[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / atresplayer.py
index 72e83bfc2c1b4fa45080538e4c87339a7a8c530a..c2cec984525fe9d318b4a615fefd01e644ce788a 100644 (file)
+# coding: utf-8
 from __future__ import unicode_literals
 
-import time
-import hmac
+import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
-    compat_str,
-    compat_urllib_request,
-    int_or_none,
-    float_or_none,
-    xpath_text,
     ExtractorError,
+    int_or_none,
+    urlencode_postdata,
 )
 
 
 class AtresPlayerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/television/[^/]+/[^/]+/[^/]+/(?P<id>.+?)_\d+\.html'
+    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/[^/]+/[^/]+/[^/]+/[^/]+/(?P<display_id>.+?)_(?P<id>[0-9a-f]{24})'
+    _NETRC_MACHINE = 'atresplayer'
     _TESTS = [
         {
-            'url': 'http://www.atresplayer.com/television/programas/el-club-de-la-comedia/temporada-4/capitulo-10-especial-solidario-nochebuena_2014122100174.html',
-            'md5': 'efd56753cda1bb64df52a3074f62e38a',
+            'url': 'https://www.atresplayer.com/antena3/series/pequenas-coincidencias/temporada-1/capitulo-7-asuntos-pendientes_5d4aa2c57ed1a88fc715a615/',
             'info_dict': {
-                'id': 'capitulo-10-especial-solidario-nochebuena',
+                'id': '5d4aa2c57ed1a88fc715a615',
                 'ext': 'mp4',
-                'title': 'Especial Solidario de Nochebuena',
-                'description': 'md5:e2d52ff12214fa937107d21064075bf1',
-                'duration': 5527.6,
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'title': 'CapĂ­tulo 7: Asuntos pendientes',
+                'description': 'md5:7634cdcb4d50d5381bedf93efb537fbc',
+                'duration': 3413,
             },
+            'params': {
+                'format': 'bestvideo',
+            },
+            'skip': 'This video is only available for registered users'
         },
         {
-            'url': 'http://www.atresplayer.com/television/series/el-secreto-de-puente-viejo/el-chico-de-los-tres-lunares/capitulo-977-29-12-14_2014122400174.html',
+            'url': 'https://www.atresplayer.com/lasexta/programas/el-club-de-la-comedia/temporada-4/capitulo-10-especial-solidario-nochebuena_5ad08edf986b2855ed47adc4/',
+            'only_matching': True,
+        },
+        {
+            'url': 'https://www.atresplayer.com/antena3/series/el-secreto-de-puente-viejo/el-chico-de-los-tres-lunares/capitulo-977-29-12-14_5ad51046986b2886722ccdea/',
             'only_matching': True,
         },
     ]
-
-    _USER_AGENT = 'Dalvik/1.6.0 (Linux; U; Android 4.3; GT-I9300 Build/JSS15J'
-    _MAGIC = 'QWtMLXs414Yo+c#_+Q#K@NN)'
-    _TIMESTAMP_SHIFT = 30000
-
-    _TIME_API_URL = 'http://servicios.atresplayer.com/api/admin/time.json'
-    _URL_VIDEO_TEMPLATE = 'https://servicios.atresplayer.com/api/urlVideo/{1}/{0}/{1}|{2}|{3}.json'
-    _PLAYER_URL_TEMPLATE = 'https://servicios.atresplayer.com/episode/getplayer.json?episodePk=%s'
-    _EPISODE_URL_TEMPLATE = 'http://www.atresplayer.com/episodexml/%s'
+    _API_BASE = 'https://api.atresplayer.com/'
+
+    def _real_initialize(self):
+        self._login()
+
+    def _handle_error(self, e, code):
+        if isinstance(e.cause, compat_HTTPError) and e.cause.code == code:
+            error = self._parse_json(e.cause.read(), None)
+            if error.get('error') == 'required_registered':
+                self.raise_login_required()
+            raise ExtractorError(error['error_description'], expected=True)
+        raise
+
+    def _login(self):
+        username, password = self._get_login_info()
+        if username is None:
+            return
+
+        self._request_webpage(
+            self._API_BASE + 'login', None, 'Downloading login page')
+
+        try:
+            target_url = self._download_json(
+                'https://account.atresmedia.com/api/login', None,
+                'Logging in', headers={
+                    'Content-Type': 'application/x-www-form-urlencoded'
+                }, data=urlencode_postdata({
+                    'username': username,
+                    'password': password,
+                }))['targetUrl']
+        except ExtractorError as e:
+            self._handle_error(e, 400)
+
+        self._request_webpage(target_url, None, 'Following Target URL')
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
 
