[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / tv4.py

diff --git a/youtube_dl/extractor/tv4.py b/youtube_dl/extractor/tv4.py
index 7918e3d86f9f88ff91274a08d8fd4d06885c12b8..c498b0191623220071d38764f04d3ba1fc114558 100644 (file)
--- a/youtube_dl/extractor/tv4.py
+++ b/youtube_dl/extractor/tv4.py
@@ -1,13 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
-from ..compat import compat_str
 from ..utils import (
     int_or_none,
     parse_iso8601,
-    try_get,
-    determine_ext,
 )
 
 
@@ -18,12 +17,13 @@ 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',
@@ -62,67 +62,63 @@ 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')
-
-        # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
-        if info.get('is_geo_restricted'):
-            self.report_warning('This content might not be available in your country due to licensing restrictions.')
+            'https://playback-api.b17g.net/asset/%s' % video_id,
+            video_id, 'Downloading video info JSON', query={
+                'service': 'tv4',
+                'device': 'browser',
+                'protocol': 'hls,dash',
+                'drm': 'widevine',
+            })['metadata']
 
         title = info['title']
 
-        subtitles = {}
-        formats = []
-        # http formats are linked with unresolvable host
-        for kind in ('hls3', ''):
-            data = self._download_json(
-                'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
-                video_id, 'Downloading sources JSON', query={
-                    'protocol': kind,
-                    'videoFormat': 'MP4+WEBVTT',
-                })
-            items = try_get(data, lambda x: x['playback']['items']['item'])
-            if not items:
-                continue
-            if isinstance(items, dict):
-                items = [items]
-            for item in items:
-                manifest_url = item.get('url')
-                if not isinstance(manifest_url, compat_str):
-                    continue
-                ext = determine_ext(manifest_url)
-                if ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
-                        m3u8_id=kind, fatal=False))
-                elif ext == 'f4m':
-                    formats.extend(self._extract_akamai_formats(
-                        manifest_url, video_id, {
-                            'hls': 'tv4play-i.akamaihd.net',
-                        }))
-                elif ext == 'webvtt':
-                    subtitles = self._merge_subtitles(
-                        subtitles, {
-                            'sv': [{
-                                'url': manifest_url,
-                                'ext': 'vtt',
-                            }]})
+        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))
+
+        if not formats and info.get('is_geo_restricted'):
+            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
+
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
-            'subtitles': subtitles,
+            # 'subtitles': subtitles,
             'description': info.get('description'),
             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
             'duration': int_or_none(info.get('duration')),
             'thumbnail': info.get('image'),
-            'is_live': info.get('is_live') is True,
+            'is_live': info.get('isLive') is True,
+            'series': info.get('seriesTitle'),
+            'season_number': int_or_none(info.get('seasonNumber')),
+            'episode': info.get('episodeTitle'),
+            'episode_number': int_or_none(info.get('episodeNumber')),
         }