from test.helper import FakeYDL
from youtube_dl import YoutubeDL
+from youtube_dl.extractor import YoutubeIE
class YDL(FakeYDL):
{u'ext': u'mp4', u'height': 460},
]
info_dict = {u'formats': formats, u'extractor': u'test'}
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
ydl.process_ie_result(info_dict)
downloaded = ydl.downloaded_info_dicts[0]
self.assertEqual(downloaded[u'ext'], u'webm')
{u'ext': u'mp4', u'height': 1080},
]
info_dict[u'formats'] = formats
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
ydl.process_ie_result(info_dict)
downloaded = ydl.downloaded_info_dicts[0]
self.assertEqual(downloaded[u'ext'], u'mp4')
- # No prefer_free_formats => keep original formats order
+ # No prefer_free_formats => prefer mp4 and flv for greater compatibilty
ydl = YDL()
ydl.params['prefer_free_formats'] = False
formats = [
{u'ext': u'webm', u'height': 720},
+ {u'ext': u'mp4', u'height': 720},
{u'ext': u'flv', u'height': 720},
]
info_dict[u'formats'] = formats
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
+ ydl.process_ie_result(info_dict)
+ downloaded = ydl.downloaded_info_dicts[0]
+ self.assertEqual(downloaded[u'ext'], u'mp4')
+
+ ydl = YDL()
+ ydl.params['prefer_free_formats'] = False
+ formats = [
+ {u'ext': u'flv', u'height': 720},
+ {u'ext': u'webm', u'height': 720},
+ ]
+ info_dict[u'formats'] = formats
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
ydl.process_ie_result(info_dict)
downloaded = ydl.downloaded_info_dicts[0]
self.assertEqual(downloaded[u'ext'], u'flv')
def test_format_limit(self):
formats = [
- {u'format_id': u'meh', u'url': u'http://example.com/meh'},
- {u'format_id': u'good', u'url': u'http://example.com/good'},
- {u'format_id': u'great', u'url': u'http://example.com/great'},
- {u'format_id': u'excellent', u'url': u'http://example.com/exc'},
+ {u'format_id': u'meh', u'url': u'http://example.com/meh', 'preference': 1},
+ {u'format_id': u'good', u'url': u'http://example.com/good', 'preference': 2},
+ {u'format_id': u'great', u'url': u'http://example.com/great', 'preference': 3},
+ {u'format_id': u'excellent', u'url': u'http://example.com/exc', 'preference': 4},
]
info_dict = {
u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
def test_format_selection(self):
formats = [
- {u'format_id': u'35', u'ext': u'mp4'},
- {u'format_id': u'45', u'ext': u'webm'},
- {u'format_id': u'47', u'ext': u'webm'},
- {u'format_id': u'2', u'ext': u'flv'},
+ {u'format_id': u'35', u'ext': u'mp4', 'preference': 1},
+ {u'format_id': u'45', u'ext': u'webm', 'preference': 2},
+ {u'format_id': u'47', u'ext': u'webm', 'preference': 3},
+ {u'format_id': u'2', u'ext': u'flv', 'preference': 4},
]
info_dict = {u'formats': formats, u'extractor': u'test'}
downloaded = ydl.downloaded_info_dicts[0]
self.assertEqual(downloaded['format_id'], u'35')
+ def test_youtube_format_selection(self):
+ order = [
+ '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '36', '17', '13',
+ # Apple HTTP Live Streaming
+ '96', '95', '94', '93', '92', '132', '151',
+ # 3D
+ '85', '84', '102', '83', '101', '82', '100',
+ # Dash video
+ '138', '137', '248', '136', '247', '135', '246',
+ '245', '244', '134', '243', '133', '242', '160',
+ # Dash audio
+ '141', '172', '140', '139', '171',
+ ]
+
+ for f1id, f2id in zip(order, order[1:]):
+ f1 = YoutubeIE._formats[f1id].copy()
+ f1['format_id'] = f1id
+ f2 = YoutubeIE._formats[f2id].copy()
+ f2['format_id'] = f2id
+
+ info_dict = {'formats': [f1, f2], 'extractor': 'youtube'}
+ ydl = YDL()
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
+ ydl.process_ie_result(info_dict)
+ downloaded = ydl.downloaded_info_dicts[0]
+ self.assertEqual(downloaded['format_id'], f1id)
+
+ info_dict = {'formats': [f2, f1], 'extractor': 'youtube'}
+ ydl = YDL()
+ yie = YoutubeIE(ydl)
+ yie._sort_formats(info_dict['formats'])
