2014-12-06 10:24:20 -05:00
|
|
|
# coding: utf-8
|
2014-04-05 11:56:36 -04:00
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
2014-07-25 05:46:51 -04:00
|
|
|
import itertools
|
2013-06-23 16:07:27 -04:00
|
|
|
import re
|
2015-01-22 12:34:40 -05:00
|
|
|
import random
|
2013-06-23 16:07:27 -04:00
|
|
|
|
|
|
|
from .common import InfoExtractor
|
2014-12-13 06:24:42 -05:00
|
|
|
from ..compat import (
|
2015-08-02 19:54:21 -04:00
|
|
|
compat_parse_qs,
|
2015-01-22 12:11:22 -05:00
|
|
|
compat_str,
|
2016-03-25 15:46:57 -04:00
|
|
|
compat_urllib_parse_urlencode,
|
2015-08-02 19:54:21 -04:00
|
|
|
compat_urllib_parse_urlparse,
|
2015-08-18 10:26:45 -04:00
|
|
|
compat_urlparse,
|
2014-12-13 06:24:42 -05:00
|
|
|
)
|
|
|
|
from ..utils import (
|
2013-06-23 16:07:27 -04:00
|
|
|
ExtractorError,
|
2015-08-17 10:20:04 -04:00
|
|
|
int_or_none,
|
2016-06-13 13:06:31 -04:00
|
|
|
js_to_json,
|
2016-03-03 11:11:26 -05:00
|
|
|
orderedSet,
|
2015-08-02 19:54:21 -04:00
|
|
|
parse_duration,
|
2014-10-16 11:23:35 -04:00
|
|
|
parse_iso8601,
|
2015-11-21 11:18:17 -05:00
|
|
|
sanitized_Request,
|
2016-03-25 16:19:24 -04:00
|
|
|
urlencode_postdata,
|
2013-06-23 16:07:27 -04:00
|
|
|
)
|
|
|
|
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
class TwitchBaseIE(InfoExtractor):
|
2015-01-22 12:11:22 -05:00
|
|
|
_VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv'
|
2015-01-21 11:27:21 -05:00
|
|
|
|
2014-10-17 11:58:18 -04:00
|
|
|
_API_BASE = 'https://api.twitch.tv'
|
2015-01-22 12:11:22 -05:00
|
|
|
_USHER_BASE = 'http://usher.twitch.tv'
|
2015-10-17 08:28:21 -04:00
|
|
|
_LOGIN_URL = 'http://www.twitch.tv/login'
|
2015-03-03 06:59:17 -05:00
|
|
|
_NETRC_MACHINE = 'twitch'
|
2013-06-23 16:07:27 -04:00
|
|
|
|
2014-10-16 11:23:35 -04:00
|
|
|
def _handle_error(self, response):
|
|
|
|
if not isinstance(response, dict):
|
|
|
|
return
|
|
|
|
error = response.get('error')
|
|
|
|
if error:
|
|
|
|
raise ExtractorError(
|
|
|
|
'%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
|
|
|
|
expected=True)
|
|
|
|
|
|
|
|
def _download_json(self, url, video_id, note='Downloading JSON metadata'):
|
2015-02-28 11:59:55 -05:00
|
|
|
headers = {
|
|
|
|
'Referer': 'http://api.twitch.tv/crossdomain/receiver.html?v=2',
|
|
|
|
'X-Requested-With': 'XMLHttpRequest',
|
|
|
|
}
|
|
|
|
for cookie in self._downloader.cookiejar:
|
|
|
|
if cookie.name == 'api_token':
|
|
|
|
headers['Twitch-Api-Token'] = cookie.value
|
2015-11-21 11:18:17 -05:00
|
|
|
request = sanitized_Request(url, headers=headers)
|
2015-02-28 11:59:55 -05:00
|
|
|
response = super(TwitchBaseIE, self)._download_json(request, video_id, note)
|
2014-10-16 11:23:35 -04:00
|
|
|
self._handle_error(response)
|
|
|
|
return response
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
def _real_initialize(self):
|
|
|
|
self._login()
|
|
|