-        episode_id = self._search_regex(
-            r'episode="([^"]+)"', webpage, 'episode id')
+        try:
+            episode = self._download_json(
+                self._API_BASE + 'client/v1/player/episode/' + video_id, video_id)
+        except ExtractorError as e:
+            self._handle_error(e, 403)
 
-        timestamp = int_or_none(self._download_webpage(
-            self._TIME_API_URL,
-            video_id, 'Downloading timestamp', fatal=False), 1000, time.time())
-        timestamp_shifted = compat_str(timestamp + self._TIMESTAMP_SHIFT)
-        token = hmac.new(
-            self._MAGIC.encode('ascii'),
-            (episode_id + timestamp_shifted).encode('utf-8')
-        ).hexdigest()
+        title = episode['titulo']
 
         formats = []
-        for fmt in ['windows', 'android_tablet']:
-            request = compat_urllib_request.Request(
-                self._URL_VIDEO_TEMPLATE.format(fmt, episode_id, timestamp_shifted, token))
-            request.add_header('Youtubedl-user-agent', self._USER_AGENT)
-
-            fmt_json = self._download_json(
-                request, video_id, 'Downloading %s video JSON' % fmt)
-
-            result = fmt_json.get('resultDes')
-            if result.lower() != 'ok':
-                raise ExtractorError(
-                    '%s returned error: %s' % (self.IE_NAME, result), expected=True)
-
-            for _, video_url in fmt_json['resultObject'].items():
-                if video_url.endswith('/Manifest'):
-                    formats.extend(self._extract_f4m_formats(video_url[:-9] + '/manifest.f4m', video_id))
-                else:
-                    formats.append({
-                        'url': video_url,
-                        'format_id': 'android',
-                        'preference': 1,
-                    })
+        for source in episode.get('sources', []):
+            src = source.get('src')
+            if not src:
+                continue
+            src_type = source.get('type')
+            if src_type == 'application/vnd.apple.mpegurl':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif src_type == 'application/dash+xml':
+                formats.extend(self._extract_mpd_formats(
+                    src, video_id, mpd_id='dash', fatal=False))
         self._sort_formats(formats)
 
-        player = self._download_json(
-            self._PLAYER_URL_TEMPLATE % episode_id,
-            episode_id)
-
-        path_data = player.get('pathData')
-
-        episode = self._download_xml(
-            self._EPISODE_URL_TEMPLATE % path_data,
-            video_id, 'Downloading episode XML')
-
-        duration = float_or_none(xpath_text(
-            episode, './media/asset/info/technical/contentDuration', 'duration'))
-
-        art = episode.find('./media/asset/info/art')
-        title = xpath_text(art, './name', 'title')
-        description = xpath_text(art, './description', 'description')
-        thumbnail = xpath_text(episode, './media/asset/files/background', 'thumbnail')
+        heartbeat = episode.get('heartbeat') or {}
+        omniture = episode.get('omniture') or {}
+        get_meta = lambda x: heartbeat.get(x) or omniture.get(x)
 
         return {
+            'display_id': display_id,
             'id': video_id,
             'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
+            'description': episode.get('descripcion'),
+            'thumbnail': episode.get('imgPoster'),
+            'duration': int_or_none(episode.get('duration')),
             'formats': formats,
+            'channel': get_meta('channel'),
+            'season': get_meta('season'),
+            'episode_number': int_or_none(get_meta('episodeNumber')),
         }