+ ydl.process_ie_result(info_dict)
+ downloaded = ydl.downloaded_info_dicts[0]
+ self.assertEqual(downloaded['format_id'], f1id)
+
def test_add_extra_info(self):
test_dict = {
'extractor': 'Foo',
formats = list(takewhile_inclusive(
lambda f: f['format_id'] != format_limit, formats
))
- if self.params.get('prefer_free_formats'):
- def _free_formats_key(f):
- try:
- ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
- except ValueError:
- ext_ord = -1
- # We only compare the extension if they have the same height and width
- return (f.get('height') if f.get('height') is not None else -1,
- f.get('width') if f.get('width') is not None else -1,
- ext_ord)
- formats = sorted(formats, key=_free_formats_key)
+
+ # TODO Central sorting goes here
if formats[0] is not info_dict:
# only set the 'formats' fields if the original info_dict list them
def format_resolution(format, default='unknown'):
if format.get('vcodec') == 'none':
return 'audio only'
- if format.get('_resolution') is not None:
- return format['_resolution']
+ if format.get('resolution') is not None:
+ return format['resolution']
if format.get('height') is not None:
if format.get('width') is not None:
res = u'%sx%s' % (format['width'], format['height'])
else:
res = u'%sp' % format['height']
+ elif format.get('width') is not None:
+ res = u'?x%d' % format['width']
else:
res = default
return res
def list_formats(self, info_dict):
def format_note(fdict):
res = u''
+ if f.get('ext') in ['f4f', 'f4m']:
+ res += u'(unsupported) '
if fdict.get('format_note') is not None:
res += fdict['format_note'] + u' '
+ if fdict.get('tbr') is not None:
+ res += u'%4dk ' % fdict['tbr']
if (fdict.get('vcodec') is not None and
fdict.get('vcodec') != 'none'):
- res += u'%-5s' % fdict['vcodec']
- elif fdict.get('vbr') is not None:
- res += u'video'
+ res += u'%-5s@' % fdict['vcodec']
+ elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
+ res += u'video@'
if fdict.get('vbr') is not None:
- res += u'@%4dk' % fdict['vbr']
+ res += u'%4dk' % fdict['vbr']
if fdict.get('acodec') is not None:
if res:
res += u', '
header_line = line({
'format_id': u'format code', 'ext': u'extension',
- '_resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
+ 'resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
(info_dict['id'], header_line, u"\n".join(formats_s)))
'width': format['width'],
'height': int(format['height']),
})
- formats = sorted(formats, key=lambda f: (f['height'], f['width']))
+
+ self._sort_formats(formats)
playlist.append({
'_type': 'video',
from ..utils import (
compat_http_client,
compat_urllib_error,
+ compat_urllib_parse_urlparse,
compat_str,
clean_html,
id: Video identifier.
title: Video title, unescaped.
- Additionally, it must contain either a formats entry or url and ext:
+ Additionally, it must contain either a formats entry or a url one:
- formats: A list of dictionaries for each format available, it must
- be ordered from worst to best quality. Potential fields:
+ formats: A list of dictionaries for each format available, ordered
+ from worst to best quality.
+
+ Potential fields:
* url Mandatory. The URL of the video file
* ext Will be calculated from url if missing
* format A human-readable description of the format
("3D" or "DASH video")
* width Width of the video, if known
* height Height of the video, if known
+ * resolution Textual description of width and height
+ * tbr Average bitrate of audio and video in KBit/s
* abr Average audio bitrate in KBit/s
* acodec Name of the audio codec in use
* vbr Average video bitrate in KBit/s
* vcodec Name of the video codec in use
* filesize The number of bytes, if known in advance
* player_url SWF Player URL (used for rtmpdump).
+ * protocol The protocol that will be used for the actual
+ download, lower-case.
+ "http", "https", "rtsp", "rtmp" or so.