|
|
|
|
|
def _login(self):
|
|
|
|
(username, password) = self._get_login_info()
|
|
|
|
if username is None:
|
|
|
|
return
|
|
|
|
|
2015-10-17 08:28:21 -04:00
|
|
|
login_page, handle = self._download_webpage_handle(
|
2015-01-21 11:27:21 -05:00
|
|
|
self._LOGIN_URL, None, 'Downloading login page')
|
|
|
|
|
2015-07-14 12:36:30 -04:00
|
|
|
login_form = self._hidden_inputs(login_page)
|
2015-01-21 11:27:21 -05:00
|
|
|
|
2015-07-10 10:55:53 -04:00
|
|
|
login_form.update({
|
2015-10-17 08:28:21 -04:00
|
|
|
'username': username,
|
|
|
|
'password': password,
|
2015-07-10 10:55:53 -04:00
|
|
|
})
|
2015-01-21 11:27:21 -05:00
|
|
|
|
2015-10-17 08:28:21 -04:00
|
|
|
redirect_url = handle.geturl()
|
|
|
|
|
2015-08-18 10:26:45 -04:00
|
|
|
post_url = self._search_regex(
|
|
|
|
r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
|
2015-10-17 08:28:21 -04:00
|
|
|
'post url', default=redirect_url, group='url')
|
2015-08-18 10:26:45 -04:00
|
|
|
|
|
|
|
if not post_url.startswith('http'):
|
2015-10-17 08:28:21 -04:00
|
|
|
post_url = compat_urlparse.urljoin(redirect_url, post_url)
|
2015-08-18 10:26:45 -04:00
|
|
|
|
2015-11-21 11:18:17 -05:00
|
|
|
request = sanitized_Request(
|
2016-03-25 16:19:24 -04:00
|
|
|
post_url, urlencode_postdata(login_form))
|
2015-10-17 08:28:21 -04:00
|
|
|
request.add_header('Referer', redirect_url)
|
2015-01-21 11:27:21 -05:00
|
|
|
response = self._download_webpage(
|
|
|
|
request, None, 'Logging in as %s' % username)
|
|
|
|
|
2015-07-10 11:15:09 -04:00
|
|
|
error_message = self._search_regex(
|
|
|
|
r'<div[^>]+class="subwindow_notice"[^>]*>([^<]+)</div>',
|
|
|
|
response, 'error message', default=None)
|
|
|
|
if error_message:
|
2015-01-21 11:27:21 -05:00
|
|
|
raise ExtractorError(
|
2015-07-10 11:15:09 -04:00
|
|
|
'Unable to login. Twitch said: %s' % error_message, expected=True)
|
2015-01-21 11:27:21 -05:00
|
|
|
|
2015-07-10 11:16:42 -04:00
|
|
|
if '>Reset your password<' in response:
|
|
|
|
self.report_warning('Twitch asks you to reset your password, go to https://secure.twitch.tv/reset/submit')
|
|
|
|
|
2015-03-06 04:53:49 -05:00
|
|
|
def _prefer_source(self, formats):
|
|
|
|
try:
|
|
|
|
source = next(f for f in formats if f['format_id'] == 'Source')
|
|
|
|
source['preference'] = 10
|
|
|
|
except StopIteration:
|
|
|
|
pass # No Source stream present
|
|
|
|
self._sort_formats(formats)
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
|
|
|
|
class TwitchItemBaseIE(TwitchBaseIE):
|
|
|
|
def _download_info(self, item, item_id):
|
|
|
|
return self._extract_info(self._download_json(
|
2014-10-16 11:23:35 -04:00
|
|
|
'%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
|
2015-01-21 11:27:21 -05:00
|
|
|
'Downloading %s info JSON' % self._ITEM_TYPE))
|
2014-12-18 10:25:42 -05:00
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
