[streamcz] Update extractor

youtube_dl/extractor/streamcz.py

@@ -5,14 +5,16 @@ import re
 import json
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+)
 from ..utils import (
     int_or_none,
-    compat_str,
 )
 
 
 class StreamCZIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<videoid>.+)'
+    _VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<id>[0-9]+)'
 
     _TESTS = [{
         'url': 'http://www.stream.cz/peklonataliri/765767-ecka-pro-deti',
@@ -21,61 +23,63 @@ class StreamCZIE(InfoExtractor):
             'id': '765767',
             'ext': 'mp4',
             'title': 'Peklo na talíři: Éčka pro děti',
-            'description': 'md5:49ace0df986e95e331d0fe239d421519',
+            'description': 'Taška s grónskou pomazánkou a další pekelnosti ZDE',
-            'thumbnail': 'http://im.stream.cz/episode/52961d7e19d423f8f06f0100',
+            'thumbnail': 're:^http://im.stream.cz/episode/52961d7e19d423f8f06f0100',
             'duration': 256,
         },
     }, {
         'url': 'http://www.stream.cz/blanik/10002447-tri-roky-pro-mazanka',
-        'md5': '246272e753e26bbace7fcd9deca0650c',
+        'md5': 'e54a254fb8b871968fd8403255f28589',
         'info_dict': {
             'id': '10002447',
             'ext': 'mp4',
             'title': 'Kancelář Blaník: Tři roky pro Mazánka',
-            'description': 'md5:9177695a8b756a0a8ab160de4043b392',
+            'description': 'md5:3862a00ba7bf0b3e44806b544032c859',
-            'thumbnail': 'http://im.stream.cz/episode/537f838c50c11f8d21320000',
+            'thumbnail': 're:^http://im.stream.cz/episode/537f838c50c11f8d21320000',
             'duration': 368,
         },
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('videoid')
+        data = self._download_json(
-
+            'http://www.stream.cz/API/episode/%s' % video_id, video_id)
-        webpage = self._download_webpage(url, video_id)
-
-        data = self._html_search_regex(r'Stream\.Data\.Episode\((.+?)\);', webpage, 'stream data')
-
-        jsonData = json.loads(data)
 
         formats = []
-        for video in jsonData['instances']:
+        for quality, video in enumerate(data['video_qualities']):
-            for video_format in video['instances']:
+            for f in video['formats']:
-                format_id = video_format['quality']
+                typ = f['type'].partition('/')[2]
-
+                qlabel = video.get('quality_label')
-                if format_id == '240p':
-                    quality = 0
-                elif format_id == '360p':
-                    quality = 1
-                elif format_id == '480p':
-                    quality = 2
-                elif format_id == '720p':
-                    quality = 3
-
                 formats.append({
-                    'format_id': '%s-%s' % (video_format['type'].split('/')[1], format_id),
+                    'format_note': '%s-%s' % (qlabel, typ) if qlabel else typ,
-                    'url': video_format['source'],
+                    'format_id': '%s-%s' % (typ, f['quality']),
+                    'url': f['source'],
+                    'height': int_or_none(f['quality'].rstrip('p')),
                     'quality': quality,
                 })
-
         self._sort_formats(formats)
 
+        image = data.get('image')
+        if image:
+            thumbnail = self._proto_relative_url(
+                image.replace('{width}', '1240').replace('{height}', '697'),
+                scheme='http:',
+            )
+        else:
+            thumbnail = None
+
+        stream = data.get('_embedded', {}).get('stream:show', {}).get('name')
+        if stream:
+            title = '%s: %s' % (stream, data['name'])
+        else:
+            title = data['name']
+
         return {
-            'id': compat_str(jsonData['episode_id']),
+            'id': video_id,
-            'title': self._og_search_title(webpage),
+            'title': title,
-            'thumbnail': jsonData['episode_image_original_url'].replace('//', 'http://'),
+            'thumbnail': thumbnail,
             'formats': formats,
-            'description': self._og_search_description(webpage),
+            'description': data.get('web_site_text'),
-            'duration': int_or_none(jsonData['duration']),
+            'duration': int_or_none(data.get('duration')),
-            'view_count': int_or_none(jsonData['stats_total']),
+            'view_count': int_or_none(data.get('views')),
         }