+ * preference Order number of this format. If this field is
+ present, the formats get sorted by this field.
+ -1 for default (order by other properties),
+ -2 or smaller for less than default.
url: Final video URL.
ext: Video filename extension.
format: The video format, defaults to ext (used for --get-format)
}
return RATING_TABLE.get(rating.lower(), None)
+ def _sort_formats(self, formats):
+ def _formats_key(f):
+ # TODO remove the following workaround
+ from ..utils import determine_ext
+ if not f.get('ext') and 'url' in f:
+ f['ext'] = determine_ext(f['url'])
+
+ preference = f.get('preference')
+ if preference is None:
+ proto = f.get('protocol')
+ if proto is None:
+ proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme
+
+ preference = 0 if proto in ['http', 'https'] else -0.1
+ if f.get('ext') in ['f4f', 'f4m']: # Not yet supported
+ preference -= 0.5
+
+ if f.get('vcodec') == 'none': # audio only
+ if self._downloader.params.get('prefer_free_formats'):
+ ORDER = [u'aac', u'mp3', u'm4a', u'webm', u'ogg', u'opus']
+ else:
+ ORDER = [u'webm', u'opus', u'ogg', u'mp3', u'aac', u'm4a']
+ ext_preference = 0
+ try:
+ audio_ext_preference = ORDER.index(f['ext'])
+ except ValueError:
+ audio_ext_preference = -1
+ else:
+ if self._downloader.params.get('prefer_free_formats'):
+ ORDER = [u'flv', u'mp4', u'webm']
+ else:
+ ORDER = [u'webm', u'flv', u'mp4']
+ try:
+ ext_preference = ORDER.index(f['ext'])
+ except ValueError:
+ ext_preference = -1
+ audio_ext_preference = 0
+
+ return (
+ preference,
+ f.get('height') if f.get('height') is not None else -1,
+ f.get('width') if f.get('width') is not None else -1,
+ ext_preference,
+ f.get('vbr') if f.get('vbr') is not None else -1,
+ f.get('abr') if f.get('abr') is not None else -1,
+ audio_ext_preference,
+ f.get('filesize') if f.get('filesize') is not None else -1,
+ f.get('format_id'),
+ )
+ formats.sort(key=_formats_key)
class SearchInfoExtractor(InfoExtractor):
'width': int(fe.find('./width').text),
'height': int(fe.find('./height').text),
'url': fe.find('./url').text,
- 'ext': determine_ext(fe.find('./url').text),
'filesize': int(fe.find('./filesize').text),
'video_bitrate': int(fe.find('./videoBitrate').text),
- '3sat_qualityname': fe.find('./quality').text,
} for fe in format_els
if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
- def _sortkey(format):
- qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
- prefer_http = 1 if 'rtmp' in format['url'] else 0
- return (qidx, prefer_http, format['video_bitrate'])
- formats.sort(key=_sortkey)
+ self._sort_formats(formats)
return {
'_type': 'video',
'format_id': u'%s-%d' % (media_type, vbr),
})
formats.append(format)
- formats.sort(key=lambda f: (f.get('vbr'), f['abr']))
if not formats:
raise ExtractorError(u'Could not find any valid formats')
+ self._sort_formats(formats)
+
return {
'id': video_id,
'title': title,
from .common import InfoExtractor
from ..utils import (
+ compat_str,
clean_html,
get_element_by_id,
)
raw_page, u'base url')
formats_json = self._search_regex(r'bitrates: (\[.+?\])', raw_page,
u'video formats')
- formats = json.loads(formats_json)
- formats = sorted(formats, key=lambda f: f['bitrate'])
+ formats_mit = json.loads(formats_json)
+ formats = [
+ {
+ 'format_id': f['label'],
+ 'url': base_url + f['url'].partition(':')[2],
+ 'ext': f['url'].partition(':')[0],
+ 'format': f['label'],
+ 'width': f['width'],
+ 'vbr': f['bitrate'],
+ }
+ for f in formats_mit
+ ]
title = get_element_by_id('edit-title', clean_page)
description = clean_html(get_element_by_id('edit-description', clean_page))
return {'id': video_id,
'title': title,
- 'url': base_url + formats[-1]['url'].replace('mp4:', ''),
- 'ext': 'mp4',
+ 'formats': formats,
'description': description,
'thumbnail': thumbnail,
}
# Blacklist type 6, it's extremely LQ and not available on the same server
if n.tag.startswith('type') and n.tag != 'type6'
]
- formats.sort(key=lambda f: f['vbr'])
duration = float(idoc[0].findall('./duration')[0].text)
+ self._sort_formats(formats)
+
info = {
'id': video_id,
'title': video_title,
'filesize': a['size'],
'ext': a['ext'],
})
