[rtbf] improve extraction

- add support for audio and live streams(closes #11923)(closes #9638)
- extract HLS, DASH and all HTTP formats
- extract subtitles
- fixup specific http urls(fixes #16101)

youtube_dl/extractor/rtbf.py

@@ -1,10 +1,14 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
-    int_or_none,
     ExtractorError,
+    float_or_none,
+    int_or_none,
+    strip_or_none,
 )
 
 
@@ -14,20 +18,19 @@ class RTBFIE(InfoExtractor):
         (?:
             video/[^?]+\?.*\bid=|
             ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=|
-            auvio/[^/]+\?.*id=
+            auvio/[^/]+\?.*\b(?P<live>l)?id=
         )(?P<id>\d+)'''
     _TESTS = [{
         'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
-        'md5': '799f334ddf2c0a582ba80c44655be570',
+        'md5': '8c876a1cceeb6cf31b476461ade72384',
         'info_dict': {
             'id': '1921274',
             'ext': 'mp4',
             'title': 'Les Diables au coeur (épisode 2)',
-            'description': 'Football - Diables Rouges',
+            'description': '(du 25/04/2014)',
-            'duration': 3099,
+            'duration': 3099.54,
             'upload_date': '20140425',
-            'timestamp': 1398456336,
+            'timestamp': 1398456300,
-            'uploader': 'rtbfsport',
         }
     }, {
         # geo restricted
@@ -39,6 +42,18 @@ class RTBFIE(InfoExtractor):
     }, {
         'url': 'http://www.rtbf.be/auvio/detail_jeudi-en-prime-siegfried-bracke?id=2102996',
         'only_matching': True,
+    }, {
+        # Live
+        'url': 'https://www.rtbf.be/auvio/direct_pure-fm?lid=134775',
+        'only_matching': True,
+    }, {
+        # Audio
+        'url': 'https://www.rtbf.be/auvio/detail_cinq-heures-cinema?id=2360811',
+        'only_matching': True,
+    }, {
+        # With Subtitle
+        'url': 'https://www.rtbf.be/auvio/detail_les-carnets-du-bourlingueur?id=2361588',
+        'only_matching': True,
     }]
     _IMAGE_HOST = 'http://ds1.ds.static.rtbf.be'
     _PROVIDERS = {
@@ -53,46 +68,94 @@ class RTBFIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        live, media_id = re.match(self._VALID_URL, url).groups()
-        data = self._download_json(
+        embed_page = self._download_webpage(
-            'http://www.rtbf.be/api/media/video?method=getVideoDetail&args[]=%s' % video_id, video_id)
+            'https://www.rtbf.be/auvio/embed/' + ('direct' if live else 'media'),
+            media_id, query={'id': media_id})
+        data = self._parse_json(self._html_search_regex(
+            r'data-media="([^"]+)"', embed_page, 'media data'), media_id)
 
         error = data.get('error')
         if error:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
 
-        data = data['data']
-
         provider = data.get('provider')
         if provider in self._PROVIDERS:
             return self.url_result(data['url'], self._PROVIDERS[provider])
 
+        title = data['title']
+        is_live = data.get('isLive')
+        if is_live:
+            title = self._live_title(title)
+        height_re = r'-(\d+)p\.'
         formats = []
-        for key, format_id in self._QUALITIES:
+
-            format_url = data.get(key + 'Url')
+        m3u8_url = data.get('urlHlsAes128') or data.get('urlHls')
-            if format_url:
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False))
+
+        fix_url = lambda x: x.replace('//rtbf-vod.', '//rtbf.') if '/geo/drm/' in x else x
+        http_url = data.get('url')
+        if formats and http_url and re.search(height_re, http_url):
+            http_url = fix_url(http_url)
+            for m3u8_f in formats.copy():
+                height = m3u8_f.get('height')
+                if not height:
+                    continue
+                f = m3u8_f.copy()
+                del f['protocol']
+                f.update({
+                    'format_id': m3u8_f['format_id'].replace('hls-', 'http-'),
+                    'url': re.sub(height_re, '-%dp.' % height, http_url),
+                })
+                formats.append(f)
+        else:
+            sources = data.get('sources') or {}
+            for key, format_id in self._QUALITIES:
+                format_url = sources.get(key)
+                if not format_url:
+                    continue
+                height = int_or_none(self._search_regex(
+                    height_re, format_url, 'height', default=None))
                 formats.append({
                     'format_id': format_id,
-                    'url': format_url,
+                    'url': fix_url(format_url),
+                    'height': height,
                 })
 
-        thumbnails = []
+        mpd_url = data.get('urlDash')
-        for thumbnail_id, thumbnail_url in data.get('thumbnail', {}).items():
+        if not data.get('drm') and mpd_url:
-            if thumbnail_id != 'default':
+            formats.extend(self._extract_mpd_formats(
-                thumbnails.append({
+                mpd_url, media_id, mpd_id='dash', fatal=False))
-                    'url': self._IMAGE_HOST + thumbnail_url,
+
-                    'id': thumbnail_id,
+        audio_url = data.get('urlAudio')
-                })
+        if audio_url:
+            formats.append({
+                'format_id': 'audio',
+                'url': audio_url,
+                'vcodec': 'none',
+            })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for track in (data.get('tracks') or {}).values():
+            sub_url = track.get('url')
+            if not sub_url:
+                continue
+            subtitles.setdefault(track.get('lang') or 'fr', []).append({
+                'url': sub_url,
+            })
 
         return {
-            'id': video_id,
+            'id': media_id,
             'formats': formats,
-            'title': data['title'],
+            'title': title,
-            'description': data.get('description') or data.get('subtitle'),
+            'description': strip_or_none(data.get('description')),
-            'thumbnails': thumbnails,
+            'thumbnail': data.get('thumbnail'),
-            'duration': data.get('duration') or data.get('realDuration'),
+            'duration': float_or_none(data.get('realDuration')),
-            'timestamp': int_or_none(data.get('created')),
+            'timestamp': int_or_none(data.get('liveFrom')),
-            'view_count': int_or_none(data.get('viewCount')),
+            'series': data.get('programLabel'),
-            'uploader': data.get('channel'),
+            'subtitles': subtitles,
-            'tags': data.get('tags'),
+            'is_live': is_live,
         }