Merge branch 'subtitles_rework' (closes PR #1326)

This commit is contained in:
Jaime Marquínez Ferrándiz 2013-09-11 17:37:57 +02:00
commit 542cca0e8c
8 changed files with 259 additions and 163 deletions

8
.gitignore vendored
View File

@ -17,4 +17,10 @@ youtube-dl.tar.gz
.coverage
cover/
updates_key.pem
*.egg-info
*.egg-info
*.srt
*.sbv
*.vtt
*.flv
*.mp4
*.part

View File

@ -38,7 +38,6 @@
"writedescription": false,
"writeinfojson": true,
"writesubtitles": false,
"onlysubtitles": false,
"allsubtitles": false,
"listssubtitles": false
}

View File

@ -0,0 +1,69 @@
#!/usr/bin/env python
import sys
import unittest
import json
import io
import hashlib
# Allow direct execution
import os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from youtube_dl.extractor import DailymotionIE
from youtube_dl.utils import *
from helper import FakeYDL
md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
class TestDailymotionSubtitles(unittest.TestCase):
def setUp(self):
self.DL = FakeYDL()
self.url = 'http://www.dailymotion.com/video/xczg00'
def getInfoDict(self):
IE = DailymotionIE(self.DL)
info_dict = IE.extract(self.url)
return info_dict
def getSubtitles(self):
info_dict = self.getInfoDict()
return info_dict[0]['subtitles']
def test_no_writesubtitles(self):
subtitles = self.getSubtitles()
self.assertEqual(subtitles, None)
def test_subtitles(self):
self.DL.params['writesubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
def test_subtitles_lang(self):
self.DL.params['writesubtitles'] = True
self.DL.params['subtitleslangs'] = ['fr']
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
def test_allsubtitles(self):
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(len(subtitles.keys()), 5)
def test_list_subtitles(self):
self.DL.params['listsubtitles'] = True
info_dict = self.getInfoDict()
self.assertEqual(info_dict, None)
def test_automatic_captions(self):
self.DL.params['writeautomaticsub'] = True
self.DL.params['subtitleslang'] = ['en']
subtitles = self.getSubtitles()
self.assertTrue(len(subtitles.keys()) == 0)
def test_nosubtitles(self):
self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv'
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(len(subtitles), 0)
def test_multiple_langs(self):
self.DL.params['writesubtitles'] = True
langs = ['es', 'fr', 'de']
self.DL.params['subtitleslangs'] = langs
subtitles = self.getSubtitles()
for lang in langs:
self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
if __name__ == '__main__':
unittest.main()