- formats.sort(key=lambda a: a['filesize'])
+
+ self._sort_formats(formats)
return {
'id': video_id,
from ..utils import (
compat_urllib_parse,
compat_urlparse,
- determine_ext,
clean_html,
+ int_or_none,
)
formats = []
for s in info['streams']:
format_info = {
- 'width': s.get('width'),
- 'height': s.get('height'),
- 'bitrate': s.get('bitrate'),
+ 'width': int_or_none(s.get('width')),
+ 'height': int_or_none(s.get('height')),
+ 'tbr': int_or_none(s.get('bitrate')),
}
host = s['host']
else:
format_url = compat_urlparse.urljoin(host, path)
format_info['url'] = format_url
- format_info['ext'] = determine_ext(format_url)
formats.append(format_info)
- formats = sorted(formats, key=lambda f:(f['height'], f['width']))
+
+ self._sort_formats(formats)
return {
'id': video_id,
(?(1).+)? # if we found the ID, everything can follow
$"""
_NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
- # Listed in order of quality
- _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13',
- # Apple HTTP Live Streaming
- '96', '95', '94', '93', '92', '132', '151',
- # 3D
- '85', '84', '102', '83', '101', '82', '100',
- # Dash video
- '138', '137', '248', '136', '247', '135', '246',
- '245', '244', '134', '243', '133', '242', '160',
- # Dash audio
- '141', '172', '140', '171', '139',
- ]
- _video_extensions = {
- '13': '3gp',
- '17': '3gp',
- '18': 'mp4',
- '22': 'mp4',
- '36': '3gp',
- '37': 'mp4',
- '38': 'mp4',
- '43': 'webm',
- '44': 'webm',
- '45': 'webm',
- '46': 'webm',
+ _formats = {
+ '5': {'ext': 'flv', 'width': 400, 'height': 240},
+ '6': {'ext': 'flv', 'width': 450, 'height': 270},
+ '13': {'ext': '3gp'},
+ '17': {'ext': '3gp', 'width': 176, 'height': 144},
+ '18': {'ext': 'mp4', 'width': 640, 'height': 360},
+ '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
+ '34': {'ext': 'flv', 'width': 640, 'height': 360},
+ '35': {'ext': 'flv', 'width': 854, 'height': 480},
+ '36': {'ext': '3gp', 'width': 320, 'height': 240},
+ '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
+ '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
+ '43': {'ext': 'webm', 'width': 640, 'height': 360},
+ '44': {'ext': 'webm', 'width': 854, 'height': 480},
+ '45': {'ext': 'webm', 'width': 1280, 'height': 720},
+ '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
+
# 3d videos
- '82': 'mp4',
- '83': 'mp4',
- '84': 'mp4',
- '85': 'mp4',
- '100': 'webm',
- '101': 'webm',
- '102': 'webm',
+ '82': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20},
+ '83': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20},
+ '84': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20},
+ '85': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': '3D', 'preference': -20},
+ '100': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20},
+ '101': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20},
+ '102': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20},
# Apple HTTP Live Streaming
- '92': 'mp4',
- '93': 'mp4',
- '94': 'mp4',
- '95': 'mp4',
- '96': 'mp4',
- '132': 'mp4',
- '151': 'mp4',
-
- # Dash mp4
- '133': 'mp4',
- '134': 'mp4',
- '135': 'mp4',
- '136': 'mp4',
- '137': 'mp4',
- '138': 'mp4',
- '160': 'mp4',
+ '92': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10},
+ '93': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'HLS', 'preference': -10},
+ '94': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'HLS', 'preference': -10},
+ '95': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'HLS', 'preference': -10},
+ '96': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'HLS', 'preference': -10},
+ '132': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10},
+ '151': {'ext': 'mp4', 'height': 72, 'resolution': '72p', 'format_note': 'HLS', 'preference': -10},
+
+ # DASH mp4 video
+ '133': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'DASH video', 'preference': -40},
+ '134': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'DASH video', 'preference': -40},
+ '135': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'DASH video', 'preference': -40},
+ '136': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'DASH video', 'preference': -40},