def _extract_media(self, item_id):
|
|
|
|
info = self._download_info(self._ITEM_SHORTCUT, item_id)
|
2014-10-16 11:23:35 -04:00
|
|
|
response = self._download_json(
|
2015-01-21 11:27:21 -05:00
|
|
|
'%s/api/videos/%s%s' % (self._API_BASE, self._ITEM_SHORTCUT, item_id), item_id,
|
|
|
|
'Downloading %s playlist JSON' % self._ITEM_TYPE)
|
2014-10-16 11:23:35 -04:00
|
|
|
entries = []
|
|
|
|
chunks = response['chunks']
|
|
|
|
qualities = list(chunks.keys())
|
|
|
|
for num, fragment in enumerate(zip(*chunks.values()), start=1):
|
|
|
|
formats = []
|
|
|
|
for fmt_num, fragment_fmt in enumerate(fragment):
|
|
|
|
format_id = qualities[fmt_num]
|
|
|
|
fmt = {
|
|
|
|
'url': fragment_fmt['url'],
|
|
|
|
'format_id': format_id,
|
|
|
|
'quality': 1 if format_id == 'live' else 0,
|
|
|
|
}
|
|
|
|
m = re.search(r'^(?P<height>\d+)[Pp]', format_id)
|
|
|
|
if m:
|
|
|
|
fmt['height'] = int(m.group('height'))
|
|
|
|
formats.append(fmt)
|
|
|
|
self._sort_formats(formats)
|
|
|
|
entry = dict(info)
|
2014-10-18 10:57:24 -04:00
|
|
|
entry['id'] = '%s_%d' % (entry['id'], num)
|
2014-10-16 11:23:35 -04:00
|
|
|
entry['title'] = '%s part %d' % (entry['title'], num)
|
|
|
|
entry['formats'] = formats
|
|
|
|
entries.append(entry)
|
2014-10-17 12:46:53 -04:00
|
|
|
return self.playlist_result(entries, info['id'], info['title'])
|
2014-10-16 11:23:35 -04:00
|
|
|
|
|
|
|
def _extract_info(self, info):
|
|
|
|
return {
|
|
|
|
'id': info['_id'],
|
2015-08-17 10:16:43 -04:00
|
|
|
'title': info.get('title') or 'Untitled Broadcast',
|
2015-08-17 10:20:04 -04:00
|
|
|
'description': info.get('description'),
|
|
|
|
'duration': int_or_none(info.get('length')),
|
|
|
|
'thumbnail': info.get('preview'),
|
|
|
|
'uploader': info.get('channel', {}).get('display_name'),
|
|
|
|
'uploader_id': info.get('channel', {}).get('name'),
|
|
|
|
'timestamp': parse_iso8601(info.get('recorded_at')),
|
|
|
|
'view_count': int_or_none(info.get('views')),
|
2014-10-16 11:23:35 -04:00
|
|
|
}
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
def _real_extract(self, url):
|
|
|
|
return self._extract_media(self._match_id(url))
|
2014-12-06 10:24:20 -05:00
|
|
|
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
class TwitchVideoIE(TwitchItemBaseIE):
|
|
|
|
IE_NAME = 'twitch:video'
|
2015-03-22 10:39:35 -04:00
|
|
|
_VALID_URL = r'%s/[^/]+/b/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE
|
2015-01-21 11:27:21 -05:00
|
|
|
_ITEM_TYPE = 'video'
|
|
|
|
_ITEM_SHORTCUT = 'a'
|
2014-12-06 10:24:20 -05:00
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
_TEST = {
|
|
|
|
'url': 'http://www.twitch.tv/riotgames/b/577357806',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'a577357806',
|
|
|
|
'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG',
|
|
|
|
},
|
|
|
|
'playlist_mincount': 12,
|
2016-05-12 13:57:52 -04:00