View File

@ -18,85 +18,63 @@ md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
class TestYoutubeSubtitles(unittest.TestCase):
def setUp(self):
DL = FakeYDL()
DL.params['allsubtitles'] = False
DL.params['writesubtitles'] = False
DL.params['subtitlesformat'] = 'srt'
DL.params['listsubtitles'] = False
def test_youtube_no_subtitles(self):
DL = FakeYDL()
DL.params['writesubtitles'] = False
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.DL = FakeYDL()
self.url = 'QRS8MkLhQmM'
def getInfoDict(self):
IE = YoutubeIE(self.DL)
info_dict = IE.extract(self.url)
return info_dict
def getSubtitles(self):
info_dict = self.getInfoDict()
return info_dict[0]['subtitles']
def test_youtube_no_writesubtitles(self):
self.DL.params['writesubtitles'] = False
subtitles = self.getSubtitles()
self.assertEqual(subtitles, None)
def test_youtube_subtitles(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_subtitles_it(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
DL.params['subtitleslangs'] = ['it']
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles']['it']
self.assertEqual(md5(sub), '164a51f16f260476a05b50fe4c2f161d')
def test_youtube_onlysubtitles(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
DL.params['onlysubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub), '4cd9278a35ba2305f47354ee13472260')
self.DL.params['writesubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_subtitles_lang(self):
self.DL.params['writesubtitles'] = True
self.DL.params['subtitleslangs'] = ['it']
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
def test_youtube_allsubtitles(self):
DL = FakeYDL()
DL.params['allsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(len(subtitles.keys()), 13)
def test_youtube_subtitles_sbv_format(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'sbv'
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub), '13aeaa0c245a8bed9a451cb643e3ad8b')
self.DL.params['writesubtitles'] = True
self.DL.params['subtitlesformat'] = 'sbv'
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
def test_youtube_subtitles_vtt_format(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'vtt'
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles']['en']
self.assertEqual(md5(sub), '356cdc577fde0c6783b9b822e7206ff7')
self.DL.params['writesubtitles'] = True
self.DL.params['subtitlesformat'] = 'vtt'
subtitles = self.getSubtitles()
self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
def test_youtube_list_subtitles(self):
DL = FakeYDL()
DL.params['listsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
self.DL.params['listsubtitles'] = True
info_dict = self.getInfoDict()
self.assertEqual(info_dict, None)
def test_youtube_automatic_captions(self):
DL = FakeYDL()
DL.params['writeautomaticsub'] = True
DL.params['subtitleslangs'] = ['it']
IE = YoutubeIE(DL)
info_dict = IE.extract('8YoUxe5ncPo')
sub = info_dict[0]['subtitles']['it']
self.assertTrue(sub is not None)
self.url = '8YoUxe5ncPo'
self.DL.params['writeautomaticsub'] = True
self.DL.params['subtitleslangs'] = ['it']
subtitles = self.getSubtitles()
self.assertTrue(subtitles['it'] is not None)
def test_youtube_nosubtitles(self):
self.url = 'sAjKT8FhjI8'
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(len(subtitles), 0)
def test_youtube_multiple_langs(self):
DL = FakeYDL()
DL.params['writesubtitles'] = True
self.url = 'QRS8MkLhQmM'
self.DL.params['writesubtitles'] = True
langs = ['it', 'fr', 'de']
DL.params['subtitleslangs'] = langs
IE = YoutubeIE(DL)
subtitles = IE.extract('QRS8MkLhQmM')[0]['subtitles']
self.DL.params['subtitleslangs'] = langs
subtitles = self.getSubtitles()
for lang in langs:
self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)

View File

@ -205,13 +205,10 @@ def parseOpts(overrideArguments=None):
subtitles.add_option('--write-sub', '--write-srt',
action='store_true', dest='writesubtitles',
help='write subtitle file (currently youtube only)', default=False)
help='write subtitle file', default=False)
subtitles.add_option('--write-auto-sub', '--write-automatic-sub',
action='store_true', dest='writeautomaticsub',
help='write automatic subtitle file (currently youtube only)', default=False)
subtitles.add_option('--only-sub',
action='store_true', dest='skip_download',
help='[deprecated] alias of --skip-download', default=False)
help='write automatic subtitle file (youtube only)', default=False)
subtitles.add_option('--all-subs',
action='store_true', dest='allsubtitles',
help='downloads all the available subtitles of the video', default=False)
@ -222,7 +219,7 @@ def parseOpts(overrideArguments=None):
action='store', dest='subtitlesformat', metavar='FORMAT',
help='subtitle format (default=srt) ([sbv/vtt] youtube only)', default='srt')
subtitles.add_option('--sub-lang', '--sub-langs', '--srt-lang',
action='callback', dest='subtitleslang', metavar='LANGS', type='str',
action='callback', dest='subtitleslangs', metavar='LANGS', type='str',
default=[], callback=_comma_separated_values_options_callback,
help='languages of the subtitles to download (optional) separated by commas, use IETF language tags like \'en,pt\'')
@ -593,7 +590,7 @@ def _real_main(argv=None):
'allsubtitles': opts.allsubtitles,
'listsubtitles': opts.listsubtitles,
'subtitlesformat': opts.subtitlesformat,
'subtitleslangs': opts.subtitleslang,
'subtitleslangs': opts.subtitleslangs,
'matchtitle': decodeOption(opts.matchtitle),
'rejecttitle': decodeOption(opts.rejecttitle),
'max_downloads': opts.max_downloads,