+ '137': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40},
+ '138': {'ext': 'mp4', 'height': 1081, 'resolution': '>1080p', 'format_note': 'DASH video', 'preference': -40},
+ '160': {'ext': 'mp4', 'height': 192, 'resolution': '192p', 'format_note': 'DASH video', 'preference': -40},
# Dash mp4 audio
- '139': 'm4a',
- '140': 'm4a',
- '141': 'm4a',
+ '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
+ '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
+ '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
# Dash webm
- '171': 'webm',
- '172': 'webm',
- '242': 'webm',
- '243': 'webm',
- '244': 'webm',
- '245': 'webm',
- '246': 'webm',
- '247': 'webm',
- '248': 'webm',
- }
- _video_dimensions = {
- '5': {'width': 400, 'height': 240},
- '6': {},
- '13': {},
- '17': {'width': 176, 'height': 144},
- '18': {'width': 640, 'height': 360},
- '22': {'width': 1280, 'height': 720},
- '34': {'width': 640, 'height': 360},
- '35': {'width': 854, 'height': 480},
- '36': {'width': 320, 'height': 240},
- '37': {'width': 1920, 'height': 1080},
- '38': {'width': 4096, 'height': 3072},
- '43': {'width': 640, 'height': 360},
- '44': {'width': 854, 'height': 480},
- '45': {'width': 1280, 'height': 720},
- '46': {'width': 1920, 'height': 1080},
- '82': {'height': 360, 'display': '360p'},
- '83': {'height': 480, 'display': '480p'},
- '84': {'height': 720, 'display': '720p'},
- '85': {'height': 1080, 'display': '1080p'},
- '92': {'height': 240, 'display': '240p'},
- '93': {'height': 360, 'display': '360p'},
- '94': {'height': 480, 'display': '480p'},
- '95': {'height': 720, 'display': '720p'},
- '96': {'height': 1080, 'display': '1080p'},
- '100': {'height': 360, 'display': '360p'},
- '101': {'height': 480, 'display': '480p'},
- '102': {'height': 720, 'display': '720p'},
- '132': {'height': 240, 'display': '240p'},
- '151': {'height': 72, 'display': '72p'},
- '133': {'height': 240, 'display': '240p'},
- '134': {'height': 360, 'display': '360p'},
- '135': {'height': 480, 'display': '480p'},
- '136': {'height': 720, 'display': '720p'},
- '137': {'height': 1080, 'display': '1080p'},
- '138': {'height': 1081, 'display': '>1080p'},
- '139': {'display': '48k'},
- '140': {'display': '128k'},
- '141': {'display': '256k'},
- '160': {'height': 192, 'display': '192p'},
- '171': {'display': '128k'},
- '172': {'display': '256k'},
- '242': {'height': 240, 'display': '240p'},
- '243': {'height': 360, 'display': '360p'},
- '244': {'height': 480, 'display': '480p'},
- '245': {'height': 480, 'display': '480p'},
- '246': {'height': 480, 'display': '480p'},
- '247': {'height': 720, 'display': '720p'},
- '248': {'height': 1080, 'display': '1080p'},
- }
- _special_itags = {
- '82': '3D',
- '83': '3D',
- '84': '3D',
- '85': '3D',
- '100': '3D',
- '101': '3D',
- '102': '3D',
- '133': 'DASH Video',
- '134': 'DASH Video',
- '135': 'DASH Video',
- '136': 'DASH Video',
- '137': 'DASH Video',
- '138': 'DASH Video',
- '139': 'DASH Audio',
- '140': 'DASH Audio',
- '141': 'DASH Audio',
- '160': 'DASH Video',
- '171': 'DASH Audio',
- '172': 'DASH Audio',
- '242': 'DASH Video',
- '243': 'DASH Video',
- '244': 'DASH Video',
- '245': 'DASH Video',
- '246': 'DASH Video',
- '247': 'DASH Video',
- '248': 'DASH Video',
+ '242': {'ext': 'webm', 'height': 240, 'resolution': '240p', 'format_note': 'DASH webm', 'preference': -40},
+ '243': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': 'DASH webm', 'preference': -40},
+ '244': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
+ '245': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
+ '246': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
+ '247': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': 'DASH webm', 'preference': -40},
+ '248': {'ext': 'webm', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH webm', 'preference': -40},
+
+ # Dash webm audio
+ '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50},
+ '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 256, 'preference': -50},
}
IE_NAME = u'youtube'
Transform a dictionary in the format {itag:url} to a list of (itag, url)
with the requested formats.