|
|
|
'skip': 'HTTP Error 404: Not Found',
|
2015-01-21 11:27:21 -05:00
|
|
|
}
|
2014-12-06 10:24:20 -05:00
|
|
|
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
class TwitchChapterIE(TwitchItemBaseIE):
|
|
|
|
IE_NAME = 'twitch:chapter'
|
2015-03-22 10:39:35 -04:00
|
|
|
_VALID_URL = r'%s/[^/]+/c/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE
|
2015-01-21 11:27:21 -05:00
|
|
|
_ITEM_TYPE = 'chapter'
|
|
|
|
_ITEM_SHORTCUT = 'c'
|
2014-12-06 10:24:20 -05:00
|
|
|
|
2015-01-22 12:18:21 -05:00
|
|
|
_TESTS = [{
|
2015-01-21 11:27:21 -05:00
|
|
|
'url': 'http://www.twitch.tv/acracingleague/c/5285812',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'c5285812',
|
|
|
|
'title': 'ACRL Off Season - Sports Cars @ Nordschleife',
|
|
|
|
},
|
|
|
|
'playlist_mincount': 3,
|
2016-05-12 13:57:52 -04:00
|
|
|
'skip': 'HTTP Error 404: Not Found',
|
2015-01-22 12:18:21 -05:00
|
|
|
}, {
|
|
|
|
'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361',
|
|
|
|
'only_matching': True,
|
|
|
|
}]
|
2015-01-21 11:27:21 -05:00
|
|
|
|
|
|
|
|
|
|
|
class TwitchVodIE(TwitchItemBaseIE):
|
|
|
|
IE_NAME = 'twitch:vod'
|
2015-03-22 10:39:35 -04:00
|
|
|
_VALID_URL = r'%s/[^/]+/v/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE
|
2015-01-21 11:27:21 -05:00
|
|
|
_ITEM_TYPE = 'vod'
|
|
|
|
_ITEM_SHORTCUT = 'v'
|
|
|
|
|
2015-08-17 10:23:52 -04:00
|
|
|
_TESTS = [{
|
2015-08-02 19:54:21 -04:00
|
|
|
'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
|
2015-01-21 11:27:21 -05:00
|
|
|
'info_dict': {
|
2015-06-28 01:31:37 -04:00
|
|
|
'id': 'v6528877',
|
2015-01-21 11:27:21 -05:00
|
|
|
'ext': 'mp4',
|
2015-06-28 01:31:37 -04:00
|
|
|
'title': 'LCK Summer Split - Week 6 Day 1',
|
2015-01-21 11:27:21 -05:00
|
|
|
'thumbnail': 're:^https?://.*\.jpg$',
|
2015-06-28 01:31:37 -04:00
|
|
|
'duration': 17208,
|
|
|
|
'timestamp': 1435131709,
|
|
|
|
'upload_date': '20150624',
|
|
|
|
'uploader': 'Riot Games',
|
|
|
|
'uploader_id': 'riotgames',
|
2015-01-21 11:27:21 -05:00
|
|
|
'view_count': int,
|
2015-08-02 19:54:21 -04:00
|
|
|
'start_time': 310,
|
2015-01-21 11:27:21 -05:00
|
|
|
},
|
|
|
|
'params': {
|
|
|
|
# m3u8 download
|
|
|
|
'skip_download': True,
|
|
|
|
},
|
2015-08-17 10:23:52 -04:00
|
|
|
}, {
|
|
|
|
# Untitled broadcast (title is None)
|
|
|
|
'url': 'http://www.twitch.tv/belkao_o/v/11230755',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'v11230755',
|
|
|
|
'ext': 'mp4',
|
|
|
|
'title': 'Untitled Broadcast',
|
|
|
|
'thumbnail': 're:^https?://.*\.jpg$',
|
|
|
|
'duration': 1638,
|
|
|
|
'timestamp': 1439746708,
|
|
|
|
'upload_date': '20150816',
|
|
|
|
'uploader': 'BelkAO_o',
|
|
|
|
'uploader_id': 'belkao_o',
|
|
|
|
'view_count': int,
|
|
|
|
},
|
|
|
|
'params': {
|
|
|
|
# m3u8 download
|
|
|
|
'skip_download': True,
|
|
|
|
},
|
|
|
|
}]
|
2014-12-06 10:24:20 -05:00
