Merge pull request #699 by @iemejia

Removed innecesary function to convert subtitles, improved use of the youtube api
This commit is contained in:
Filippo Valsorda 2013-03-20 11:35:55 +01:00
commit 5cb3c0b319
5 changed files with 177 additions and 58 deletions

View File

@ -29,6 +29,7 @@
"simulate": false, "simulate": false,
"skip_download": false, "skip_download": false,
"subtitleslang": null, "subtitleslang": null,
"subtitlesformat": "srt",
"test": true, "test": true,
"updatetime": true, "updatetime": true,
"usenetrc": false, "usenetrc": false,
@ -36,5 +37,8 @@
"verbose": true, "verbose": true,
"writedescription": false, "writedescription": false,
"writeinfojson": true, "writeinfojson": true,
"writesubtitles": false "writesubtitles": false,
"onlysubtitles": false,
"allsubtitles": false,
"listssubtitles": false
} }

View File

@ -38,20 +38,63 @@ class FakeDownloader(object):
md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest() md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
class TestYoutubeSubtitles(unittest.TestCase): class TestYoutubeSubtitles(unittest.TestCase):
def setUp(self):
DL = FakeDownloader()
DL.params['allsubtitles'] = False
DL.params['writesubtitles'] = False
DL.params['subtitlesformat'] = 'srt'
DL.params['listsubtitles'] = False
def test_youtube_no_subtitles(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = False
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.assertEqual(subtitles, None)
def test_youtube_subtitles(self): def test_youtube_subtitles(self):
DL = FakeDownloader() DL = FakeDownloader()
DL.params['writesubtitles'] = True DL.params['writesubtitles'] = True
IE = YoutubeIE(DL) IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM') info_dict = IE.extract('QRS8MkLhQmM')
self.assertEqual(md5(info_dict[0]['subtitles']), 'c3228550d59116f3c29fba370b55d033') sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_subtitles_it(self): def test_youtube_subtitles_it(self):
DL = FakeDownloader() DL = FakeDownloader()
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')
self.assertEqual(md5(info_dict[0]['subtitles']), '132a88a0daf8e1520f393eb58f1f646a') sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '164a51f16f260476a05b50fe4c2f161d')
def test_youtube_onlysubtitles(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
DL.params['onlysubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '4cd9278a35ba2305f47354ee13472260')
def test_youtube_allsubtitles(self):
DL = FakeDownloader()
DL.params['allsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
subtitles = info_dict[0]['subtitles']
self.assertEqual(len(subtitles), 12)
def test_youtube_subtitles_format(self):
DL = FakeDownloader()
DL.params['writesubtitles'] = True
DL.params['subtitlesformat'] = 'sbv'
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
sub = info_dict[0]['subtitles'][0]
self.assertEqual(md5(sub[2]), '13aeaa0c245a8bed9a451cb643e3ad8b')
def test_youtube_list_subtitles(self):
DL = FakeDownloader()
DL.params['listsubtitles'] = True
IE = YoutubeIE(DL)
info_dict = IE.extract('QRS8MkLhQmM')
self.assertEqual(info_dict, None)
if __name__ == '__main__': if __name__ == '__main__':
unittest.main() unittest.main()

View File

@ -78,7 +78,11 @@ class FileDownloader(object):
updatetime: Use the Last-modified header to set output file timestamps. updatetime: Use the Last-modified header to set output file timestamps.
writedescription: Write the video description to a .description file writedescription: Write the video description to a .description file
writeinfojson: Write the video description to a .info.json file writeinfojson: Write the video description to a .info.json file
writesubtitles: Write the video subtitles to a .srt file writesubtitles: Write the video subtitles to a file
onlysubtitles: Downloads only the subtitles of the video
allsubtitles: Downloads all the subtitles of the video
listsubtitles: Lists all available subtitles for the video
subtitlesformat: Subtitle format [sbv/srt] (default=srt)
subtitleslang: Language of the subtitles to download subtitleslang: Language of the subtitles to download
test: Download only first bytes to test the downloader. test: Download only first bytes to test the downloader.
keepvideo: Keep the video file after post-processing keepvideo: Keep the video file after post-processing
@ -301,9 +305,9 @@ class FileDownloader(object):
""" Report that the description file is being written """ """ Report that the description file is being written """
self.to_screen(u'[info] Writing video description to: ' + descfn) self.to_screen(u'[info] Writing video description to: ' + descfn)
def report_writesubtitles(self, srtfn): def report_writesubtitles(self, sub_filename):
""" Report that the subtitles file is being written """ """ Report that the subtitles file is being written """
self.to_screen(u'[info] Writing video subtitles to: ' + srtfn) self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
def report_writeinfojson(self, infofn): def report_writeinfojson(self, infofn):
""" Report that the metadata file has been written """ """ Report that the metadata file has been written """
@ -456,14 +460,35 @@ class FileDownloader(object):
if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']: if self.params.get('writesubtitles', False) 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]
(sub_error, sub_lang, sub) = subtitle
sub_format = self.params.get('subtitlesformat')
try: try:
srtfn = filename.rsplit('.', 1)[0] + u'.srt' sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
self.report_writesubtitles(srtfn) self.report_writesubtitles(sub_filename)
with io.open(encodeFilename(srtfn), 'w', encoding='utf-8') as srtfile: with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
srtfile.write(info_dict['subtitles']) subfile.write(sub)
except (OSError, IOError): except (OSError, IOError):
self.trouble(u'ERROR: Cannot write subtitles file ' + descfn) self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
return return
if self.params.get('onlysubtitles', False):
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
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.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
return
if self.params.get('onlysubtitles', False):
return
if self.params.get('writeinfojson', False): if self.params.get('writeinfojson', False):
infofn = filename + u'.info.json' infofn = filename + u'.info.json'

