Merge 156c58b07f into 00ef748cc0

youtube_dl/extractor/atvat.py

@@ -12,15 +12,15 @@ from ..utils import (
 class ATVAtIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?atv\.at/(?:[^/]+/){2}(?P<id>[dv]\d+)'
     _TESTS = [{
-        'url': 'http://atv.at/aktuell/di-210317-2005-uhr/v1698449/',
-        'md5': 'c3b6b975fb3150fc628572939df205f2',
+        'url': 'https://www.atv.at/bauer-sucht-frau-die-zweite-chance/folge-1/d3390693/',
+        'md5': 'c471605591009dfb6e6c54f7e62e2807',
         'info_dict': {
-            'id': '1698447',
+            'id': '3390684',
             'ext': 'mp4',
-            'title': 'DI, 21.03.17 | 20:05 Uhr 1/1',
+            'title': 'Bauer sucht Frau - Die zweite Chance Folge 1',
         }
     }, {
-        'url': 'http://atv.at/aktuell/meinrad-knapp/d8416/',
+        'url': 'https://www.atv.at/bauer-sucht-frau-staffel-17/fuenfte-eventfolge/d3339537/',
         'only_matching': True,
     }]
 
@@ -28,48 +28,52 @@ class ATVAtIE(InfoExtractor):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
         video_data = self._parse_json(unescapeHTML(self._search_regex(
-            [r'flashPlayerOptions\s*=\s*(["\'])(?P<json>(?:(?!\1).)+)\1',
-             r'class="[^"]*jsb_video/FlashPlayer[^"]*"[^>]+data-jsb="(?P<json>[^"]+)"'],
+            r'var\splaylist\s*=\s*(?P<json>\[.*\]);',
             webpage, 'player data', group='json')),
-            display_id)['config']['initial_video']
+            display_id)
 
-        video_id = video_data['id']
-        video_title = video_data['title']
+        first_video = video_data[0]
+        video_id = first_video['id']
+        video_title = first_video.get('tvShowTitle', first_video['title'])
 
-        parts = []
-        for part in video_data.get('parts', []):
-            part_id = part['id']
-            part_title = part['title']
+        def process_source_entry(self, source, part_id):
+            source_url = source.get('url')
+            if not source_url:
+                return None
+            ext = determine_ext(source_url)
+            if ext == 'm3u8':
+                return self._extract_m3u8_formats(
+                    source_url, part_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False)
+            else:
+                return {
+                    'url': source_url,
+                }
 
-            formats = []
-            for source in part.get('sources', []):
-                source_url = source.get('src')
-                if not source_url:
+        def process_entry(self, entry):
+            part_id = entry['id']
+            formats = [process_source_entry(self, source, part_id) for source in entry["sources"]]
+            formats_flat = []
+            for f in formats:
+                if f is None:
                     continue
-                ext = determine_ext(source_url)
-                if ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        source_url, part_id, 'mp4', 'm3u8_native',
-                        m3u8_id='hls', fatal=False))
+                elif type(f) is list:
+                    formats_flat.extend(f)
                 else:
-                    formats.append({
-                        'format_id': source.get('delivery'),
-                        'url': source_url,
-                    })
-            self._sort_formats(formats)
+                    formats_flat.append(f)
+            self._sort_formats(formats_flat)
+            return {
+                'id': entry['id'],
+                'title': entry['title'],
+                'duration': int_or_none(entry.get('duration')),
+                'formats': formats_flat
+            }
 
-            parts.append({
-                'id': part_id,
-                'title': part_title,
-                'thumbnail': part.get('preview_image_url'),
-                'duration': int_or_none(part.get('duration')),
-                'is_live': part.get('is_livestream'),
-                'formats': formats,
-            })
+        entries = [process_entry(self, entry) for entry in video_data]
 
         return {
             '_type': 'multi_video',
             'id': video_id,
             'title': video_title,
-            'entries': parts,
+            'entries': entries,
         }