View File

@ -3,15 +3,19 @@ import json
import itertools
from .common import InfoExtractor
from .subtitles import SubtitlesInfoExtractor
from ..utils import (
compat_urllib_request,
compat_str,
get_element_by_attribute,
get_element_by_id,
ExtractorError,
)
class DailymotionIE(InfoExtractor):
class DailymotionIE(SubtitlesInfoExtractor):
"""Information Extractor for Dailymotion"""
_VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
@ -73,6 +77,12 @@ class DailymotionIE(InfoExtractor):
raise ExtractorError(u'Unable to extract video URL')
video_url = info[max_quality]
# subtitles
video_subtitles = self.extract_subtitles(video_id)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id)
return
return [{
'id': video_id,
'url': video_url,
@ -80,9 +90,25 @@ class DailymotionIE(InfoExtractor):
'upload_date': video_upload_date,
'title': self._og_search_title(webpage),
'ext': video_extension,
'subtitles': video_subtitles,
'thumbnail': info['thumbnail_url']
}]
def _get_available_subtitles(self, video_id):
try:
sub_list = self._download_webpage(
'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
video_id, note=False)
except ExtractorError as err:
self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
return {}
info = json.loads(sub_list)
if (info['total'] > 0):
sub_lang_list = dict((l['language'], l['url']) for l in info['list'])
return sub_lang_list
self._downloader.report_warning(u'video doesn\'t have subtitles')
return {}
class DailymotionPlaylistIE(InfoExtractor):
_VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'

View File

@ -0,0 +1,86 @@
from .common import InfoExtractor
from ..utils import (
compat_str,
ExtractorError,
)
class SubtitlesInfoExtractor(InfoExtractor):
def _list_available_subtitles(self, video_id):
""" outputs the available subtitles for the video """
sub_lang_list = self._get_available_subtitles(video_id)
sub_lang = ",".join(list(sub_lang_list.keys()))
self.to_screen(u'%s: Available subtitles for video: %s' %
(video_id, sub_lang))
def _extract_subtitles(self, video_id):
""" returns {sub_lang: sub} or {} if subtitles not found """
available_subs_list = self._get_available_subtitles(video_id)
if not available_subs_list: # error, it didn't get the available subtitles
return {}
if self._downloader.params.get('allsubtitles', False):
sub_lang_list = available_subs_list
else:
if self._downloader.params.get('writesubtitles', False):
if self._downloader.params.get('subtitleslangs', False):
requested_langs = self._downloader.params.get('subtitleslangs')
elif 'en' in available_subs_list:
requested_langs = ['en']
else:
requested_langs = [list(available_subs_list.keys())[0]]
sub_lang_list = {}
for sub_lang in requested_langs:
if not sub_lang in available_subs_list:
self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
continue
sub_lang_list[sub_lang] = available_subs_list[sub_lang]
subtitles = {}
for sub_lang, url in sub_lang_list.items():
subtitle = self._request_subtitle_url(sub_lang, url)
if subtitle:
subtitles[sub_lang] = subtitle
return subtitles
def _request_subtitle_url(self, sub_lang, url):
""" makes the http request for the subtitle """
try:
sub = self._download_webpage(url, None, note=False)
except ExtractorError as err:
self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
return
if not sub:
self._downloader.report_warning(u'Did not fetch video subtitles')
return
return sub
def _get_available_subtitles(self, video_id):
"""
returns {sub_lang: url} or {} if not available
Must be redefined by the subclasses
"""
pass
def _request_automatic_caption(self, video_id, webpage):
"""
returns {sub_lang: sub} or {} if not available
Must be redefined by the subclasses that support automatic captions,
otherwise it will return {}
"""
self._downloader.report_warning(u'Automatic Captions not supported by this server')
return {}
def extract_subtitles(self, video_id, video_webpage=None):
"""
Extract the subtitles and/or the automatic captions if requested.
Returns None or a dictionary in the format {sub_lang: sub}
"""
video_subtitles = None
if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_subtitles(video_id)
elif self._downloader.params.get('writeautomaticsub', False):
video_subtitles = self._request_automatic_caption(video_id, video_webpage)
return video_subtitles

