0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-04 04:07:26 +00:00
yt-dlp/yt_dlp/extractor/ustream.py

273 lines
10 KiB
Python
Raw Normal View History

import random
2013-06-23 20:08:28 +00:00
import re
import urllib.parse
2013-06-23 20:08:28 +00:00
from .common import InfoExtractor
2015-10-05 16:30:38 +00:00
from ..utils import (
ExtractorError,
encode_data_uri,
2015-10-05 16:30:38 +00:00
float_or_none,
int_or_none,
2021-11-06 01:05:24 +00:00
join_nonempty,
mimetype2ext,
str_or_none,
2015-10-05 16:30:38 +00:00
)
2013-06-23 20:08:28 +00:00
class UstreamIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?(?:ustream\.tv|video\.ibm\.com)/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)'
IE_NAME = 'ustream'
_EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?(?:ustream\.tv|video\.ibm\.com)/embed/.+?)\1']
_TESTS = [{
'url': 'http://www.ustream.tv/recorded/20274954',
'md5': '088f151799e8f572f84eb62f17d73e5c',
'info_dict': {
2014-05-29 13:22:36 +00:00
'id': '20274954',
'ext': 'flv',
'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
2015-10-05 16:52:51 +00:00
'description': 'Young Americans for Liberty February 7, 2012 2:28 AM',
'timestamp': 1328577035,
'upload_date': '20120207',
'uploader': 'yaliberty',
'uploader_id': '6780869',
},
}, {
# From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
# Title and uploader available only from params JSON
'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
'md5': '5a2abf40babeac9812ed20ae12d34e10',
'info_dict': {
'id': '59307601',
'ext': 'flv',
'title': '-CG11- Canada Games Figure Skating',
'uploader': 'sportscanadatv',
2015-10-05 16:52:51 +00:00
},
'skip': 'This Pro Broadcaster has chosen to remove this video from the ustream.tv site.',
}, {
'url': 'http://www.ustream.tv/embed/10299409',
'info_dict': {
'id': '10299409',
},
'playlist_count': 3,
}, {
'url': 'http://www.ustream.tv/recorded/91343263',
'info_dict': {
'id': '91343263',
'ext': 'mp4',
'title': 'GitHub Universe - General Session - Day 1',
'upload_date': '20160914',
'description': 'GitHub Universe - General Session - Day 1',
'timestamp': 1473872730,
'uploader': 'wa0dnskeqkr',
'uploader_id': '38977840',
},
'params': {
'skip_download': True, # m3u8 download
},
}, {
'url': 'https://video.ibm.com/embed/recorded/128240221?&autoplay=true&controls=true&volume=100',
'only_matching': True,
}]
2013-06-23 20:08:28 +00:00
def _get_stream_info(self, url, video_id, app_id_ver, extra_note=None):
def num_to_hex(n):
return hex(n)[2:]
rnd = random.randrange
if not extra_note:
extra_note = ''
conn_info = self._download_json(
f'http://r{rnd(1e8)}-1-{video_id}-recorded-lp-live.ums.ustream.tv/1/ustream',
video_id, note='Downloading connection info' + extra_note,
query={
'type': 'viewer',
'appId': app_id_ver[0],
'appVersion': app_id_ver[1],
'rsid': f'{num_to_hex(rnd(1e8))}:{num_to_hex(rnd(1e8))}',
'rpin': f'_rpin.{rnd(1e15)}',
'referrer': url,
'media': video_id,
'application': 'recorded',
})
host = conn_info[0]['args'][0]['host']
connection_id = conn_info[0]['args'][0]['connectionId']
return self._download_json(
f'http://{host}/1/ustream?connectionId={connection_id}',
video_id, note='Downloading stream info' + extra_note)
def _get_streams(self, url, video_id, app_id_ver):
# Sometimes the return dict does not have 'stream'
for trial_count in range(3):
stream_info = self._get_stream_info(
url, video_id, app_id_ver,
extra_note=f' (try {trial_count + 1})' if trial_count > 0 else '')
if 'stream' in stream_info[0]['args'][0]:
return stream_info[0]['args'][0]['stream']
return []
def _parse_segmented_mp4(self, dash_stream_info):
def resolve_dash_template(template, idx, chunk_hash):
return template.replace('%', str(idx), 1).replace('%', chunk_hash)
formats = []
for stream in dash_stream_info['streams']:
# Use only one provider to avoid too many formats
provider = dash_stream_info['providers'][0]
fragments = [{
'url': resolve_dash_template(
provider['url'] + stream['initUrl'], 0, dash_stream_info['hashes']['0']),
}]
for idx in range(dash_stream_info['videoLength'] // dash_stream_info['chunkTime']):
fragments.append({
'url': resolve_dash_template(
provider['url'] + stream['segmentUrl'], idx,
dash_stream_info['hashes'][str(idx // 10 * 10)]),
})
content_type = stream['contentType']
kind = content_type.split('/')[0]
