youtube-dl

commit d64ec1242e9dec03ea2aa86b6e913db78c8619e0
parent 3ec86619e33a3d1e29c14ec053d7e420ac8b62ae
Author: Remita Amine <remitamine@gmail.com>
Date:   Wed,  6 Nov 2019 10:44:19 +0100

[onionstudios] fix extraction

Diffstat:
M
youtube_dl/extractor/onionstudios.py
 | 
76
+++++++++++++++++++++++++++++++++++++++++++++-------------------------------

1 file changed, 45 insertions(+), 31 deletions(-)
diff --git a/youtube_dl/extractor/onionstudios.py b/youtube_dl/extractor/onionstudios.py
@@ -5,10 +5,11 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
+    compat_str,
     int_or_none,
-    float_or_none,
-    mimetype2ext,
+    js_to_json,
+    parse_iso8601,
+    try_get,
 )
 
 
@@ -17,14 +18,16 @@ class OnionStudiosIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937',
-        'md5': '719d1f8c32094b8c33902c17bcae5e34',
+        'md5': '5a118d466d62b5cd03647cf2c593977f',
         'info_dict': {
             'id': '2937',
             'ext': 'mp4',
             'title': 'Hannibal charges forward, stops for a cocktail',
+            'description': 'md5:545299bda6abf87e5ec666548c6a9448',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'uploader': 'The A.V. Club',
-            'uploader_id': 'the-av-club',
+            'uploader': 'a.v. club',
+            'upload_date': '20150619',
+            'timestamp': 1434728546,
         },
     }, {
         'url': 'http://www.onionstudios.com/embed?id=2855&autoplay=true',
@@ -44,38 +47,49 @@ class OnionStudiosIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        webpage = self._download_webpage(
+            'http://onionstudios.com/embed/dc94dc2899fe644c0e7241fa04c1b732.js',
+            video_id)
+        mcp_id = compat_str(self._parse_json(self._search_regex(
+            r'window\.mcpMapping\s*=\s*({.+?});', webpage,
+            'MCP Mapping'), video_id, js_to_json)[video_id]['mcp_id'])
         video_data = self._download_json(
-            'http://www.onionstudios.com/video/%s.json' % video_id, video_id)
+            'https://api.vmh.univision.com/metadata/v1/content/' + mcp_id,
+            mcp_id)['videoMetadata']
+        iptc = video_data['photoVideoMetadataIPTC']
+        title = iptc['title']['en']
+        fmg = video_data.get('photoVideoMetadata_fmg') or {}
+        tvss_domain = fmg.get('tvssDomain') or 'https://auth.univision.com'
+        data = self._download_json(
+            tvss_domain + '/api/v3/video-auth/url-signature-tokens',
+            mcp_id, query={'mcpids': mcp_id})['data'][0]
+        formats = []
 
-        title = video_data['title']
+        rendition_url = data.get('renditionUrl')
+        if rendition_url:
+            formats = self._extract_m3u8_formats(
+                rendition_url, mcp_id, 'mp4',
+                'm3u8_native', m3u8_id='hls', fatal=False)
+
+        fallback_rendition_url = data.get('fallbackRenditionUrl')
+        if fallback_rendition_url:
+            formats.append({
+                'format_id': 'fallback',
+                'tbr': int_or_none(self._search_regex(
+                    r'_(\d+)\.mp4', fallback_rendition_url,
+                    'bitrate', default=None)),
+                'url': fallback_rendition_url,
+            })
 
-        formats = []
-        for source in video_data.get('sources', []):
-            source_url = source.get('url')
-            if not source_url:
-                continue
-            ext = mimetype2ext(source.get('content_type')) or determine_ext(source_url)
-            if ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    source_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
-            else:
-                tbr = int_or_none(source.get('bitrate'))
-                formats.append({
-                    'format_id': ext + ('-%d' % tbr if tbr else ''),
-                    'url': source_url,
-                    'width': int_or_none(source.get('width')),
-                    'tbr': tbr,
-                    'ext': ext,
-                })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,
-            'thumbnail': video_data.get('poster_url'),
-            'uploader': video_data.get('channel_name'),
-            'uploader_id': video_data.get('channel_slug'),
-            'duration': float_or_none(video_data.get('duration', 1000)),
-            'tags': video_data.get('tags'),
+            'thumbnail': try_get(iptc, lambda x: x['cloudinaryLink']['link'], compat_str),
+            'uploader': fmg.get('network'),
+            'duration': int_or_none(iptc.get('fileDuration')),
             'formats': formats,
+            'description': try_get(iptc, lambda x: x['description']['en'], compat_str),
+            'timestamp': parse_iso8601(iptc.get('dateReleased')),
         }