View File

@ -7,6 +7,7 @@ import socket
import itertools
from .common import InfoExtractor, SearchInfoExtractor
from .subtitles import SubtitlesInfoExtractor
from ..utils import (
compat_http_client,
compat_parse_qs,
@ -130,7 +131,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
return
self._confirm_age()
class YoutubeIE(YoutubeBaseInfoExtractor):
class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
IE_DESC = u'YouTube.com'
_VALID_URL = r"""^
(
@ -397,19 +399,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video info webpage' % video_id)
def report_video_subtitles_download(self, video_id):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Checking available subtitles' % video_id)
def report_video_subtitles_request(self, video_id, sub_lang, format):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
def report_video_subtitles_available(self, video_id, sub_lang_list):
"""Report available subtitles."""
sub_lang = ",".join(list(sub_lang_list.keys()))
self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
def report_information_extraction(self, video_id):
"""Report attempt to extract video information."""
self.to_screen(u'%s: Extracting video information' % video_id)
@ -464,48 +453,31 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Fallback to the other algortihms
return self._decrypt_signature(s)
def _get_available_subtitles(self, video_id):
self.report_video_subtitles_download(video_id)
request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
try:
sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
sub_list = self._download_webpage(
'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
video_id, note=False)
except ExtractorError as err:
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 = dict((l[1], l[0]) for l in sub_lang_list)
lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
sub_lang_list = {}
for l in lang_list:
lang = l[1]
params = compat_urllib_parse.urlencode({
'lang': lang,
'v': video_id,
'fmt': self._downloader.params.get('subtitlesformat'),
})
url = u'http://www.youtube.com/api/timedtext?' + params
sub_lang_list[lang] = url
if not sub_lang_list:
self._downloader.report_warning(u'video doesn\'t have subtitles')
return {}
return sub_lang_list
def _list_available_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
self.report_video_subtitles_available(video_id, sub_lang_list)
def _request_subtitle(self, sub_lang, sub_name, video_id, format):
"""
Return the subtitle as a string or None if they are not found
"""
self.report_video_subtitles_request(video_id, sub_lang, format)
params = compat_urllib_parse.urlencode({
'lang': sub_lang,
'name': sub_name,
'v': video_id,
'fmt': format,
})
url = 'http://www.youtube.com/api/timedtext?' + params
try:
sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
return
if not sub:
self._downloader.report_warning(u'Did not fetch video subtitles')
return
return sub
def _request_automatic_caption(self, video_id, webpage):
"""We need the webpage for getting the captions url, pass it as an
argument to speed up the process."""
@ -537,38 +509,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
except (KeyError, ExtractorError):
self._downloader.report_warning(err_msg)
return {}
def _extract_subtitles(self, video_id):
"""
Return a dictionary: {language: subtitles} or {} if the subtitles
couldn't be found
"""
available_subs_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
if not available_subs_list: #There was some error, it didn't get the available subtitles
return {}
if self._downloader.params.get('allsubtitles', False):
sub_lang_list = available_subs_list
else:
if self._downloader.params.get('subtitleslangs', False):
reqested_langs = self._downloader.params.get('subtitleslangs')
elif 'en' in available_subs_list:
reqested_langs = ['en']
else:
reqested_langs = [list(available_subs_list.keys())[0]]
sub_lang_list = {}
for sub_lang in reqested_langs:
if not sub_lang in available_subs_list:
self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
continue
sub_lang_list[sub_lang] = available_subs_list[sub_lang]
subtitles = {}
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)
if subtitle:
subtitles[sub_lang] = subtitle
return subtitles
def _print_formats(self, formats):
print('Available formats:')
@ -768,12 +708,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
video_description = u''
# subtitles
video_subtitles = None
if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_subtitles(video_id)
elif self._downloader.params.get('writeautomaticsub', False):
video_subtitles = self._request_automatic_caption(video_id, video_webpage)
video_subtitles = self.extract_subtitles(video_id, video_webpage)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id)