[konserthusetplay] Improve and extract all formats (Closes #8381)

1 files changed, 87 insertions, 24 deletions

diff --git a/youtube_dl/extractor/konserthusetplay.py b/youtube_dl/extractor/konserthusetplay.py
index 8200942d3..f21d050d7 100644
--- a/youtube_dl/extractor/konserthusetplay.py
+++ b/youtube_dl/extractor/konserthusetplay.py
@@ -2,44 +2,107 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
 
 
 class KonserthusetPlayIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?konserthusetplay\.se/\?m=(?P<id>[0-9A-Za-z_-]+)'
-
-    _TESTS = [{
+    _VALID_URL = r'https?://(?:www\.)?konserthusetplay\.se/\?m=(?P<id>[^&]+)'
+    _TEST = {
         'url': 'http://www.konserthusetplay.se/?m=CKDDnlCY-dhWAAqiMERd-A',
-        'md5': 'e272a765e0d12a0226199e5f32d76116',
         'info_dict': {
             'id': 'CKDDnlCY-dhWAAqiMERd-A',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Orkesterns instrument: Valthornen',
             'description': 'md5:f10e1f0030202020396a4d712d2fa827',
-            'thumbnail': 'http://csp.picsearch.com/img/C/K/D/D/title_CKDDnlCY-dhWAAqiMERd-A'
-        }
-    }, {
-        'url': 'http://www.konserthusetplay.se/?m=IyQcMOEpmKqT91SVT5OP8Q',
-        'md5': 'c4adb8ca76fdd33d4cbdcc7c3d181f22',
-        'info_dict': {
-            'id': 'IyQcMOEpmKqT91SVT5OP8Q',
-            'ext': 'mp4',
-            'title': 'Eliasson Einsame Fahrt, violinkonsert',
-            'description': 'md5:a8dcc8dfd9473d52433b2c5f588ba191',
-            'thumbnail': 'http://csp.picsearch.com/img/I/y/Q/c/title_IyQcMOEpmKqT91SVT5OP8Q'
-        }
-    }]
+            'thumbnail': 're:^https?://.*$',
+            'duration': 398.8,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        description = self._og_search_description(webpage)
-        title = self._og_search_title(webpage)
-        main_video = self._html_search_regex(r'<link rel="video_src" href="(.+?)" />', webpage, 'url')
-        thumbnail = self._og_search_thumbnail(webpage)
+
+        e = self._search_regex(
+            r'https?://csp\.picsearch\.com/rest\?.*\be=(.+?)[&"\']', webpage, 'e')
+
+        rest = self._download_json(
+            'http://csp.picsearch.com/rest?e=%s&containerId=mediaplayer&i=object' % e,
+            video_id, transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
+
+        media = rest['media']
+        player_config = media['playerconfig']
+        playlist = player_config['playlist']
+
+        source = next(f for f in playlist if f.get('bitrates'))
+
+        connection_url = (player_config.get('rtmp', {}).get(
+            'netConnectionUrl') or player_config.get(
+            'plugins', {}).get('bwcheck', {}).get('netConnectionUrl'))
+
+        FORMAT_ID_REGEX = r'_([^_]+)_h264m\.mp4'
+
+        formats = []
+
+        fallback_url = source.get('fallbackUrl')
+        fallback_format_id = None
+        if fallback_url:
+            fallback_format_id = self._search_regex(
+                FORMAT_ID_REGEX, fallback_url, 'format id', default=None)
+
+        if connection_url:
+            for f in source['bitrates']:
+                video_url = f.get('url')
+                if not video_url:
+                    continue
+                format_id = self._search_regex(
+                    FORMAT_ID_REGEX, video_url, 'format id', default=None)
+                f_common = {
+                    'vbr': int_or_none(f.get('bitrate')),
+                    'width': int_or_none(f.get('width')),
+                    'height': int_or_none(f.get('height')),
+                }
+                f = f_common.copy()
+                f.update({
+                    'url': connection_url,
+                    'play_path': video_url,
+                    'format_id': 'rtmp-%s' % format_id if format_id else 'rtmp',
+                    'ext': 'flv',
+                })
+                formats.append(f)
+                if format_id and format_id == fallback_format_id:
+                    f = f_common.copy()
+                    f.update({
+                        'url': fallback_url,
+                        'format_id': 'http-%s' % format_id if format_id else 'http',
+                    })
+                    formats.append(f)
+
+        if not formats and fallback_url:
+            formats.append({
+                'url': fallback_url,
+            })
+
+        self._sort_formats(formats)
+
+        title = player_config.get('title') or media['title']
+        description = player_config.get('mediaInfo', {}).get('description')
+        thumbnail = media.get('image')
+        duration = float_or_none(media.get('duration'), 1000)
+
         return {
             'id': video_id,
             'title': title,
             'description': description,
-            'url': main_video,
-            'thumbnail': thumbnail
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
         }