"""
- existing_formats = [x for x in self._available_formats if x in url_map]
+ existing_formats = [x for x in self._formats if x in url_map]
if len(existing_formats) == 0:
raise ExtractorError(u'no known formats available for video')
video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
formats = []
for itag, video_real_url in video_url_list:
- # Extension
- video_extension = self._video_extensions.get(itag, 'flv')
- resolution = self._video_dimensions.get(itag, {}).get('display')
- width = self._video_dimensions.get(itag, {}).get('width')
- height = self._video_dimensions.get(itag, {}).get('height')
- note = self._special_itags.get(itag)
-
- video_format = '{0} - {1}{2}'.format(itag if itag else video_extension,
- '%dx%d' % (width, height) if width is not None and height is not None else (resolution if resolution is not None else '???'),
- ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '')
-
- formats.append({
- 'url': video_real_url,
- 'ext': video_extension,
- 'format': video_format,
- 'format_id': itag,
- 'player_url': player_url,
- '_resolution': resolution,
- 'width': width,
- 'height': height,
- 'format_note': note,
- })
+ dct = {
+ 'format_id': itag,
+ 'url': video_real_url,
+ 'player_url': player_url,
+ }
+ dct.update(self._formats[itag])
+ formats.append(dct)
- def _formats_key(f):
- note = f.get('format_note')
- if note is None:
- note = u''
- is_dash = u'DASH' in note
- return (
- 0 if is_dash else 1,
- f.get('height') if f.get('height') is not None else -1,
- f.get('width') if f.get('width') is not None else -1)
- formats.sort(key=_formats_key)
+ self._sort_formats(formats)
return {
'id': video_id,
''', format_id)
ext = format_m.group('container')
- is_supported = ext != 'f4f'
-
- PROTO_ORDER = ['http', 'rtmp', 'rtsp']
- try:
- proto_pref = -PROTO_ORDER.index(format_m.group('proto'))
- except ValueError:
- proto_pref = -999
+ proto = format_m.group('proto')
quality = fnode.find('./quality').text
- QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low']
- try:
- quality_pref = -QUALITY_ORDER.index(quality)
- except ValueError:
- quality_pref = -999
-
abr = int(fnode.find('./audioBitrate').text) // 1000
vbr = int(fnode.find('./videoBitrate').text) // 1000
- pref = (is_available, is_supported,
- proto_pref, quality_pref, vbr, abr)
format_note = u''
- if not is_supported:
- format_note += u'(unsupported)'
if not format_note:
format_note = None
'height': int(fnode.find('./height').text),
'filesize': int(fnode.find('./filesize').text),
'format_note': format_note,
- '_pref': pref,
+ 'protocol': format_m.group('proto').lower(),
'_available': is_available,
}
format_nodes = doc.findall('.//formitaeten/formitaet')
- formats = sorted(filter(lambda f: f['_available'],
- map(xml_to_format, format_nodes)),
- key=operator.itemgetter('_pref'))
+ formats = list(filter(
+ lambda f: f['_available'],
+ map(xml_to_format, format_nodes)))
+
+ self._sort_formats(formats)
return {
'id': video_id,
class HEADRequest(compat_urllib_request.Request):
def get_method(self):
return "HEAD"
+
+
+def int_or_none(v):
+ return v if v is None else int(v)
projects / youtube-dl / commitdiff