1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
|
from __future__ import unicode_literals
import itertools
import re
from .common import InfoExtractor
from ..utils import (
ExtractorError,
parse_iso8601,
)
class TwitchIE(InfoExtractor):
# TODO: One broadcast may be split into multiple videos. The key
# 'broadcast_id' is the same for all parts, and 'broadcast_part'
# starts at 1 and increases. Can we treat all parts as one video?
_VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?twitch\.tv/
(?:
(?P<channelid>[^/]+)|
(?:(?:[^/]+)/b/(?P<videoid>[^/]+))|
(?:(?:[^/]+)/c/(?P<chapterid>[^/]+))
)
/?(?:\#.*)?$
"""
_PAGE_LIMIT = 100
_API_BASE = 'https://api.twitch.tv'
_TESTS = [{
'url': 'http://www.twitch.tv/riotgames/b/577357806',
'info_dict': {
'id': 'a577357806',
'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG',
},
'playlist_mincount': 12,
}, {
'url': 'http://www.twitch.tv/acracingleague/c/5285812',
'info_dict': {
'id': 'c5285812',
'title': 'ACRL Off Season - Sports Cars @ Nordschleife',
},
'playlist_mincount': 3,
}, {
'url': 'http://www.twitch.tv/vanillatv',
'info_dict': {
'id': 'vanillatv',
'title': 'VanillaTV',
},
'playlist_mincount': 412,
}]
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'):
response = super(TwitchIE, self)._download_json(url, video_id, note)
self._handle_error(response)
return response
def _extract_media(self, item, item_id):
ITEMS = {
'a': 'video',
'c': 'chapter',
}
info = self._extract_info(self._download_json(
'%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
'Downloading %s info JSON' % ITEMS[item]))
response = self._download_json(
'%s/api/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
'Downloading %s playlist JSON' % ITEMS[item])
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)
entry['id'] = '%s_%d' % (entry['id'], num)
entry['title'] = '%s part %d' % (entry['title'], num)
entry['formats'] = formats
entries.append(entry)
return self.playlist_result(entries, info['id'], info['title'])
def _extract_info(self, info):
return {
'id': info['_id'],
'title': info['title'],
'description': info['description'],
'duration': info['length'],
'thumbnail': info['preview'],
'uploader': info['channel']['display_name'],
'uploader_id': info['channel']['name'],
'timestamp': parse_iso8601(info['recorded_at']),
'view_count': info['views'],
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj.group('chapterid'):
return self._extract_media('c', mobj.group('chapterid'))
"""
webpage = self._download_webpage(url, chapter_id)
m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage)
if not m:
raise ExtractorError('Cannot find archive of a chapter')
archive_id = m.group(1)
api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
doc = self._download_xml(
api, chapter_id,
note='Downloading chapter information',
errnote='Chapter information download failed')
for a in doc.findall('.//archive'):
if archive_id == a.find('./id').text:
break
else:
raise ExtractorError('Could not find chapter in chapter information')
video_url = a.find('./video_file_url').text
video_ext = video_url.rpartition('.')[2] or 'flv'
chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id
chapter_info = self._download_json(
chapter_api_url, 'c' + chapter_id,
note='Downloading chapter metadata',
errnote='Download of chapter metadata failed')
bracket_start = int(doc.find('.//bracket_start').text)
bracket_end = int(doc.find('.//bracket_end').text)
# TODO determine start (and probably fix up file)
# youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457
#video_url += '?start=' + TODO:start_timestamp
# bracket_start is 13290, but we want 51670615
self._downloader.report_warning('Chapter detected, but we can just download the whole file. '
'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end)))
info = {
'id': 'c' + chapter_id,
'url': video_url,
'ext': video_ext,
'title': chapter_info['title'],
'thumbnail': chapter_info['preview'],
'description': chapter_info['description'],
'uploader': chapter_info['channel']['display_name'],
'uploader_id': chapter_info['channel']['name'],
}
return info
"""
elif mobj.group('videoid'):
return self._extract_media('a', mobj.group('videoid'))
elif mobj.group('channelid'):
channel_id = mobj.group('channelid')
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
for counter in itertools.count(1):
response = self._download_json(
'%s/kraken/channels/%s/videos/?offset=%d&limit=%d'
% (self._API_BASE, channel_id, offset, limit),
channel_id, 'Downloading channel videos JSON page %d' % counter)
videos = response['videos']
if not videos:
break
entries.extend([self.url_result(video['url'], 'Twitch') for video in videos])
offset += limit
return self.playlist_result(entries, channel_id, channel_name)
|
