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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
|
# coding: utf-8
from __future__ import unicode_literals
from uuid import uuid4
import re
from .common import InfoExtractor
from ..utils import (
compat_str,
ExtractorError,
sanitized_Request,
urlencode_postdata,
)
class ZattooBaseIE(InfoExtractor):
_NETRC_MACHINE = 'zattoo'
_HOST_URL = 'https://zattoo.com'
_power_guide_hash = None
def _login(self, uuid, session_id):
(username, password) = self._get_login_info()
if not username or not password:
raise ExtractorError(
'A valid %s account is needed to access this media.' % self._NETRC_MACHINE,
expected=True)
login_form = {
'login': username,
'password': password,
'remember': True,
}
request = sanitized_Request(
'%s/zapi/v2/account/login' % self._HOST_URL,
urlencode_postdata(login_form))
request.add_header(
'Referer', '%s/login' % self._HOST_URL)
request.add_header(
'Content-Type', 'application/x-www-form-urlencoded; charset=UTF-8')
request.add_header(
'Cookie', 'uuid=%s; beaker.session.id=%s' % (uuid, session_id))
response = self._request_webpage(
request, None, 'Logging in')
data = self._parse_json(response.read(), None)
return data['session']['power_guide_hash']
def _get_app_token_and_version(self):
host_webpage = self._download_webpage(
self._HOST_URL, None, 'Downloading %s' % self._HOST_URL)
app_token = self._html_search_regex(
r'<script.+window\.appToken\s*=\s*\'(.+)\'', host_webpage, 'app token')
app_version = self._html_search_regex(
r'<!--\w+-(.+?)-', host_webpage, 'app version', default='2.8.2')
return app_token, app_version
def _say_hello(self, uuid, app_token, app_version):
postdata = {
'client_app_token': app_token,
'uuid': uuid,
'lang': 'en',
'app_version': app_version,
'format': 'json',
}
request = sanitized_Request(
'%s/zapi/v2/session/hello' % self._HOST_URL,
urlencode_postdata(postdata))
response = self._request_webpage(
request, None, 'Say hello')
cookie = response.headers.get('Set-Cookie')
session_id = self._search_regex(
r'beaker\.session\.id\s*=\s*(.+?);', cookie, 'session id')
return session_id
def _extract_cid(self, video_id, channel_name):
channel_groups = self._download_json(
'%s/zapi/v2/cached/channels/%s' % (self._HOST_URL,
self._power_guide_hash),
video_id,
'Downloading available channel list',
query={'details': False})['channel_groups']
channel_list = []
for chgrp in channel_groups:
channel_list.extend(chgrp['channels'])
try:
return next(
chan['cid'] for chan in channel_list
if chan['display_alias'] == channel_name or chan['cid'] == channel_name)
except StopIteration:
raise ExtractorError('Could not extract channel id')
def _extract_cid_and_video_info(self, video_id):
data = self._download_json(
'%s/zapi/program/details' % self._HOST_URL,
video_id,
'Downloading video information',
query={
'program_id': video_id,
'complete': True
})
info_dict = {
'id': video_id,
'title': data['program']['title'],
'description': data['program'].get('description'),
'thumbnail': data['program'].get('image_url')
}
cid = data['program']['cid']
return cid, info_dict
def _extract_formats(self, cid, video_id, record_id=None, is_live=False):
postdata = {
'stream_type': 'dash',
'https_watch_urls': True,
}
if record_id:
url = '%s/zapi/watch/recording/%s' % (self._HOST_URL, record_id)
else:
url = '%s/zapi/watch/recall/%s/%s' % (self._HOST_URL, cid, video_id)
if is_live:
postdata.update({'timeshift': 10800})
url = '%s/zapi/watch/live/%s' % (self._HOST_URL, cid)
data = self._download_json(
sanitized_Request(url, urlencode_postdata(postdata)),
video_id, 'Downloading dash formats')
formats = []
for elem in data['stream']['watch_urls']:
audio_channel = elem.get('audio_channel')
maxrate = elem.get('maxrate')
formats.extend(
self._extract_mpd_formats(
elem['url'], video_id,
mpd_id='dash-maxrate-%s-channel-%s' % (maxrate, audio_channel), fatal=False))
postdata.update({'stream_type': 'hls'})
request = sanitized_Request(
url, urlencode_postdata(postdata))
data = self._download_json(
request, video_id, 'Downloading hls formats')
for elem in data['stream']['watch_urls']:
audio_channel = elem.get('audio_channel')
preference = None
# Prefer audio channel A:
if audio_channel == 'A':
preference = 1
maxrate = elem.get('maxrate')
formats.extend(
self._extract_m3u8_formats(
elem['url'], video_id, 'mp4', entry_protocol='m3u8_native',
preference=preference,
m3u8_id='hls-maxrate-%s-channel-%s' % (maxrate, audio_channel),
fatal=False))
self._sort_formats(formats)
return formats
def _real_initialize(self):
uuid = compat_str(uuid4())
app_token, app_version = self._get_app_token_and_version()
session_id = self._say_hello(uuid, app_token, app_version)
self._power_guide_hash = self._login(uuid, session_id)
def _extract_video(self, channel_name, video_id, record_id=None, is_live=False):
if is_live:
cid = self._extract_cid(video_id, channel_name)
info_dict = {
'id': channel_name,
'title': self._live_title(channel_name),
'is_live': True,
}
else:
cid, info_dict = self._extract_cid_and_video_info(video_id)
formats = self._extract_formats(
cid, video_id, record_id=record_id, is_live=is_live)
info_dict['formats'] = formats
return info_dict
class QuicklineBaseIE(ZattooBaseIE):
_NETRC_MACHINE = 'quickline'
_HOST_URL = 'https://mobiltv.quickline.com'
class QuicklineIE(QuicklineBaseIE):
_VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<channel>[^/]+)/(?P<id>[0-9]+)'
def _real_extract(self, url):
channel_name, video_id = re.match(self._VALID_URL, url).groups()
return self._extract_video(channel_name, video_id)
class QuicklineLiveIE(QuicklineBaseIE):
_VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<id>[^/]+)$'
def _real_extract(self, url):
channel_name = video_id = self._match_id(url)
return self._extract_video(channel_name, video_id, is_live=True)
class ZattooIE(ZattooBaseIE):
_VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<channel>[^/]+?)/(?P<id>[0-9]+)[^/]+(?:/(?P<recid>[0-9]+))?'
# Since regular videos are only available for 7 days and recorded videos
# are only available for a specific user, we cannot have detailed tests.
_TESTS = [{
'url': 'https://zattoo.com/watch/prosieben/130671867-maze-runner-die-auserwaehlten-in-der-brandwueste',
'only_matching': True,
}, {
'url': 'https://zattoo.com/watch/srf_zwei/132905652-eishockey-spengler-cup/102791477/1512211800000/1514433500000/92000',
'only_matching': True,
}]
def _real_extract(self, url):
channel_name, video_id, record_id = re.match(self._VALID_URL, url).groups()
return self._extract_video(channel_name, video_id, record_id)
class ZattooLiveIE(ZattooBaseIE):
_VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<id>[^/]+)$'
_TEST = {
'url': 'https://zattoo.com/watch/srf1',
'only_matching': True,
}
def _real_extract(self, url):
channel_name = video_id = self._match_id(url)
return self._extract_video(channel_name, video_id, is_live=True)
|