Merge pull request #937 from jaimeMF/subtitles_rework

Subtitles rework
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-08-23 02:40:25 -07:00
commit 9e9c164052
4 changed files with 67 additions and 92 deletions

View file

@ -35,47 +35,47 @@ class TestYoutubeSubtitles(unittest.TestCase):
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260') self.assertEqual(md5(sub), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_subtitles_it(self): def test_youtube_subtitles_it(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
DL.params['subtitleslang'] = 'it' DL.params['subtitleslang'] = 'it'
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['it']
self.assertEqual(md5(sub[2]), '164a51f16f260476a05b50fe4c2f161d') self.assertEqual(md5(sub), '164a51f16f260476a05b50fe4c2f161d')
def test_youtube_onlysubtitles(self): def test_youtube_onlysubtitles(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
DL.params['onlysubtitles'] = True DL.params['onlysubtitles'] = True
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260') self.assertEqual(md5(sub), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_allsubtitles(self): def test_youtube_allsubtitles(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['allsubtitles'] = True DL.params['allsubtitles'] = True
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles'] subtitles = info_dict[0]['subtitles']
self.assertEqual(len(subtitles), 13) self.assertEqual(len(subtitles.keys()), 13)
def test_youtube_subtitles_sbv_format(self): def test_youtube_subtitles_sbv_format(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'sbv' DL.params['subtitlesformat'] = 'sbv'
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub[2]), '13aeaa0c245a8bed9a451cb643e3ad8b') self.assertEqual(md5(sub), '13aeaa0c245a8bed9a451cb643e3ad8b')
def test_youtube_subtitles_vtt_format(self): def test_youtube_subtitles_vtt_format(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'vtt' DL.params['subtitlesformat'] = 'vtt'
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub[2]), '356cdc577fde0c6783b9b822e7206ff7') self.assertEqual(md5(sub), '356cdc577fde0c6783b9b822e7206ff7')
def test_youtube_list_subtitles(self): def test_youtube_list_subtitles(self):
DL = FakeYDL() DL = FakeYDL()
DL.params['listsubtitles'] = True DL.params['listsubtitles'] = True
@ -88,8 +88,8 @@ class TestYoutubeSubtitles(unittest.TestCase):
DL.params['subtitleslang'] = 'it' DL.params['subtitleslang'] = 'it'
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('8YoUxe5ncPo') info_dict = IE.extract('8YoUxe5ncPo')
sub = info_dict[0]['subtitles'][0] sub = info_dict[0]['subtitles']['it']
self.assertTrue(sub[2] is not None) self.assertTrue(sub is not None)
if __name__ == '__main__': if __name__ == '__main__':
unittest.main() unittest.main()

View file

@ -483,41 +483,28 @@ class YoutubeDL(object):
self.report_error(u'Cannot write description file ' + descfn) self.report_error(u'Cannot write description file ' + descfn)
return return
if (self.params.get('writesubtitles', False) or self.params.get('writeautomaticsub')) and 'subtitles' in info_dict and info_dict['subtitles']: subtitles_are_requested = any([self.params.get('writesubtitles', False),
self.params.get('writeautomaticsub'),
self.params.get('allsubtitles', False)])
if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
# subtitles download errors are already managed as troubles in relevant IE # subtitles download errors are already managed as troubles in relevant IE
# that way it will silently go on when used with unsupporting IE # that way it will silently go on when used with unsupporting IE
subtitle = info_dict['subtitles'][0] subtitles = info_dict['subtitles']
(sub_error, sub_lang, sub) = subtitle
sub_format = self.params.get('subtitlesformat') sub_format = self.params.get('subtitlesformat')
if sub_error: for sub_lang in subtitles.keys():
self.report_warning("Some error while getting the subtitles") sub = subtitles[sub_lang]
else: if sub is None:
continue
try: try:
sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
self.report_writesubtitles(sub_filename) self.report_writesubtitles(sub_filename)
with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
subfile.write(sub) subfile.write(sub)
except (OSError, IOError): except (OSError, IOError):
self.report_error(u'Cannot write subtitles file ' + descfn) self.report_error(u'Cannot write subtitles file ' + descfn)
return return
if self.params.get('allsubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
subtitles = info_dict['subtitles']
sub_format = self.params.get('subtitlesformat')
for subtitle in subtitles:
(sub_error, sub_lang, sub) = subtitle
if sub_error:
self.report_warning("Some error while getting the subtitles")
else:
try:
sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
self.report_writesubtitles(sub_filename)
with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
subfile.write(sub)
except (OSError, IOError):
self.report_error(u'Cannot write subtitles file ' + descfn)
return
if self.params.get('writeinfojson', False): if self.params.get('writeinfojson', False):
infofn = filename + u'.info.json' infofn = filename + u'.info.json'
self.report_writeinfojson(infofn) self.report_writeinfojson(infofn)

View file

@ -47,7 +47,8 @@ class InfoExtractor(object):
uploader_id: Nickname or id of the video uploader. uploader_id: Nickname or id of the video uploader.
location: Physical location of the video. location: Physical location of the video.
player_url: SWF Player URL (used for rtmpdump). player_url: SWF Player URL (used for rtmpdump).
subtitles: The subtitle file contents. subtitles: The subtitle file contents as a dictionary in the format
{language: subtitles}.
view_count: How many users have watched the video on the platform. view_count: How many users have watched the video on the platform.
urlhandle: [internal] The urlHandle to be used to download the file, urlhandle: [internal] The urlHandle to be used to download the file,
like returned by urllib.request.urlopen like returned by urllib.request.urlopen