View File

@ -48,7 +48,7 @@ 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 .srt file contents. subtitles: The subtitle file contents.
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
@ -224,7 +224,16 @@ class YoutubeIE(InfoExtractor):
def report_video_subtitles_download(self, video_id): def report_video_subtitles_download(self, video_id):
"""Report attempt to download video info webpage.""" """Report attempt to download video info webpage."""
self._downloader.to_screen(u'[youtube] %s: Downloading video subtitles' % video_id) self._downloader.to_screen(u'[youtube] %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._downloader.to_screen(u'[youtube] %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._downloader.to_screen(u'[youtube] %s: Available subtitles for video: %s' % (video_id, sub_lang))
def report_information_extraction(self, video_id): def report_information_extraction(self, video_id):
"""Report attempt to extract video information.""" """Report attempt to extract video information."""
@ -238,55 +247,63 @@ class YoutubeIE(InfoExtractor):
"""Indicate the download will use the RTMP protocol.""" """Indicate the download will use the RTMP protocol."""
self._downloader.to_screen(u'[youtube] RTMP download detected') self._downloader.to_screen(u'[youtube] RTMP download detected')
def _closed_captions_xml_to_srt(self, xml_string): def _get_available_subtitles(self, video_id):
srt = ''
texts = re.findall(r'<text start="([\d\.]+)"( dur="([\d\.]+)")?>([^<]+)</text>', xml_string, re.MULTILINE)
# TODO parse xml instead of regex
for n, (start, dur_tag, dur, caption) in enumerate(texts):
if not dur: dur = '4'
start = float(start)
end = start + float(dur)
start = "%02i:%02i:%02i,%03i" %(start/(60*60), start/60%60, start%60, start%1*1000)
end = "%02i:%02i:%02i,%03i" %(end/(60*60), end/60%60, end%60, end%1*1000)
caption = unescapeHTML(caption)
caption = unescapeHTML(caption) # double cycle, intentional
srt += str(n+1) + '\n'
srt += start + ' --> ' + end + '\n'
srt += caption + '\n\n'
return srt
def _extract_subtitles(self, video_id):
self.report_video_subtitles_download(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) request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
try: try:
srt_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'WARNING: unable to download video subtitles: %s' % compat_str(err), None) return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
srt_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', srt_list) sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
srt_lang_list = dict((l[1], l[0]) for l in srt_lang_list) sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
if not srt_lang_list: if not sub_lang_list:
return (u'WARNING: video has no closed captions', None) return (u'WARNING: video doesn\'t have subtitles', None)
if self._downloader.params.get('subtitleslang', False): return sub_lang_list
srt_lang = self._downloader.params.get('subtitleslang')
elif 'en' in srt_lang_list: def _list_available_subtitles(self, video_id):
srt_lang = 'en' sub_lang_list = self._get_available_subtitles(video_id)
else: self.report_video_subtitles_available(video_id, sub_lang_list)
srt_lang = list(srt_lang_list.keys())[0]
if not srt_lang in srt_lang_list: def _request_subtitle(self, sub_lang, sub_name, video_id, format):
return (u'WARNING: no closed captions found in the specified language', None) self.report_video_subtitles_request(video_id, sub_lang, format)
params = compat_urllib_parse.urlencode({ params = compat_urllib_parse.urlencode({
'lang': srt_lang, 'lang': sub_lang,
'name': srt_lang_list[srt_lang].encode('utf-8'), 'name': sub_name,
'v': video_id, 'v': video_id,
'fmt': format,
}) })
url = 'http://www.youtube.com/api/timedtext?' + params url = 'http://www.youtube.com/api/timedtext?' + params
try: try:
srt_xml = 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'WARNING: unable to download video subtitles: %s' % compat_str(err), None) return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
if not srt_xml: if not sub:
return (u'WARNING: Did not fetch video subtitles', None) return (u'WARNING: Did not fetch video subtitles', None)
return (None, self._closed_captions_xml_to_srt(srt_xml)) return (None, sub_lang, sub)
def _extract_subtitle(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
if self._downloader.params.get('subtitleslang', False):
sub_lang = self._downloader.params.get('subtitleslang')
elif 'en' in sub_lang_list:
sub_lang = 'en'
else:
sub_lang = list(sub_lang_list.keys())[0]
if not sub_lang in sub_lang_list:
return (u'WARNING: no closed captions found in the specified language "%s"' % sub_lang, None)
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
return [subtitle]
def _extract_all_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
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)
subtitles.append(subtitle)
return subtitles
def _print_formats(self, formats): def _print_formats(self, formats):
print('Available formats:') print('Available formats:')
@ -507,12 +524,26 @@ class YoutubeIE(InfoExtractor):
else: else:
video_description = '' video_description = ''
# closed captions # subtitles
video_subtitles = None video_subtitles = None
if self._downloader.params.get('writesubtitles', False): if self._downloader.params.get('writesubtitles', False):
(srt_error, video_subtitles) = self._extract_subtitles(video_id) video_subtitles = self._extract_subtitle(video_id)
if srt_error: if video_subtitles:
self._downloader.trouble(srt_error) (sub_error, sub_lang, sub) = video_subtitles[0]
if sub_error:
self._downloader.trouble(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.trouble(sub_error)
if self._downloader.params.get('listsubtitles', False):
sub_lang_list = self._list_available_subtitles(video_id)
return
if 'length_seconds' not in video_info: if 'length_seconds' not in video_info:
self._downloader.trouble(u'WARNING: unable to extract video duration') self._downloader.trouble(u'WARNING: unable to extract video duration')