|
|
|
|
2013-06-23 16:07:27 -04:00
|
|
|
def _real_extract(self, url):
|
2015-01-21 11:27:21 -05:00
|
|
|
item_id = self._match_id(url)
|
2015-10-17 08:29:54 -04:00
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
info = self._download_info(self._ITEM_SHORTCUT, item_id)
|
|
|
|
access_token = self._download_json(
|
2015-10-17 08:43:12 -04:00
|
|
|
'%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
|
|
|
|
'Downloading %s access token' % self._ITEM_TYPE)
|
2015-10-17 08:29:54 -04:00
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
formats = self._extract_m3u8_formats(
|
2015-10-17 08:43:12 -04:00
|
|
|
'%s/vod/%s?%s' % (
|
|
|
|
self._USHER_BASE, item_id,
|
2016-03-25 15:46:57 -04:00
|
|
|
compat_urllib_parse_urlencode({
|
2015-10-17 08:29:54 -04:00
|
|
|
'allow_source': 'true',
|
2016-03-07 22:03:24 -05:00
|
|
|
'allow_audio_only': 'true',
|
2015-10-17 08:29:54 -04:00
|
|
|
'allow_spectre': 'true',
|
|
|
|
'player': 'twitchweb',
|
|
|
|
'nauth': access_token['token'],
|
|
|
|
'nauthsig': access_token['sig'],
|
|
|
|
})),
|
2016-06-04 16:19:44 -04:00
|
|
|
item_id, 'mp4', entry_protocol='m3u8_native')
|
2015-10-17 08:29:54 -04:00
|
|
|
|
2015-03-06 04:53:49 -05:00
|
|
|
self._prefer_source(formats)
|
2015-01-21 11:27:21 -05:00
|
|
|
info['formats'] = formats
|
2015-08-02 19:54:21 -04:00
|
|
|
|
|
|
|
parsed_url = compat_urllib_parse_urlparse(url)
|
|
|
|
query = compat_parse_qs(parsed_url.query)
|
|
|
|
if 't' in query:
|
|
|
|
info['start_time'] = parse_duration(query['t'][0])
|
|
|
|
|
2015-01-21 11:27:21 -05:00
|
|
|
return info
|
|
|
|
|
|
|
|
|
|
|
|
class TwitchPlaylistBaseIE(TwitchBaseIE):
|
|
|
|
_PLAYLIST_URL = '%s/kraken/channels/%%s/videos/?offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
|
2016-03-03 11:12:55 -05:00
|
|
|
_PAGE_LIMIT = 100
|
2015-01-21 11:27:21 -05:00
|
|
|
|
|
|
|
def _extract_playlist(self, channel_id):
|
|
|
|
info = self._download_json(
|
|
|
|
'%s/kraken/channels/%s' % (self._API_BASE, channel_id),
|
|
|
|
channel_id, 'Downloading channel info JSON')
|
|
|
|
channel_name = info.get('display_name') or info.get('name')
|
|
|
|
entries = []
|
|
|
|
offset = 0
|
|
|
|
limit = self._PAGE_LIMIT
|
2016-03-03 11:10:36 -05:00
|
|
|
broken_paging_detected = False
|
|
|
|
counter_override = None
|
2015-01-21 11:27:21 -05:00
|
|
|
for counter in itertools.count(1):
|
|
|
|
response = self._download_json(
|
|
|
|
self._PLAYLIST_URL % (channel_id, offset, limit),
|
2016-03-03 11:10:36 -05:00
|
|
|
channel_id,
|
|
|
|
'Downloading %s videos JSON page %s'
|
|
|
|
% (self._PLAYLIST_TYPE, counter_override or counter))
|
2015-01-23 10:58:40 -05:00
|
|
|
page_entries = self._extract_playlist_page(response)
|
|
|
|
if not page_entries:
|
2015-01-21 11:27:21 -05:00
|
|
|
break
|
2016-03-03 11:10:36 -05:00
|
|
|
total = int_or_none(response.get('_total'))