f = {
2021-11-06 01:05:24 +00:00
'format_id': join_nonempty(
'dash', kind, str_or_none(stream.get('bitrate'))),
'protocol': 'http_dash_segments',
# TODO: generate a MPD doc for external players?
'url': encode_data_uri(b'<MPD/>', 'text/xml'),
'ext': mimetype2ext(content_type),
'height': stream.get('height'),
'width': stream.get('width'),
'fragments': fragments,
}
if kind == 'video':
f.update({
'vcodec': stream.get('codec'),
'acodec': 'none',
'vbr': stream.get('bitrate'),
})
else:
f.update({
'vcodec': 'none',
'acodec': stream.get('codec'),
'abr': stream.get('bitrate'),
})
formats.append(f)
return formats
2013-06-23 20:08:28 +00:00
def _real_extract(self, url):
m = self._match_valid_url(url)
2015-10-05 16:40:20 +00:00
video_id = m.group('id')
2019-03-09 12:14:41 +00:00
# some sites use this embed format (see: https://github.com/ytdl-org/youtube-dl/issues/2990)
2014-05-29 13:22:36 +00:00
if m.group('type') == 'embed/recorded':
2015-10-05 16:40:20 +00:00
video_id = m.group('id')
desktop_url = 'http://www.ustream.tv/recorded/' + video_id
return self.url_result(desktop_url, 'Ustream')
if m.group('type') == 'embed':
2015-10-05 16:40:20 +00:00
video_id = m.group('id')
webpage = self._download_webpage(url, video_id)
content_video_ids = self._parse_json(self._search_regex(
r'ustream\.vars\.offAirContentVideoIds=([^;]+);', webpage,
'content video IDs'), video_id)
return self.playlist_result(
(self.url_result('http://www.ustream.tv/recorded/' + u, 'Ustream') for u in content_video_ids),
video_id)
2015-10-05 16:40:20 +00:00
params = self._download_json(
f'https://api.ustream.tv/videos/{video_id}.json', video_id)
2015-10-05 16:30:38 +00:00
error = params.get('error')
if error:
raise ExtractorError(
f'{self.IE_NAME} returned error: {error}', expected=True)
2013-06-23 20:08:28 +00:00
2015-10-05 16:30:38 +00:00
video = params['video']
2015-10-05 16:48:47 +00:00
title = video['title']
filesize = float_or_none(video.get('file_size'))
2015-10-05 16:30:38 +00:00
formats = [{
2015-10-05 16:51:04 +00:00
'id': video_id,
2015-10-05 16:30:38 +00:00
'url': video_url,
'ext': format_id,
2015-10-05 16:48:47 +00:00
'filesize': filesize,
} for format_id, video_url in video['media_urls'].items() if video_url]
if not formats:
hls_streams = self._get_streams(url, video_id, app_id_ver=(11, 2))
if hls_streams:
# m3u8_native leads to intermittent ContentTooShortError
formats.extend(self._extract_m3u8_formats(
hls_streams[0]['url'], video_id, ext='mp4', m3u8_id='hls'))
'''
# DASH streams handling is incomplete as 'url' is missing
dash_streams = self._get_streams(url, video_id, app_id_ver=(3, 1))
if dash_streams:
formats.extend(self._parse_segmented_mp4(dash_streams))
'''
2015-10-05 16:30:38 +00:00
description = video.get('description')
timestamp = int_or_none(video.get('created_at'))
duration = float_or_none(video.get('length'))
view_count = int_or_none(video.get('views'))
2015-10-05 16:30:38 +00:00
uploader = video.get('owner', {}).get('username')
uploader_id = video.get('owner', {}).get('id')
2013-06-23 20:08:28 +00:00
2015-10-05 16:30:38 +00:00
thumbnails = [{
'id': thumbnail_id,
'url': thumbnail_url,
} for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()]
return {
'id': video_id,
2015-10-05 16:30:38 +00:00
'title': title,
'description': description,
'thumbnails': thumbnails,
'timestamp': timestamp,
'duration': duration,
'view_count': view_count,
'uploader': uploader,
2015-10-05 16:30:38 +00:00
'uploader_id': uploader_id,
'formats': formats,
}
2013-09-12 10:30:14 +00:00
class UstreamChannelIE(InfoExtractor):
2016-09-08 11:29:05 +00:00
_VALID_URL = r'https?://(?:www\.)?ustream\.tv/channel/(?P<slug>.+)'
IE_NAME = 'ustream:channel'
_TEST = {
'url': 'http://www.ustream.tv/channel/channeljapan',
'info_dict': {
'id': '10874166',
},
'playlist_mincount': 17,
}
2013-09-12 10:30:14 +00:00
def _real_extract(self, url):
m = self._match_valid_url(url)
display_id = m.group('slug')
webpage = self._download_webpage(url, display_id)
channel_id = self._html_search_meta('ustream:channel_id', webpage)
2013-09-12 10:30:14 +00:00
BASE = 'http://www.ustream.tv'
next_url = f'/ajax/socialstream/videos/{channel_id}/1.json'
video_ids = []
2013-09-12 10:30:14 +00:00
while next_url:
reply = self._download_json(
urllib.parse.urljoin(BASE, next_url), display_id,
note=f'Downloading video information (next: {len(video_ids) + 1})')
video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
2013-09-12 10:30:14 +00:00
next_url = reply['nextUrl']
entries = [
self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
for vid in video_ids]
return {
'_type': 'playlist',
'id': channel_id,
'display_id': display_id,
'entries': entries,
}