[streamango] Improve extraction (closes #12643)

youtube_dl/extractor/streamango.py

@@ -1,11 +1,18 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    js_to_json,
+)
 
 
 class StreamangoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?streamango\.com/(?:f|embed)/(?P<id>.+?)/(?:.+)'
+    _VALID_URL = r'https?://(?:www\.)?streamango\.com/(?:f|embed)/(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://streamango.com/f/clapasobsptpkdfe/20170315_150006_mp4',
         'md5': 'e992787515a182f55e38fc97588d802a',
@@ -13,7 +20,6 @@ class StreamangoIE(InfoExtractor):
             'id': 'clapasobsptpkdfe',
             'ext': 'mp4',
             'title': '20170315_150006.mp4',
-            'url': r're:https://streamango\.com/v/d/clapasobsptpkdfe~[0-9]{10}~(?:[0-9]+\.){3}[0-9]+~.{8}/720',
         }
     }, {
         'url': 'https://streamango.com/embed/clapasobsptpkdfe/20170315_150006_mp4',
@@ -21,29 +27,33 @@ class StreamangoIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        def extract_url(urltype):
-            return self._search_regex(
-                r'type\s*:\s*["\']{}["\']\s*,\s*src\s*:\s*["\'](?P<url>.+?)["\'].*'.format(urltype),
-                webpage, 'video URL', group='url')
-
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
 
         title = self._og_search_title(webpage)
-        url = 'https:' + extract_url('video/mp4')
-        dashurl = extract_url(r'application/dash\+xml')
-
-        formats = [{
-            'url': url,
-            'ext': 'mp4',
-            'width': 1280,
-            'height': 720,
-            'format_id': 'mp4',
-        }]
-
-        formats.extend(self._extract_mpd_formats(
-            dashurl, video_id, mpd_id='dash', fatal=False))
 
+        formats = []
+        for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
+            video = self._parse_json(
+                format_, video_id, transform_source=js_to_json, fatal=False)
+            if not video:
+                continue
+            src = video.get('src')
+            if not src:
+                continue
+            ext = determine_ext(src, default_ext=None)
+            if video.get('type') == 'application/dash+xml' or ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    src, video_id, mpd_id='dash', fatal=False))
+            else:
+                formats.append({
+                    'url': src,
+                    'ext': ext or 'mp4',
+                    'width': int_or_none(video.get('width')),
+                    'height': int_or_none(video.get('height')),
+                    'tbr': int_or_none(video.get('bitrate')),
+                })
         self._sort_formats(formats)
 
         return {