Upstream/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-12-15 21:57:53 +01:00
Code Issues Projects Releases Wiki Activity

[twitter] Fix extraction of test_Twitter and test_Twitter_1

Browse Source
This commit is contained in:
Yen Chi Hsuan 2016-03-03 14:24:24 +08:00
parent 0ae937a798
commit b8972bd69d
1 changed files with 51 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

53
youtube_dl/extractor/twitter.py
View File

@ -95,6 +95,8 @@ class TwitterCardIE(TwitterBaseIE):
config = None
formats = []
duration = None
for user_agent in USER_AGENTS:
request = sanitized_Request(url)
request.add_header('User-Agent', user_agent)
@ -109,16 +111,10 @@ class TwitterCardIE(TwitterBaseIE):
config = self._parse_json(self._html_search_regex(
r'data-(?:player-)?config="([^"]+)"', webpage, 'data player config'),
video_id)
if 'playlist' not in config:
vmap_url = config.get('vmapUrl') or config.get('vmap_url')
if vmap_url:
formats.append({
'url': self._get_vmap_video_url(vmap_url, video_id),
})
break # same video regardless of UA
continue
video_url = config['playlist'][0]['source']
playlist = config.get('playlist')
if playlist:
video_url = playlist[0]['source']
f = {
'url': video_url,
@ -131,11 +127,44 @@ class TwitterCardIE(TwitterBaseIE):
'height': int(m.group('height')),
})
formats.append(f)
continue
vmap_url = config.get('vmapUrl') or config.get('vmap_url')
if vmap_url:
formats.append({
'url': self._get_vmap_video_url(vmap_url, video_id),
})
break # same video regardless of UA
media_info = config.get('status', {}).get('entities', [{}])[0].get('mediaInfo', {})
if media_info:
for media_variant in media_info['variants']:
media_url = media_variant['url']
if media_url.endswith('.m3u8'):
formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
elif media_url.endswith('.mpd'):
formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
else:
vbr = int_or_none(media_variant.get('bitRate'), scale=1000)
a_format = {
'url': media_url,
'format_id': 'http-%d' % vbr if vbr else 'http',
'vbr': vbr,
}
# Reported bitRate may be zero
if not a_format['vbr']:
del a_format['vbr']
formats.append(a_format)
duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
break # same video regardless of UA
self._sort_formats(formats)
title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
thumbnail = config.get('posterImageUrl') or config.get('image_src')
duration = float_or_none(config.get('duration'))
duration = float_or_none(config.get('duration')) or duration
return {
'id': video_id,
@ -153,7 +182,6 @@ class TwitterIE(InfoExtractor):
_TESTS = [{
'url': 'https://twitter.com/freethenipple/status/643211948184596480',
# MD5 checksums are different in different places
'info_dict': {
'id': '643211948184596480',
'ext': 'mp4',
@ -164,6 +192,9 @@ class TwitterIE(InfoExtractor):
'uploader': 'FREE THE NIPPLE',
'uploader_id': 'freethenipple',
},
'params': {
'skip_download': True, # requires ffmpeg
},
}, {
'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',