w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[YoutubeIE] Externalize format selection

Browse source
This commit is contained in:
rzhxeo 2013-12-18 03:30:55 +01:00
parent 3e78514568
commit 4ea3be0a5c
2 changed files with 34 additions and 93 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
youtube_dl/YoutubeDL.py
View file

@ -636,7 +636,7 @@ class YoutubeDL(object):
info_dict['playlist_index'] = None info_dict['playlist_index'] = None
# This extractors handle format selection themselves # This extractors handle format selection themselves
if info_dict['extractor'] in [u'youtube', u'Youku']: if info_dict['extractor'] in [u'Youku']:
if download: if download:
self.process_info(info_dict) self.process_info(info_dict)
return info_dict return info_dict

125
youtube_dl/extractor/youtube.py
View file

@ -162,23 +162,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
# Dash audio # Dash audio
'141', '172', '140', '171', '139', '141', '172', '140', '171', '139',
] ]
_available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13',
# Apple HTTP Live Streaming
'96', '95', '94', '93', '92', '132', '151',
# 3D
'85', '102', '84', '101', '83', '100', '82',
# Dash video
'138', '248', '137', '247', '136', '246', '245',
'244', '135', '243', '134', '242', '133', '160',
# Dash audio
'172', '141', '171', '140', '139',
]
_video_formats_map = {
'flv': ['35', '34', '6', '5'],
'3gp': ['36', '17', '13'],
'mp4': ['38', '37', '22', '18'],
'webm': ['46', '45', '44', '43'],
}
_video_extensions = { _video_extensions = {
'13': '3gp', '13': '3gp',
'17': '3gp', '17': '3gp',
@ -1153,13 +1136,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
self._downloader.report_warning(err_msg) self._downloader.report_warning(err_msg)
return {} return {}
def _print_formats(self, formats):
print('Available formats:')
for x in formats:
print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
self._video_dimensions.get(x, '???'),
' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
def _extract_id(self, url): def _extract_id(self, url):
mobj = re.match(self._VALID_URL, url, re.VERBOSE) mobj = re.match(self._VALID_URL, url, re.VERBOSE)
if mobj is None: if mobj is None:
@ -1172,48 +1148,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
Transform a dictionary in the format {itag:url} to a list of (itag, url) Transform a dictionary in the format {itag:url} to a list of (itag, url)
with the requested formats. with the requested formats.
""" """
req_format = self._downloader.params.get('format', None) existing_formats = [x for x in self._available_formats if x in url_map]
format_limit = self._downloader.params.get('format_limit', None)
available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
if format_limit is not None and format_limit in available_formats:
format_list = available_formats[available_formats.index(format_limit):]
else:
format_list = available_formats
existing_formats = [x for x in format_list if x in url_map]
if len(existing_formats) == 0: if len(existing_formats) == 0:
raise ExtractorError(u'no known formats available for video') raise ExtractorError(u'no known formats available for video')
if self._downloader.params.get('listformats', None): video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
self._print_formats(existing_formats) video_url_list.reverse() # order worst to best
return
if req_format is None or req_format == 'best':
video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
elif req_format == 'worst':
video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
elif req_format in ('-1', 'all'):
video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
else:
# Specific formats. We pick the first in a slash-delimeted sequence.
# Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality
# available in the specified format. For example,
# if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
# if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'.
# if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'.
req_formats = req_format.split('/')
video_url_list = None
for rf in req_formats:
if rf in url_map:
video_url_list = [(rf, url_map[rf])]
break
if rf in self._video_formats_map:
for srf in self._video_formats_map[rf]:
if srf in url_map:
video_url_list = [(srf, url_map[srf])]
break
else:
continue
break
if video_url_list is None:
raise ExtractorError(u'requested format not available')
return video_url_list return video_url_list
def _extract_from_m3u8(self, manifest_url, video_id): def _extract_from_m3u8(self, manifest_url, video_id):
@ -1462,19 +1401,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
url += '&ratebypass=yes' url += '&ratebypass=yes'
url_map[url_data['itag'][0]] = url url_map[url_data['itag'][0]] = url
video_url_list = self._get_video_url_list(url_map) video_url_list = self._get_video_url_list(url_map)
if not video_url_list:
return
elif video_info.get('hlsvp'): elif video_info.get('hlsvp'):
manifest_url = video_info['hlsvp'][0] manifest_url = video_info['hlsvp'][0]
url_map = self._extract_from_m3u8(manifest_url, video_id) url_map = self._extract_from_m3u8(manifest_url, video_id)
video_url_list = self._get_video_url_list(url_map) video_url_list = self._get_video_url_list(url_map)
if not video_url_list:
return
else: else:
raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
results = [] formats = []
for itag, video_real_url in video_url_list: for itag, video_real_url in video_url_list:
# Extension # Extension
video_extension = self._video_extensions.get(itag, 'flv') video_extension = self._video_extensions.get(itag, 'flv')
@ -1482,30 +1416,37 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, video_format = '{0} - {1}{2}'.format(itag if itag else video_extension,
self._video_dimensions.get(itag, '???'), self._video_dimensions.get(itag, '???'),
' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '')
note = self._special_itags.get(itag, None)
resolution = self._video_dimensions.get(itag, None)
results.append({ formats.append({
'id': video_id, 'url': video_real_url,
'url': video_real_url, 'ext': video_extension,
'uploader': video_uploader, 'format': video_format,
'uploader_id': video_uploader_id, 'format_id': itag,
'upload_date': upload_date, 'player_url': player_url,
'title': video_title, '_resolution': resolution,
'ext': video_extension, 'format_note': note,
'format': video_format,
'format_id': itag,
'thumbnail': video_thumbnail,
'description': video_description,
'player_url': player_url,
'subtitles': video_subtitles,
'duration': video_duration,
'age_limit': 18 if age_gate else 0,
'annotations': video_annotations,
'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
}) })
return results
return {
'id': video_id,
'uploader': video_uploader,
'uploader_id': video_uploader_id,
'upload_date': upload_date,
'title': video_title,
'thumbnail': video_thumbnail,
'description': video_description,
'subtitles': video_subtitles,
'duration': video_duration,
'age_limit': 18 if age_gate else 0,
'annotations': video_annotations,
'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
'formats': formats,
}
class YoutubePlaylistIE(YoutubeBaseInfoExtractor): class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
IE_DESC = u'YouTube.com playlists' IE_DESC = u'YouTube.com playlists'