]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/noco.py
8b83e1f760141bc1a6a0ff4c56761e09b6e3d920
[youtubedl] / youtube_dl / extractor / noco.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import time
6 import hashlib
7
8 from .common import InfoExtractor
9 from ..compat import (
10 compat_str,
11 compat_urlparse,
12 )
13 from ..utils import (
14 clean_html,
15 ExtractorError,
16 int_or_none,
17 float_or_none,
18 parse_iso8601,
19 sanitized_Request,
20 urlencode_postdata,
21 )
22
23
24 class NocoIE(InfoExtractor):
25 _VALID_URL = r'https?://(?:(?:www\.)?noco\.tv/emission/|player\.noco\.tv/\?idvideo=)(?P<id>\d+)'
26 _LOGIN_URL = 'https://noco.tv/do.php'
27 _API_URL_TEMPLATE = 'https://api.noco.tv/1.1/%s?ts=%s&tk=%s'
28 _SUB_LANG_TEMPLATE = '&sub_lang=%s'
29 _NETRC_MACHINE = 'noco'
30
31 _TESTS = [
32 {
33 'url': 'http://noco.tv/emission/11538/nolife/ami-ami-idol-hello-france/',
34 'md5': '0a993f0058ddbcd902630b2047ef710e',
35 'info_dict': {
36 'id': '11538',
37 'ext': 'mp4',
38 'title': 'Ami Ami Idol - Hello! France',
39 'description': 'md5:4eaab46ab68fa4197a317a88a53d3b86',
40 'upload_date': '20140412',
41 'uploader': 'Nolife',
42 'uploader_id': 'NOL',
43 'duration': 2851.2,
44 },
45 'skip': 'Requires noco account',
46 },
47 {
48 'url': 'http://noco.tv/emission/12610/lbl42/the-guild/s01e01-wake-up-call',
49 'md5': 'c190f1f48e313c55838f1f412225934d',
50 'info_dict': {
51 'id': '12610',
52 'ext': 'mp4',
53 'title': 'The Guild #1 - Wake-Up Call',
54 'timestamp': 1403863200,
55 'upload_date': '20140627',
56 'uploader': 'LBL42',
57 'uploader_id': 'LBL',
58 'duration': 233.023,
59 },
60 'skip': 'Requires noco account',
61 }
62 ]
63
64 def _real_initialize(self):
65 self._login()
66
67 def _login(self):
68 (username, password) = self._get_login_info()
69 if username is None:
70 return
71
72 login = self._download_json(
73 self._LOGIN_URL, None, 'Logging in as %s' % username,
74 data=urlencode_postdata({
75 'a': 'login',
76 'cookie': '1',
77 'username': username,
78 'password': password,
79 }),
80 headers={
81 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
82 })
83
84 if 'erreur' in login:
85 raise ExtractorError('Unable to login: %s' % clean_html(login['erreur']), expected=True)
86
87 @staticmethod
88 def _ts():
89 return int(time.time() * 1000)
90
91 def _call_api(self, path, video_id, note, sub_lang=None):
92 ts = compat_str(self._ts() + self._ts_offset)
93 tk = hashlib.md5((hashlib.md5(ts.encode('ascii')).hexdigest() + '#8S?uCraTedap6a').encode('ascii')).hexdigest()
94 url = self._API_URL_TEMPLATE % (path, ts, tk)
95 if sub_lang:
96 url += self._SUB_LANG_TEMPLATE % sub_lang
97
98 request = sanitized_Request(url)
99 request.add_header('Referer', self._referer)
100
101 resp = self._download_json(request, video_id, note)
102
103 if isinstance(resp, dict) and resp.get('error'):
104 self._raise_error(resp['error'], resp['description'])
105
106 return resp
107
108 def _raise_error(self, error, description):
109 raise ExtractorError(
110 '%s returned error: %s - %s' % (self.IE_NAME, error, description),
111 expected=True)
112
113 def _real_extract(self, url):
114 video_id = self._match_id(url)
115
116 # Timestamp adjustment offset between server time and local time
117 # must be calculated in order to use timestamps closest to server's
118 # in all API requests (see https://github.com/rg3/youtube-dl/issues/7864)
119 webpage = self._download_webpage(url, video_id)
120
121 player_url = self._search_regex(
122 r'(["\'])(?P<player>https?://noco\.tv/(?:[^/]+/)+NocoPlayer.+?\.swf.*?)\1',
123 webpage, 'noco player', group='player',