|
|
|
|
# Since the beginning of March 2016 twitch's paging mechanism
|
|
|
|
# is completely broken on the twitch side. It simply ignores
|
|
|
|
# a limit and returns the whole offset number of videos.
|
|
|
|
# Working around by just requesting all videos at once.
|
2016-03-15 11:45:43 -04:00
|
|
|
# Upd: pagination bug was fixed by twitch on 15.03.2016.
|
2016-03-03 11:10:36 -05:00
|
|
|
if not broken_paging_detected and total and len(page_entries) > limit:
|
|
|
|
self.report_warning(
|
2016-03-15 11:45:43 -04:00
|
|
|
'Twitch pagination is broken on twitch side, requesting all videos at once',
|
2016-03-03 11:10:36 -05:00
|
|
|
channel_id)
|
|
|
|
broken_paging_detected = True
|
2016-03-03 11:18:32 -05:00
|
|
|
offset = total
|
2016-03-03 11:10:36 -05:00
|
|
|
counter_override = '(all at once)'
|
|
|
|
continue
|
2015-01-23 10:58:40 -05:00
|
|
|
entries.extend(page_entries)
|
2016-03-03 11:10:36 -05:00
|
|
|
if broken_paging_detected or total and len(page_entries) >= total:
|
|
|
|
break
|
2015-01-21 11:27:21 -05:00
|
|
|
offset += limit
|
2015-01-23 10:58:40 -05:00
|
|
|
return self.playlist_result(
|
2016-03-03 11:11:26 -05:00
|
|
|
[self.url_result(entry) for entry in orderedSet(entries)],
|
2015-01-23 10:58:40 -05:00
|
|
|
channel_id, channel_name)
|
|
|
|
|
|
|
|
def _extract_playlist_page(self, response):
|
|
|
|
videos = response.get('videos')
|
|
|
|
return [video['url'] for video in videos] if videos else []
|
2015-01-21 11:27:21 -05:00
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
return self._extract_playlist(self._match_id(url))
|
|
|
|
|
|
|
|
|
|
|
|
class TwitchProfileIE(TwitchPlaylistBaseIE):
|
|
|
|
IE_NAME = 'twitch:profile'
|
|
|
|
_VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
|
|
|
|
_PLAYLIST_TYPE = 'profile'
|
|
|
|
|
|
|
|
_TEST = {
|
|
|
|
'url': 'http://www.twitch.tv/vanillatv/profile',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'vanillatv',
|
|
|
|
'title': 'VanillaTV',
|
|
|
|
},
|
|
|
|
'playlist_mincount': 412,
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE):
|
2015-01-21 12:06:16 -05:00
|
|
|
