Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / tv4.py

diff --git a/youtube_dl/extractor/tv4.py b/youtube_dl/extractor/tv4.py
index 343edf20663d172a4e071a77f228fc4d9962003d..51923e44afcc7913c32406c940bba6aec9edbf13 100644 (file)
--- a/youtube_dl/extractor/tv4.py
+++ b/youtube_dl/extractor/tv4.py
@@ -1,9 +1,11 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
+    int_or_none,
     parse_iso8601,
 )
 
@@ -15,33 +17,34 @@ class TV4IE(InfoExtractor):
             tv4\.se/(?:[^/]+)/klipp/(?:.*)-|
             tv4play\.se/
             (?:
-                (?:program|barn)/(?:[^\?]+)\?video_id=|
+                (?:program|barn)/(?:[^/]+/|(?:[^\?]+)\?video_id=)|
                 iframe/video/|
                 film/|
                 sport/|
             )
         )(?P<id>[0-9]+)'''
+    _GEO_COUNTRIES = ['SE']
     _TESTS = [
         {
             'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650',
-            'md5': '909d6454b87b10a25aa04c4bdd416a9b',
+            'md5': 'cb837212f342d77cec06e6dad190e96d',
             'info_dict': {
                 'id': '2491650',
                 'ext': 'mp4',
                 'title': 'Kalla Fakta 5 (english subtitles)',
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'thumbnail': r're:^https?://.*\.jpg$',
                 'timestamp': int,
                 'upload_date': '20131125',
             },
         },
         {
             'url': 'http://www.tv4play.se/iframe/video/3054113',
-            'md5': '77f851c55139ffe0ebd41b6a5552489b',
+            'md5': 'cb837212f342d77cec06e6dad190e96d',
             'info_dict': {
                 'id': '3054113',
                 'ext': 'mp4',
                 'title': 'Så här jobbar ficktjuvarna - se avslöjande bilder',
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'thumbnail': r're:^https?://.*\.jpg$',
                 'description': 'Unika bilder avslöjar hur turisternas fickor vittjas mitt på Stockholms central. Två experter på ficktjuvarna avslöjar knepen du ska se upp för.',
                 'timestamp': int,
                 'upload_date': '20150130',
@@ -59,42 +62,54 @@ class TV4IE(InfoExtractor):
             'url': 'http://www.tv4play.se/barn/looney-tunes?video_id=3062412',
             'only_matching': True,
         },
+        {
+            'url': 'http://www.tv4play.se/program/farang/3922081',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         info = self._download_json(
-            'http://www.tv4play.se/player/assets/%s.json' % video_id, video_id, 'Downloading video info JSON')
+            'http://www.tv4play.se/player/assets/%s.json' % video_id,
+            video_id, 'Downloading video info JSON')
+
+        title = info['title']
 
-        # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
-        if info['is_geo_restricted']:
-            self.report_warning('This content might not be available in your country due to licensing restrictions.')
-        if info['requires_subscription']:
-            raise ExtractorError('This content requires subscription.', expected=True)
+        manifest_url = self._download_json(
+            'https://playback-api.b17g.net/media/' + video_id,
+            video_id, query={
+                'service': 'tv4',
+                'device': 'browser',
+                'protocol': 'hls',
+            })['playbackItem']['manifestUrl']
+        formats = self._extract_m3u8_formats(
+            manifest_url, video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls', fatal=False)
+        formats.extend(self._extract_mpd_formats(
+            manifest_url.replace('.m3u8', '.mpd'),
+            video_id, mpd_id='dash', fatal=False))
+        formats.extend(self._extract_f4m_formats(
+            manifest_url.replace('.m3u8', '.f4m'),
+            video_id, f4m_id='hds', fatal=False))
+        formats.extend(self._extract_ism_formats(
+            re.sub(r'\.ism/.+?\.m3u8', r'.ism/Manifest', manifest_url),
+            video_id, ism_id='mss', fatal=False))
 
-        sources_data = self._download_json(
-            'https://prima.tv4play.se/api/web/asset/%s/play.json?protocol=http&videoFormat=MP4' % video_id, video_id, 'Downloading sources JSON')
-        sources = sources_data['playback']
+        if not formats and info.get('is_geo_restricted'):
+            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 
-        formats = []
-        for item in sources.get('items', {}).get('item', []):
-            ext, bitrate = item['mediaFormat'], item['bitrate']
-            formats.append({
-                'format_id': '%s_%s' % (ext, bitrate),
-                'tbr': bitrate,
-                'ext': ext,
-                'url': item['url'],
-            })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': info['title'],
+            'title': title,
             'formats': formats,
+            # 'subtitles': subtitles,
             'description': info.get('description'),
             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
-            'duration': info.get('duration'),
+            'duration': int_or_none(info.get('duration')),
             'thumbnail': info.get('image'),
-            'is_live': sources.get('live'),
+            'is_live': info.get('is_live') is True,
         }