124 default='http://noco.tv/cdata/js/player/NocoPlayer-v1.2.40.swf')
125
126 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(player_url).query)
127 ts = int_or_none(qs.get('ts', [None])[0])
128 self._ts_offset = ts - self._ts() if ts else 0
129 self._referer = player_url
130
131 medias = self._call_api(
132 'shows/%s/medias' % video_id,
133 video_id, 'Downloading video JSON')
134
135 show = self._call_api(
136 'shows/by_id/%s' % video_id,
137 video_id, 'Downloading show JSON')[0]
138
139 options = self._call_api(
140 'users/init', video_id,
141 'Downloading user options JSON')['options']
142 audio_lang_pref = options.get('audio_language') or options.get('language', 'fr')
143
144 if audio_lang_pref == 'original':
145 audio_lang_pref = show['original_lang']
146 if len(medias) == 1:
147 audio_lang_pref = list(medias.keys())[0]
148 elif audio_lang_pref not in medias:
149 audio_lang_pref = 'fr'
150
151 qualities = self._call_api(
152 'qualities',
153 video_id, 'Downloading qualities JSON')
154
155 formats = []
156
157 for audio_lang, audio_lang_dict in medias.items():
158 preference = 1 if audio_lang == audio_lang_pref else 0
159 for sub_lang, lang_dict in audio_lang_dict['video_list'].items():
160 for format_id, fmt in lang_dict['quality_list'].items():
161 format_id_extended = 'audio-%s_sub-%s_%s' % (audio_lang, sub_lang, format_id)
162
163 video = self._call_api(
164 'shows/%s/video/%s/%s' % (video_id, format_id.lower(), audio_lang),
165 video_id, 'Downloading %s video JSON' % format_id_extended,
166 sub_lang if sub_lang != 'none' else None)
167
168 file_url = video['file']
169 if not file_url:
170 continue
171
172 if file_url in ['forbidden', 'not found']:
173 popmessage = video['popmessage']
174 self._raise_error(popmessage['title'], popmessage['message'])
175
176 formats.append({
177 'url': file_url,
178 'format_id': format_id_extended,
179 'width': int_or_none(fmt.get('res_width')),
180 'height': int_or_none(fmt.get('res_lines')),
181 'abr': int_or_none(fmt.get('audiobitrate'), 1000),
182 'vbr': int_or_none(fmt.get('videobitrate'), 1000),
183 'filesize': int_or_none(fmt.get('filesize')),
184 'format_note': qualities[format_id].get('quality_name'),
185 'quality': qualities[format_id].get('priority'),
186 'preference': preference,
187 })
188
189 self._sort_formats(formats)
190
191 timestamp = parse_iso8601(show.get('online_date_start_utc'), ' ')
192
193 if timestamp is not None and timestamp < 0:
194 timestamp = None
195
196 uploader = show.get('partner_name')
197 uploader_id = show.get('partner_key')
198 duration = float_or_none(show.get('duration_ms'), 1000)
199
200 thumbnails = []
201 for thumbnail_key, thumbnail_url in show.items():
202 m = re.search(r'^screenshot_(?P<width>\d+)x(?P<height>\d+)$', thumbnail_key)
203 if not m:
204 continue
205 thumbnails.append({
206 'url': thumbnail_url,
207 'width': int(m.group('width')),
208 'height': int(m.group('height')),
209 })
210
211 episode = show.get('show_TT') or show.get('show_OT')
212 family = show.get('family_TT') or show.get('family_OT')
213 episode_number = show.get('episode_number')
214
215 title = ''
216 if family:
217 title += family
218 if episode_number:
219 title += ' #' + compat_str(episode_number)
220 if episode:
221 title += ' - ' + compat_str(episode)
222
223 description = show.get('show_resume') or show.get('family_resume')
224
225 return {
226 'id': video_id,
227 'title': title,
228 'description': description,
229 'thumbnails': thumbnails,
230 'timestamp': timestamp,
231 'uploader': uploader,
232 'uploader_id': uploader_id,
233 'duration': duration,
234 'formats': formats,
235 }