[ellentube] Fix extraction (closes #14407)

3 files changed, 145 insertions, 104 deletions

diff --git a/youtube_dl/extractor/ellentube.py b/youtube_dl/extractor/ellentube.py
new file mode 100644
index 0000000..68fe172
--- /dev/null
+++ b/youtube_dl/extractor/ellentube.py
@@ -0,0 +1,140 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    int_or_none,
+)
+
+
+class EllenTubeIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    (?:
+                        https://api-prod\.ellentube\.com/ellenapi/api/item/
+                        |ellentube:
+                    )
+                    (?P<id>
+                        [\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}
+                    )'''
+
+    _TESTS = [{
+        'url': 'https://api-prod.ellentube.com/ellenapi/api/item/75c64c16-aefd-4558-b4f5-3de09b22e6fc',
+        'match_only': True,
+    }, {
+        'url': 'ellentube:734a3353-f697-4e79-9ca9-bfc3002dc1e0',
+        'match_only': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        data = self._download_json(
+            'https://api-prod.ellentube.com/ellenapi/api/item/%s' % video_id, video_id)
+        title = data['title']
+        description = data.get('description')
+        publish_time = int_or_none(data.get('publishTime'))
+        thumbnail = data.get('thumbnail')
+
+        formats = []
+        duration = None
+        for entry in data.get('media'):
+            if entry.get('id') == 'm3u8':
+                formats = self._extract_m3u8_formats(
+                    entry.get('url'), video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
+                duration = int_or_none(entry.get('duration'))
+                break
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'thumbnail': thumbnail,
+            'timestamp': publish_time,
+            'formats': formats,
+        }
+
+
+class EllenTubeVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ellentube\.com/video/(?P<id>.+)\.html'
+
+    _TEST = {
+        'url': 'https://www.ellentube.com/video/ellen-meets-las-vegas-survivors-jesus-campos-and-stephen-schuck.html',
+        'md5': '2fabc277131bddafdd120e0fc0f974c9',
+        'info_dict': {
+            'id': '0822171c-3829-43bf-b99f-d77358ae75e3',
+            'ext': 'mp4',
+            'title': 'Ellen Meets Las Vegas Survivors Jesus Campos and Stephen Schuck',
+            'description': 'md5:76e3355e2242a78ad9e3858e5616923f',
+            'duration': 514,
+            'timestamp': 1508505120000,
+            'thumbnail': 'https://warnerbros-h.assetsadobe.com/is/image/content/dam/ellen/videos/episodes/season15/32/video--2728751654987218111',
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._html_search_regex(
+            r'(?s)<!--\s*CONTENT\s*-->.*data-config.+([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
+            webpage, 'video id')
+        return self.url_result('ellentube:%s' % video_id, 'EllenTube')
+
+
+class EllenTubePlaylistIE(InfoExtractor):
+    def _extract_videos_from_json(self, data, display_id):
+        return [self.url_result('ellentube:%s' % elem['id'], 'EllenTube')
+                for elem in data if elem.get('type') == 'VIDEO']
+
+    def _extract_playlist(self, url, display_id, extract_description=True):
+        webpage = self._download_webpage(url, display_id)
+        playlist_data = self._html_search_regex(
+            r'<div\s+data-component\s*=\s*"Details"(.+)</div>', webpage, 'playlist data')
+        playlist_title = self._search_regex(
+            r'"title"\s*:\s*"(.+?)"', playlist_data, 'playlist title')
+        playlist_description = clean_html(self._search_regex(
+            r'"description"\s*:\s*"(.+?)"', playlist_data, 'playlist description',
+            fatal=False)) if extract_description else None
+        api_search = self._search_regex(
+            r'"filter"\s*:\s*"(.+?)"', playlist_data, 'playlist api request')
+        api_data = self._download_json(
+            'https://api-prod.ellentube.com/ellenapi/api/feed/?%s' % api_search,
+            display_id)
+        return self.playlist_result(
+            self._extract_videos_from_json(api_data, display_id),
+            display_id, playlist_title, playlist_description)
+
+
+class EllenTubeEpisodeIE(EllenTubePlaylistIE):
+    _VALID_URL = r'https?://(?:www\.)?ellentube\.com/episode/(?P<id>.+)\.html'
+
+    _TEST = {
+        'url': 'https://www.ellentube.com/episode/dax-shepard-jordan-fisher-haim.html',
+        'info_dict': {
+            'id': 'dax-shepard-jordan-fisher-haim',
+            'title': 'Dax Shepard, \'DWTS\' Team Jordan Fisher & Lindsay Arnold, HAIM',
+            'description': 'md5:aed85d42892f6126e71ec5ed2aea2a0d'
+        },
+        'playlist_count': 6,
