte order marker)
# see "Mind your charset" on this page:
# https://docs.djangoproject.com/en/1.3/topics/i18n/localization/
ENCODING = 'utf_8'
def __init__ (self, pathname):
self.pathname = pathname
self.parse()
def parse (self):
with codecs.open(self.pathname, 'r', self.ENCODING) as stream:
text = stream.read()
msgs = text.split('\n\n')
self.msgs = [Msg.parse(m) for m in msgs]
return msgs
def write (self, out_pathname=None):
if out_pathname == None:
out_pathname = self.pathname
with codecs.open(out_pathname, 'w', self.ENCODING) as stream:
for msg in self.msgs:
msg.write(stream)
class Msg:
# A PoFile is parsed into a list of Msg objects, each of which corresponds
# to an externalized string entry.
# Each Msg object may contain multiple comment lines, capturing metadata
# Each Msg has a property list (self.props) with a dict of key-values.
# Each value is a list of strings
kwords = ['msgid', 'msgstr', 'msgctxt', 'msgid_plural']
# Line might begin with "msgid ..." or "msgid[2] ..."
pattern = re.compile('^(\w+)(\[(\d+)\])?')
@classmethod
def parse (cls, string):
'''
String is a fragment of a pofile (.po) source file.
This returns a Msg object created by parsing string.
'''
lines = string.strip().split('\n')
msg = Msg()
msg.comments = []
msg.props = {}
last_kword = None
for line in lines:
if line[0]=='#':
msg.comments.append(line)
elif line[0]=='"' and last_kword != None:
msg.add_string(last_kword, line)
else:
match = cls.pattern.search(line)
if match:
kword = match.group(1)
last_kword = kword
if kword in cls.kwords:
if match.group(3):
key = '%s[%s]' % (kword, match.group(3))
msg.add_string(key, line[len(key):])
else:
msg.add_string(kword, line[len(kword):])
return msg
def get_property (self, kword):
'''returns value for kword. Typically returns a list of strings'''
return self.props.get(kword, [])
def set_property (self, kword, value):
'''sets value for kword. Typically returns a list of strings'''
self.props[kword] = value
def add_string (self, kword, line):
'''Append line to the list of values stored for the property kword'''
props = self.props
value = self.get_property(kword)
value.append(self.cleanup_string(line))
self.set_property(kword, value)
def cleanup_string(self, string):
string = string.strip()
if len(string)>1 and string[0]=='"' and string[-1]=='"':
return string[1:-1]
else:
return string
def write (self, stream):
'''Write a Msg to stream'''
for comment in self.comments:
stream.write(comment)
stream.write('\n')
for (key, values) in self.sort(self.props.items()):
stream.write(key + ' ')
for value in values:
stream.write('"'+value+'"')
stream.write('\n')
stream.write('\n')
# Preferred ordering of key output
# Always print 'msgctxt' first, then 'msgid', etc.
KEY_ORDER = ('msgctxt', 'msgid', 'msgid_plural', 'msgstr', 'msgstr[0]', 'msgstr[1]')
def keyword_compare (self, k1, k2):
for key in self.KEY_ORDER:
if key == k1:
return -1
if key == k2:
return 1
return 0
def sort (self, plist):
'''sorts a propertylist to bring the high-priority keys to the beginning of the list'''
return sorted(plist, key=itemgetter(0), cmp=self.keyword_compare)
# Testing
#
# >>> file = 'mitx/conf/locale/en/LC_MESSAGES/django.po'
# >>> file1 = 'mitx/conf/locale/en/LC_MESSAGES/django1.po'
# >>> po = PoFile(file)
# >>> po.write(file1)
# $ diff file file1
i18n/update.py
......@@ -42,8 +42,8 @@ BABEL_OUT = MSGS_DIR + '/mako.po'
# These are the shell commands invoked by main()
COMMANDS = {
'babel_mako': 'pybabel extract -F %s -c "TRANSLATORS:" . -o %s' % (BABEL_CONFIG, BABEL_OUT),
'make_django': 'django-admin.py makemessages --all --extension html -l en',
'make_djangojs': 'django-admin.py makemessages --all -d djangojs --extension js -l en',
'make_django': 'django-admin.py makemessages --all --ignore=src/* --extension html -l en',
'make_djangojs': 'django-admin.py makemessages --all -d djangojs --ignore=src/* --extension js -l en',
'msgcat' : 'msgcat -o merged.po django.po %s' % BABEL_OUT,
'rename_django' : 'mv django.po django_old.po',
'rename_merged' : 'mv merged.po django.po',
......@@ -81,6 +81,15 @@ def main ():
create_dir_if_necessary(LOCALE_DIR)
log.info('Executing all commands from %s' % BASE_DIR)
remove_files = ['django.po', 'djangojs.po', 'nonesuch']
for filename in remove_files:
path = MSGS_DIR + '/' + filename
log.info('Deleting file %s' % path)
if not os.path.exists(path):
log.warn("File does not exist: %s" % path)
else:
os.remove(path)
# Generate or update human-readable .po files from all source code.
execute('babel_mako', log=log)
execute('make_django', log=log)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment