Merge pull request #3 from unit193/master

[youtubedl] / youtube_dl / extractor / roosterteeth.py

diff --git a/youtube_dl/extractor/roosterteeth.py b/youtube_dl/extractor/roosterteeth.py
index 857434540d89a41dae80d9b4270e7bd6e6192acb..8883639b290b9712eb94ef05ab41ecaff7b06ee3 100644 (file)
--- a/youtube_dl/extractor/roosterteeth.py
+++ b/youtube_dl/extractor/roosterteeth.py
@@ -1,35 +1,34 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

-import re
-

 from .common import InfoExtractor
 from .common import InfoExtractor

+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+)

 from ..utils import (
     ExtractorError,
     int_or_none,
 from ..utils import (
     ExtractorError,
     int_or_none,

-    strip_or_none,
-    unescapeHTML,
+    str_or_none,

     urlencode_postdata,
 )
 
 
 class RoosterTeethIE(InfoExtractor):
     urlencode_postdata,
 )
 
 
 class RoosterTeethIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/episode/(?P<id>[^/?#&]+)'
-    _LOGIN_URL = 'https://roosterteeth.com/login'
+    _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/(?:episode|watch)/(?P<id>[^/?#&]+)'

     _NETRC_MACHINE = 'roosterteeth'
     _TESTS = [{
         'url': 'http://roosterteeth.com/episode/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
         'md5': 'e2bd7764732d785ef797700a2489f212',
         'info_dict': {
     _NETRC_MACHINE = 'roosterteeth'
     _TESTS = [{
         'url': 'http://roosterteeth.com/episode/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
         'md5': 'e2bd7764732d785ef797700a2489f212',
         'info_dict': {

-            'id': '26576',
+            'id': '9156',

             'display_id': 'million-dollars-but-season-2-million-dollars-but-the-game-announcement',
             'ext': 'mp4',
             'display_id': 'million-dollars-but-season-2-million-dollars-but-the-game-announcement',
             'ext': 'mp4',

-            'title': 'Million Dollars, But...: Million Dollars, But... The Game Announcement',
-            'description': 'md5:0cc3b21986d54ed815f5faeccd9a9ca5',
+            'title': 'Million Dollars, But... The Game Announcement',
+            'description': 'md5:168a54b40e228e79f4ddb141e89fe4f5',

             'thumbnail': r're:^https?://.*\.png$',
             'series': 'Million Dollars, But...',
             'episode': 'Million Dollars, But... The Game Announcement',
             'thumbnail': r're:^https?://.*\.png$',
             'series': 'Million Dollars, But...',
             'episode': 'Million Dollars, But... The Game Announcement',

-            'comment_count': int,

         },
     }, {
         'url': 'http://achievementhunter.roosterteeth.com/episode/off-topic-the-achievement-hunter-podcast-2016-i-didn-t-think-it-would-pass-31',
         },
     }, {
         'url': 'http://achievementhunter.roosterteeth.com/episode/off-topic-the-achievement-hunter-podcast-2016-i-didn-t-think-it-would-pass-31',


@@ -47,102 +46,92 @@ class RoosterTeethIE(InfoExtractor):
         # only available for FIRST members
         'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one',
         'only_matching': True,
         # only available for FIRST members
         'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one',
         'only_matching': True,

+    }, {
+        'url': 'https://roosterteeth.com/watch/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
+        'only_matching': True,

     }]
     }]

+    _EPISODE_BASE_URL = 'https://svod-be.roosterteeth.com/api/v1/episodes/'

 
     def _login(self):
         username, password = self._get_login_info()
         if username is None:
             return
 
 
     def _login(self):
         username, password = self._get_login_info()
         if username is None:
             return
 

-        login_page = self._download_webpage(
-            self._LOGIN_URL, None,
-            note='Downloading login page',
-            errnote='Unable to download login page')
-
-        login_form = self._hidden_inputs(login_page)
-
-        login_form.update({
-            'username': username,
-            'password': password,
-        })
-
-        login_request = self._download_webpage(
-            self._LOGIN_URL, None,
-            note='Logging in',
-            data=urlencode_postdata(login_form),
-            headers={
-                'Referer': self._LOGIN_URL,
-            })
-
-        if not any(re.search(p, login_request) for p in (
-                r'href=["\']https?://(?:www\.)?roosterteeth\.com/logout"',
-                r'>Sign Out<')):
-            error = self._html_search_regex(
-                r'(?s)<div[^>]+class=(["\']).*?\balert-danger\b.*?\1[^>]*>(?:\s*<button[^>]*>.*?</button>)?(?P<error>.+?)</div>',
-                login_request, 'alert', default=None, group='error')
-            if error:
-                raise ExtractorError('Unable to login: %s' % error, expected=True)
-            raise ExtractorError('Unable to log in')
+        try:
+            self._download_json(
+                'https://auth.roosterteeth.com/oauth/token',
+                None, 'Logging in', data=urlencode_postdata({
+                    'client_id': '4338d2b4bdc8db1239360f28e72f0d9ddb1fd01e7a38fbb07b4b1f4ba4564cc5',
+                    'grant_type': 'password',
+                    'username': username,
+                    'password': password,
+                }))
+        except ExtractorError as e:
+            msg = 'Unable to login'
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                resp = self._parse_json(e.cause.read().decode(), None, fatal=False)
+                if resp:
+                    error = resp.get('extra_info') or resp.get('error_description') or resp.get('error')
+                    if error:
+                        msg += ': ' + error
+            self.report_warning(msg)