View File

@ -173,12 +173,24 @@ def parseOpts():
action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download') action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download')
video_format.add_option('-F', '--list-formats', video_format.add_option('-F', '--list-formats',
action='store_true', dest='listformats', help='list all available formats (currently youtube only)') action='store_true', dest='listformats', help='list all available formats (currently youtube only)')
video_format.add_option('--write-srt', video_format.add_option('--write-sub', '--write-srt',
action='store_true', dest='writesubtitles', action='store_true', dest='writesubtitles',
help='write video closed captions to a .srt file (currently youtube only)', default=False) help='write subtitle file (currently youtube only)', default=False)
video_format.add_option('--srt-lang', video_format.add_option('--only-sub',
action='store_true', dest='onlysubtitles',
help='downloads only the subtitles (no video)', default=False)
video_format.add_option('--all-subs',
action='store_true', dest='allsubtitles',
help='downloads all the available subtitles of the video (currently youtube only)', default=False)
video_format.add_option('--list-subs',
action='store_true', dest='listsubtitles',
help='lists all available subtitles for the video (currently youtube only)', default=False)
video_format.add_option('--sub-format',
action='store', dest='subtitlesformat', metavar='LANG',
help='subtitle format [srt/sbv] (default=srt) (currently youtube only)', default='srt')
video_format.add_option('--sub-lang', '--srt-lang',
action='store', dest='subtitleslang', metavar='LANG', action='store', dest='subtitleslang', metavar='LANG',
help='language of the closed captions to download (optional) use IETF language tags like \'en\'') help='language of the subtitles to download (optional) use IETF language tags like \'en\'')
verbosity.add_option('-q', '--quiet', verbosity.add_option('-q', '--quiet',
action='store_true', dest='quiet', help='activates quiet mode', default=False) action='store_true', dest='quiet', help='activates quiet mode', default=False)
@ -458,6 +470,10 @@ def _real_main():
'writedescription': opts.writedescription, 'writedescription': opts.writedescription,
'writeinfojson': opts.writeinfojson, 'writeinfojson': opts.writeinfojson,
'writesubtitles': opts.writesubtitles, 'writesubtitles': opts.writesubtitles,
'onlysubtitles': opts.onlysubtitles,
'allsubtitles': opts.allsubtitles,
'listsubtitles': opts.listsubtitles,
'subtitlesformat': opts.subtitlesformat,
'subtitleslang': opts.subtitleslang, 'subtitleslang': opts.subtitleslang,
'matchtitle': decodeOption(opts.matchtitle), 'matchtitle': decodeOption(opts.matchtitle),
'rejecttitle': decodeOption(opts.rejecttitle), 'rejecttitle': decodeOption(opts.rejecttitle),