IE_NAME = 'twitch:past_broadcasts'
|
2015-01-21 11:27:21 -05:00
|
|
|
_VALID_URL = r'%s/(?P<id>[^/]+)/profile/past_broadcasts/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
|
|
|
|
_PLAYLIST_URL = TwitchPlaylistBaseIE._PLAYLIST_URL + '&broadcasts=true'
|
|
|
|
_PLAYLIST_TYPE = 'past broadcasts'
|
|
|
|
|
|
|
|
_TEST = {
|
|
|
|
'url': 'http://www.twitch.tv/spamfish/profile/past_broadcasts',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'spamfish',
|
|
|
|
'title': 'Spamfish',
|
|
|
|
},
|
|
|
|
'playlist_mincount': 54,
|
|
|
|
}
|
2015-01-22 12:11:22 -05:00
|
|
|
|
|
|
|
|
|
|
|
class TwitchStreamIE(TwitchBaseIE):
|
|
|
|
IE_NAME = 'twitch:stream'
|
2015-07-19 16:16:07 -04:00
|
|
|
_VALID_URL = r'%s/(?P<id>[^/#?]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
|
2015-01-22 12:11:22 -05:00
|
|
|
|
2015-07-19 16:16:07 -04:00
|
|
|
_TESTS = [{
|
2015-01-22 12:11:22 -05:00
|
|
|
'url': 'http://www.twitch.tv/shroomztv',
|
|
|
|
'info_dict': {
|
|
|
|
'id': '12772022048',
|
|
|
|
'display_id': 'shroomztv',
|
|
|
|
'ext': 'mp4',
|
|
|
|
'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
|
|
|
|
'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
|
|
|
|
'is_live': True,
|
|
|
|
'timestamp': 1421928037,
|
|
|
|
'upload_date': '20150122',
|
|
|
|
'uploader': 'ShroomzTV',
|
|
|
|
'uploader_id': 'shroomztv',
|
|
|
|
'view_count': int,
|
|
|
|
},
|
|
|
|
'params': {
|
|
|
|
# m3u8 download
|
|
|
|
'skip_download': True,
|
|
|
|
},
|
2015-07-19 16:16:07 -04:00
|
|
|
}, {
|
|
|
|
'url': 'http://www.twitch.tv/miracle_doto#profile-0',
|
|
|
|
'only_matching': True,
|
|
|
|
}]
|
2015-01-22 12:11:22 -05:00
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
channel_id = self._match_id(url)
|
|
|
|
|
|
|
|
stream = self._download_json(
|
|
|
|
'%s/kraken/streams/%s' % (self._API_BASE, channel_id), channel_id,
|
|
|
|
'Downloading stream JSON').get('stream')
|
|
|
|
|
|
|
|
# Fallback on profile extraction if stream is offline
|
|
|
|
if not stream:
|
|
|
|
return self.url_result(
|
|
|
|
'http://www.twitch.tv/%s/profile' % channel_id,
|
|
|
|
'TwitchProfile', channel_id)