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        return self._extract_playlist(url, display_id)
+
+
+class EllenTubeStudioIE(EllenTubePlaylistIE):
+    _VALID_URL = r'https?://(?:www\.)?ellentube\.com/studios/(?P<id>.+)\.html'
+
+    _TEST = {
+        'url': 'https://www.ellentube.com/studios/macey-goes-rving0.html',
+        'info_dict': {
+            'id': 'macey-goes-rving0',
+            'title': 'Macey Goes RVing',
+        },
+        'playlist_mincount': 3,
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        return self._extract_playlist(url, display_id, False)
diff --git a/youtube_dl/extractor/ellentv.py b/youtube_dl/extractor/ellentv.py
deleted file mode 100644
index e0a13dd..0000000
--- a/youtube_dl/extractor/ellentv.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from .kaltura import KalturaIE
-from ..utils import NO_DEFAULT
-
-
-class EllenTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?:ellentv|ellentube)\.com/videos/(?P<id>[a-z0-9_-]+)'
-    _TESTS = [{
-        'url': 'http://www.ellentv.com/videos/0-ipq1gsai/',
-        'md5': '4294cf98bc165f218aaa0b89e0fd8042',
-        'info_dict': {
-            'id': '0_ipq1gsai',
-            'ext': 'mov',
-            'title': 'Fast Fingers of Fate',
-            'description': 'md5:3539013ddcbfa64b2a6d1b38d910868a',
-            'timestamp': 1428035648,
-            'upload_date': '20150403',
-            'uploader_id': 'batchUser',
-        },
-    }, {
-        # not available via http://widgets.ellentube.com/
-        'url': 'http://www.ellentv.com/videos/1-szkgu2m2/',
-        'info_dict': {
-            'id': '1_szkgu2m2',
-            'ext': 'flv',
-            'title': "Ellen's Amazingly Talented Audience",
-            'description': 'md5:86ff1e376ff0d717d7171590e273f0a5',
-            'timestamp': 1255140900,
-            'upload_date': '20091010',
-            'uploader_id': 'ellenkaltura@gmail.com',
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        URLS = ('http://widgets.ellentube.com/videos/%s' % video_id, url)
-
-        for num, url_ in enumerate(URLS, 1):
-            webpage = self._download_webpage(
-                url_, video_id, fatal=num == len(URLS))
-
-            default = NO_DEFAULT if num == len(URLS) else None
-
-            partner_id = self._search_regex(
-                r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id',
-                default=default)
-
-            kaltura_id = self._search_regex(
-                [r'id="kaltura_player_([^"]+)"',
-                 r"_wb_entry_id\s*:\s*'([^']+)",
-                 r'data-kaltura-entry-id="([^"]+)'],
-                webpage, 'kaltura id', default=default)
-
-            if partner_id and kaltura_id:
-                break
-
-        return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), KalturaIE.ie_key())
-
-
-class EllenTVClipsIE(InfoExtractor):
-    IE_NAME = 'EllenTV:clips'
-    _VALID_URL = r'https?://(?:www\.)?ellentv\.com/episodes/(?P<id>[a-z0-9_-]+)'
-    _TEST = {
-        'url': 'http://www.ellentv.com/episodes/meryl-streep-vanessa-hudgens/',
-        'info_dict': {
-            'id': 'meryl-streep-vanessa-hudgens',
-            'title': 'Meryl Streep, Vanessa Hudgens',
-        },
-        'playlist_mincount': 5,
-    }
-
-    def _real_extract(self, url):
-        playlist_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, playlist_id)
-        playlist = self._extract_playlist(webpage, playlist_id)
-
-        return {
-            '_type': 'playlist',
-            'id': playlist_id,
-            'title': self._og_search_title(webpage),
-            'entries': self._extract_entries(playlist)
-        }
-
-    def _extract_playlist(self, webpage, playlist_id):
-        json_string = self._search_regex(r'playerView.addClips\(\[\{(.*?)\}\]\);', webpage, 'json')
-        return self._parse_json('[{' + json_string + '}]', playlist_id)
-
-    def _extract_entries(self, playlist):
-        return [
-            self.url_result(
-                'kaltura:%s:%s' % (item['kaltura_partner_id'], item['kaltura_entry_id']),
-                KalturaIE.ie_key(), video_id=item['kaltura_entry_id'])
-            for item in playlist]
diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
index b9c97fa..55d2dd1 100644
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -311,9 +311,11 @@ from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .einthusan import EinthusanIE
 from .eitb import EitbIE
-from .ellentv import (
-    EllenTVIE,
-    EllenTVClipsIE,
+from .ellentube import (
+    EllenTubeIE,
+    EllenTubeEpisodeIE,
+    EllenTubeStudioIE,
+    EllenTubeVideoIE,
 )
 from .elpais import ElPaisIE
 from .embedly import EmbedlyIE