[wwe] Fix issues, extract subtitles and add support for playlists (closes #14781...

diff --git a/youtube_dl/extractor/wwe.py b/youtube_dl/extractor/wwe.py
index c471a79f5eb8ef32f9ccccacf70631f444104af3..bebc77bb544c01717b4d4b2d0281d82b1a390be7 100644 (file)
--- a/youtube_dl/extractor/wwe.py
+++ b/youtube_dl/extractor/wwe.py
@@ -1,20 +1,75 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_str
-from ..utils import urljoin
+from ..utils import (
+    try_get,
+    unescapeHTML,
+    url_or_none,
+    urljoin,
+)
+
+
+class WWEBaseIE(InfoExtractor):
+    _SUBTITLE_LANGS = {
+        'English': 'en',
+        'Deutsch': 'de',
+    }
+
+    def _extract_entry(self, data, url, video_id=None):
+        video_id = compat_str(video_id or data['nid'])
+        title = data['title']
+
+        formats = self._extract_m3u8_formats(
+            data['file'], video_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls')
 
+        description = data.get('description')
+        thumbnail = urljoin(url, data.get('image'))
+        series = data.get('show_name')
+        episode = data.get('episode_name')
 
-class WWEIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?wwe.com/(?:.*/)?videos/(?P<id>[\w-]+)'
+        subtitles = {}
+        tracks = data.get('tracks')
+        if isinstance(tracks, list):
+            for track in tracks:
+                if not isinstance(track, dict):
+                    continue
+                if track.get('kind') != 'captions':
+                    continue
+                track_file = url_or_none(track.get('file'))
+                if not track_file:
+                    continue
+                label = track.get('label')
+                lang = self._SUBTITLE_LANGS.get(label, label) or 'en'
+                subtitles.setdefault(lang, []).append({
+                    'url': track_file,
+                })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'series': series,
+            'episode': episode,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+
+class WWEIE(WWEBaseIE):
+    _VALID_URL = r'https?://(?:[^/]+\.)?wwe\.com/(?:[^/]+/)*videos/(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://www.wwe.com/videos/daniel-bryan-vs-andrade-cien-almas-smackdown-live-sept-4-2018',
-        'md5': '30cbc824b51f4010ea885bfcaec76972',
+        'md5': '92811c6a14bfc206f7a6a9c5d9140184',
         'info_dict': {
             'id': '40048199',
             'ext': 'mp4',
             'title': 'Daniel Bryan vs. Andrade "Cien" Almas: SmackDown LIVE, Sept. 4, 2018',
-            'description': 'Still fuming after he and his wife Brie Bella were attacked by The Miz and Maryse last week, Daniel Bryan takes care of some unfinished business with Andrade "Cien" Almas.',
+            'description': 'md5:2d7424dbc6755c61a0e649d2a8677f67',
             'thumbnail': r're:^https?://.*\.jpg$',
         }
     }, {
@@ -26,31 +81,60 @@ class WWEIE(InfoExtractor):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
 
-        drupal_settings = self._parse_json(
+        landing = self._parse_json(
             self._html_search_regex(
-                r'(?s)Drupal\.settings\s*,\s*({.+?})\);',
+                r'(?s)Drupal\.settings\s*,\s*({.+?})\s*\)\s*;',
                 webpage, 'drupal settings'),
-            display_id)
+            display_id)['WWEVideoLanding']
 
-        player = drupal_settings['WWEVideoLanding']['initialVideo']
-        metadata = player['playlist'][0]
+        data = landing['initialVideo']['playlist'][0]
+        video_id = landing.get('initialVideoId')
 
-        id = compat_str(metadata['nid'])
-        title = metadata.get('title') or self._og_search_title(webpage)
-        video_url = 'https:' + metadata['file']
-        thumbnail = None
-        if metadata.get('image') is not None:
-            thumbnail = urljoin(url, metadata.get('image'))
-        description = metadata.get('description')
+        info = self._extract_entry(data, url, video_id)
+        info['display_id'] = display_id
+        return info
 
-        formats = self._extract_m3u8_formats(video_url, id, 'mp4')
 
-        return {
-            'id': id,
-            'title': title,
-            'formats': formats,
-            'url': video_url,
-            'display_id': display_id,
-            'thumbnail': thumbnail,
-            'description': description,
-        }
+class WWEPlaylistIE(WWEBaseIE):
+    _VALID_URL = r'https?://(?:[^/]+\.)?wwe\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.wwe.com/shows/raw/2018-11-12',
+        'info_dict': {
+            'id': '2018-11-12',
+        },
+        'playlist_mincount': 11,
+    }, {
+        'url': 'http://www.wwe.com/article/walk-the-prank-wwe-edition',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.wwe.com/shows/wwenxt/article/matt-riddle-interview',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if WWEIE.suitable(url) else super(WWEPlaylistIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        entries = []
+        for mobj in re.finditer(
+                r'data-video\s*=\s*(["\'])(?P<data>{.+?})\1', webpage):
+            video = self._parse_json(
+                mobj.group('data'), display_id, transform_source=unescapeHTML,
+                fatal=False)
+            if not video:
+                continue
+            data = try_get(video, lambda x: x['playlist'][0], dict)
+            if not data:
+                continue
+            try:
+                entry = self._extract_entry(data, url)
+            except Exception:
+                continue
+            entry['extractor_key'] = WWEIE.ie_key()
+            entries.append(entry)
+
+        return self.playlist_result(entries, display_id)