 
     def _real_initialize(self):
 
     def _real_initialize(self):

+        if self._get_cookies(self._EPISODE_BASE_URL).get('rt_access_token'):
+            return

         self._login()
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
         self._login()
 
     def _real_extract(self, url):
         display_id = self._match_id(url)

-
-        webpage = self._download_webpage(url, display_id)
-
-        episode = strip_or_none(unescapeHTML(self._search_regex(
-            (r'videoTitle\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
-             r'<title>(?P<title>[^<]+)</title>'), webpage, 'title',
-            default=None, group='title')))
-
-        title = strip_or_none(self._og_search_title(
-            webpage, default=None)) or episode
-
-        m3u8_url = self._search_regex(
-            r'file\s*:\s*(["\'])(?P<url>http.+?\.m3u8.*?)\1',
-            webpage, 'm3u8 url', default=None, group='url')
-
-        if not m3u8_url:
-            if re.search(r'<div[^>]+class=["\']non-sponsor', webpage):
-                self.raise_login_required(
-                    '%s is only available for FIRST members' % display_id)
-
-            if re.search(r'<div[^>]+class=["\']golive-gate', webpage):
-                self.raise_login_required('%s is not available yet' % display_id)
-
-            raise ExtractorError('Unable to extract m3u8 URL')
+        api_episode_url = self._EPISODE_BASE_URL + display_id
+
+        try:
+            m3u8_url = self._download_json(
+                api_episode_url + '/videos', display_id,
+                'Downloading video JSON metadata')['data'][0]['attributes']['url']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                if self._parse_json(e.cause.read().decode(), display_id).get('access') is False:
+                    self.raise_login_required(
+                        '%s is only available for FIRST members' % display_id)
+            raise

 
         formats = self._extract_m3u8_formats(
 
         formats = self._extract_m3u8_formats(

-            m3u8_url, display_id, ext='mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls')
+            m3u8_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls')

         self._sort_formats(formats)
 
         self._sort_formats(formats)
 

-        description = strip_or_none(self._og_search_description(webpage))
-        thumbnail = self._proto_relative_url(self._og_search_thumbnail(webpage))
-
-        series = self._search_regex(
-            (r'<h2>More ([^<]+)</h2>', r'<a[^>]+>See All ([^<]+) Videos<'),
-            webpage, 'series', fatal=False)
-
-        comment_count = int_or_none(self._search_regex(
-            r'>Comments \((\d+)\)<', webpage,
-            'comment count', fatal=False))
-
-        video_id = self._search_regex(
-            (r'containerId\s*=\s*["\']episode-(\d+)\1',
-             r'<div[^<]+id=["\']episode-(\d+)'), webpage,
-            'video id', default=display_id)
+        episode = self._download_json(
+            api_episode_url, display_id,
+            'Downloading episode JSON metadata')['data'][0]
+        attributes = episode['attributes']
+        title = attributes.get('title') or attributes['display_title']
+        video_id = compat_str(episode['id'])
+
+        thumbnails = []
+        for image in episode.get('included', {}).get('images', []):
+            if image.get('type') == 'episode_image':
+                img_attributes = image.get('attributes') or {}
+                for k in ('thumb', 'small', 'medium', 'large'):
+                    img_url = img_attributes.get(k)
+                    if img_url:
+                        thumbnails.append({
+                            'id': k,
+                            'url': img_url,
+                        })

 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,

-            'description': description,
-            'thumbnail': thumbnail,
-            'series': series,
-            'episode': episode,
-            'comment_count': comment_count,
+            'description': attributes.get('description') or attributes.get('caption'),
+            'thumbnails': thumbnails,
+            'series': attributes.get('show_title'),
+            'season_number': int_or_none(attributes.get('season_number')),
+            'season_id': attributes.get('season_id'),
+            'episode': title,
+            'episode_number': int_or_none(attributes.get('number')),
+            'episode_id': str_or_none(episode.get('uuid')),

             'formats': formats,
             'formats': formats,

+            'channel_id': attributes.get('channel_id'),
+            'duration': int_or_none(attributes.get('length')),

         }
         }