|
|
|
|
|
2015-07-17 20:27:45 -04:00
|
|
|
# Channel name may be typed if different case than the original channel name
|
|
|
|
# (e.g. http://www.twitch.tv/TWITCHPLAYSPOKEMON) that will lead to constructing
|
|
|
|
# an invalid m3u8 URL. Working around by use of original channel name from stream
|
|
|
|
# JSON and fallback to lowercase if it's not available.
|
2015-07-17 20:18:03 -04:00
|
|
|
channel_id = stream.get('channel', {}).get('name') or channel_id.lower()
|
|
|
|
|
2015-01-22 12:11:22 -05:00
|
|
|
access_token = self._download_json(
|
|
|
|
'%s/api/channels/%s/access_token' % (self._API_BASE, channel_id), channel_id,
|
|
|
|
'Downloading channel access token')
|
|
|
|
|
|
|
|
query = {
|
|
|
|
'allow_source': 'true',
|
2016-03-07 22:03:24 -05:00
|
|
|
'allow_audio_only': 'true',
|
2015-01-22 12:34:40 -05:00
|
|
|
'p': random.randint(1000000, 10000000),
|
2015-01-22 12:11:22 -05:00
|
|
|
'player': 'twitchweb',
|
|
|
|
'segment_preference': '4',
|
2015-03-08 06:54:11 -04:00
|
|
|
'sig': access_token['sig'].encode('utf-8'),
|
|
|
|
'token': access_token['token'].encode('utf-8'),
|
2015-01-22 12:11:22 -05:00
|
|
|
}
|
|
|
|
formats = self._extract_m3u8_formats(
|
|
|
|
'%s/api/channel/hls/%s.m3u8?%s'
|
2016-03-25 15:46:57 -04:00
|
|
|
% (self._USHER_BASE, channel_id, compat_urllib_parse_urlencode(query)),
|
2015-01-22 12:11:22 -05:00
|
|
|
channel_id, 'mp4')
|
2015-03-06 04:53:49 -05:00
|
|
|
self._prefer_source(formats)
|
2015-01-22 12:11:22 -05:00
|
|
|
|
|
|
|
view_count = stream.get('viewers')
|
|
|
|
timestamp = parse_iso8601(stream.get('created_at'))
|
|
|
|
|
|
|
|
channel = stream['channel']
|
|
|
|
title = self._live_title(channel.get('display_name') or channel.get('name'))
|
|
|
|
description = channel.get('status')
|
|
|
|
|
|
|
|
thumbnails = []
|
|
|
|
for thumbnail_key, thumbnail_url in stream['preview'].items():
|
|
|
|
m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key)
|
|
|
|
if not m:
|
|
|
|
continue
|
|
|
|
thumbnails.append({
|
|
|
|
'url': thumbnail_url,
|
|
|
|
'width': int(m.group('width')),
|
|
|
|
'height': int(m.group('height')),
|
|
|
|
})
|
|
|
|
|
|
|
|
return {
|
|
|
|
'id': compat_str(stream['_id']),
|
|
|
|
'display_id': channel_id,
|
|
|
|
'title': title,
|
|
|
|
'description': description,
|
|
|
|
'thumbnails': thumbnails,
|
|
|
|
'uploader': channel.get('display_name'),
|
|
|
|
'uploader_id': channel.get('name'),
|
|
|
|
'timestamp': timestamp,
|
|
|
|
'view_count': view_count,
|
|
|
|
'formats': formats,
|
|
|
|
'is_live': True,
|
2015-01-23 15:05:07 -05:00
|
|
|
}
|
2016-06-13 13:06:31 -04:00
|
|
|
|
|
|
|
|
|
|
|
class TwitchClipsIE(InfoExtractor):
|
|
|
|
IE_NAME = 'twitch:clips'
|
|
|
|
_VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
|
|
|
|
|
|
|
|
_TEST = {
|
|
|
|
'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound',
|
|
|
|
'md5': '761769e1eafce0ffebfb4089cb3847cd',
|
|
|
|
'info_dict': {
|
|
|
|
'id': 'AggressiveCobraPoooound',
|
|
|
|
'ext': 'mp4',
|
|
|
|
'title': 'EA Play 2016 Live from the Novo Theatre',
|
|
|
|
'thumbnail': 're:^https?://.*\.jpg',
|
|
|
|
'creator': 'EA',
|
|
|
|
'uploader': 'stereotype_',
|
|
|
|
'uploader_id': 'stereotype_',
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
video_id = self._match_id(url)
|
|
|
|
|
|
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
|
|
|
|
clip = self._parse_json(
|
|
|
|
self._search_regex(
|
|
|
|
r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'),
|
|
|
|
video_id, transform_source=js_to_json)
|
|
|
|
|
|
|
|
video_url = clip['clip_video_url']
|
|
|
|
title = clip['channel_title']
|
|
|
|
|
|
|
|
return {
|
|
|
|
'id': video_id,
|
|
|
|
'url': video_url,
|
|
|
|
'title': title,
|
|
|
|
'thumbnail': self._og_search_thumbnail(webpage),
|
|
|
|
'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'),
|
|
|
|
'uploader': clip.get('curator_login'),
|
|
|
|
'uploader_id': clip.get('curator_display_name'),
|
|
|
|
}
|