View file

@ -458,11 +458,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
try: try:
sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'unable to download video subtitles: %s' % compat_str(err), None) self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
return {}
sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
if not sub_lang_list: if not sub_lang_list:
return (u'video doesn\'t have subtitles', None) self._downloader.report_warning(u'video doesn\'t have subtitles')
return {}
return sub_lang_list return sub_lang_list
def _list_available_subtitles(self, video_id): def _list_available_subtitles(self, video_id):
@ -471,8 +473,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _request_subtitle(self, sub_lang, sub_name, video_id, format): def _request_subtitle(self, sub_lang, sub_name, video_id, format):
""" """
Return tuple: Return the subtitle as a string or None if they are not found
(error_message, sub_lang, sub)
""" """
self.report_video_subtitles_request(video_id, sub_lang, format) self.report_video_subtitles_request(video_id, sub_lang, format)
params = compat_urllib_parse.urlencode({ params = compat_urllib_parse.urlencode({
@ -485,10 +486,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
try: try:
sub = compat_urllib_request.urlopen(url).read().decode('utf-8') sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'unable to download video subtitles: %s' % compat_str(err), None, None) self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
return
if not sub: if not sub:
return (u'Did not fetch video subtitles', None, None) self._downloader.report_warning(u'Did not fetch video subtitles')
return (None, sub_lang, sub) return
return sub
def _request_automatic_caption(self, video_id, webpage): def _request_automatic_caption(self, video_id, webpage):
"""We need the webpage for getting the captions url, pass it as an """We need the webpage for getting the captions url, pass it as an
@ -499,7 +502,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
mobj = re.search(r';ytplayer.config = ({.*?});', webpage) mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
if mobj is None: if mobj is None:
return [(err_msg, None, None)] self._downloader.report_warning(err_msg)
return {}
player_config = json.loads(mobj.group(1)) player_config = json.loads(mobj.group(1))
try: try:
args = player_config[u'args'] args = player_config[u'args']
@ -514,40 +518,38 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
}) })
subtitles_url = caption_url + '&' + params subtitles_url = caption_url + '&' + params
sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
return [(None, sub_lang, sub)] return {sub_lang: sub}
except KeyError: # An extractor error can be raise by the download process if there are
return [(err_msg, None, None)] # no automatic captions but there are subtitles
except (KeyError, ExtractorError):
self._downloader.report_warning(err_msg)
return {}
def _extract_subtitle(self, video_id): def _extract_subtitles(self, video_id):
""" """
Return a list with a tuple: Return a dictionary: {language: subtitles} or {} if the subtitles
[(error_message, sub_lang, sub)] couldn't be found
""" """
sub_lang_list = self._get_available_subtitles(video_id) sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat') sub_format = self._downloader.params.get('subtitlesformat')
if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles if not sub_lang_list: #There was some error, it didn't get the available subtitles
return [(sub_lang_list[0], None, None)] return {}
if self._downloader.params.get('subtitleslang', False): if self._downloader.params.get('writesubtitles', False):
sub_lang = self._downloader.params.get('subtitleslang') if self._downloader.params.get('subtitleslang', False):
elif 'en' in sub_lang_list: sub_lang = self._downloader.params.get('subtitleslang')
sub_lang = 'en' elif 'en' in sub_lang_list:
else: sub_lang = 'en'
sub_lang = list(sub_lang_list.keys())[0] else:
if not sub_lang in sub_lang_list: sub_lang = list(sub_lang_list.keys())[0]
return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] if not sub_lang in sub_lang_list:
self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) return {}
return [subtitle] sub_lang_list = {sub_lang: sub_lang_list[sub_lang]}
subtitles = {}
def _extract_all_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
return [(sub_lang_list[0], None, None)]
subtitles = []
for sub_lang in sub_lang_list: for sub_lang in sub_lang_list:
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
subtitles.append(subtitle) if subtitle:
subtitles[sub_lang] = subtitle
return subtitles return subtitles
def _print_formats(self, formats): def _print_formats(self, formats):
@ -738,25 +740,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# subtitles # subtitles
video_subtitles = None video_subtitles = None
if self._downloader.params.get('writesubtitles', False): if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_subtitle(video_id) video_subtitles = self._extract_subtitles(video_id)
if video_subtitles: elif self._downloader.params.get('writeautomaticsub', False):
(sub_error, sub_lang, sub) = video_subtitles[0]
if sub_error:
self._downloader.report_warning(sub_error)
if self._downloader.params.get('writeautomaticsub', False):
video_subtitles = self._request_automatic_caption(video_id, video_webpage) video_subtitles = self._request_automatic_caption(video_id, video_webpage)
(sub_error, sub_lang, sub) = video_subtitles[0]
if sub_error:
self._downloader.report_warning(sub_error)
if self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_all_subtitles(video_id)
for video_subtitle in video_subtitles:
(sub_error, sub_lang, sub) = video_subtitle
if sub_error:
self._downloader.report_warning(sub_error)
if self._downloader.params.get('listsubtitles', False): if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id) self._list_available_subtitles(video_id)