def clean_draft(self):
data = self.cleaned_data['draft']
if data.size > SUBTITLE_FILESIZE_LIMIT_KB * 1024:
raise forms.ValidationError(_(
u'File size must be less than %d kb.'
% SUBTITLE_FILESIZE_LIMIT_KB))
parts = data.name.rsplit('.', 1)
self.extension = parts[-1].lower()
if self.extension not in SUBTITLE_FILE_FORMATS:
raise forms.ValidationError(_(
u'Unsupported format. Please upload one of the following: %s'
% ", ".join(SUBTITLE_FILE_FORMATS)))
text = data.read()
encoding = chardet.detect(text)['encoding']
if not encoding:
raise forms.ValidationError(_(u'Can not detect file encoding'))
# For xml based formats we can't just convert to unicode, as the parser
# will complain that the string encoding doesn't match the encoding
# declaration in the xml file if it's not utf-8.
is_xml = self.extension in ('dfxp', 'ttml', 'xml')
decoded = force_unicode(text, encoding) if not is_xml else text
try:
parser = babelsubs.load_from(decoded, type=self.extension)
self._parsed_subtitles = parser.to_internal()
except TypeError, e:
raise forms.ValidationError(e)
def fetch_subs_vimeo(video_url, user, team):
video_id = video_url.videoid
existing_langs = set(
l.language_code
for l in video_url.video.newsubtitlelanguage_set.having_versions())
for vimeo_account in lookup_vimeo_accounts(video_url, user, team):
tracks = vimeo.get_text_tracks(vimeo_account, video_id)
versions = []
if tracks is not None and \
'data' in tracks:
for track in tracks['data']:
language_code = convert_language_code(track['language'])
if language_code and language_code not in existing_langs:
response = requests.get(track['link'])
try:
subs = load_from(response.content,
type='vtt').to_internal()
version = pipeline.add_subtitles(
video_url.video,
language_code,
subs,
note="From Vimeo",
complete=True,
origin=ORIGIN_IMPORTED)
versions.append(version)
except Exception, e:
logger.error(
"Exception while importing subtitles from Vimeo " +
str(e))
if len(versions) > 0:
subtitles_imported.send(sender=versions[0].subtitle_language,
versions=versions)
break
def version_get_subtitles(sv):
"""Return the SubtitleSet for this version.
A SubtitleSet will always be returned. It may be empty if there are no
subtitles.
"""
return load_from(decompress(sv.serialized_subtitles),
type='dfxp').to_internal()
def _retrieve(self, format):
res = self.client.post(reverse("widget:convert_subtitles"), {
'subtitles': self.subs.to_xml(),
'language_code': 'pt-br',
'format': format,
})
self.assertEqual(res.status_code , 200)
data = json.loads(res.content)
self.assertNotIn('errors', data)
parser = babelsubs.load_from(data['result'], format).to_internal()
parsed = [x for x in parser.subtitle_items()]
self.assertEqual(len(parsed), 10)
return res.content, parsed
def create_langs_and_versions(video, langs, user=None):
from subtitles import pipeline
subtitles = (babelsubs.load_from(SRT, type='srt',
language='en').to_internal())
return [pipeline.add_subtitles(video, l, subtitles) for l in langs]
def _add_language_via_pipeline(video, lang):
subtitles = babelsubs.load_from(SRT, type='srt',
language='en').to_internal()
return pipeline.add_subtitles(video, lang, subtitles)
def test_from_string(self):
subs = load_from(open(utils.get_data_file_path("simple-srt.badextension"), 'r').read() ,type='srt')
parsed = subs.to_internal()
self.assertEquals(len(parsed), 19)
def test_load_from_string(self):
filename = utils.get_data_file_path('simple.dfxp')
with open(filename) as f:
s = f.read()
load_from(s, type='dfxp').to_internal()
def test_load_from_string(self):
filename = utils.get_data_file_path('simple.dfxp')
with open(filename) as f:
s = f.read()
load_from(s, type='dfxp').to_internal()
def test_from_string(self):
subs = load_from(open(
utils.get_data_file_path("simple-srt.badextension"), 'r').read(),
type='srt')
parsed = subs.to_internal()
self.assertEquals(len(parsed), 19)
