[tvplay] fix Viafree extraction(closes #15189)(closes #24473)(closes #24789)

youtube_dl/extractor/tvplay.py

@@ -6,7 +6,6 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_str,
     compat_urlparse,
 )
 from ..utils import (
@@ -15,9 +14,7 @@ from ..utils import (
     int_or_none,
     parse_iso8601,
     qualities,
-    smuggle_url,
     try_get,
-    unsmuggle_url,
     update_url_query,
     url_or_none,
 )
@@ -235,11 +232,6 @@ class TVPlayIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        url, smuggled_data = unsmuggle_url(url, {})
-        self._initialize_geo_bypass({
-            'countries': smuggled_data.get('geo_countries'),
-        })
-
         video_id = self._match_id(url)
         geo_country = self._search_regex(
             r'https?://[^/]+\.([a-z]{2})', url,
@@ -285,8 +277,6 @@ class TVPlayIE(InfoExtractor):
                     'ext': ext,
                 }
                 if video_url.startswith('rtmp'):
-                    if smuggled_data.get('skip_rtmp'):
-                        continue
                     m = re.search(
                         r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
                     if not m:
@@ -347,115 +337,80 @@ class ViafreeIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:www\.)?
-                        viafree\.
+                        viafree\.(?P<country>dk|no|se)
-                        (?:
+                        /(?P<id>program(?:mer)?/(?:[^/]+/)+[^/?#&]+)
-                            (?:dk|no)/programmer|
-                            se/program
-                        )
-                        /(?:[^/]+/)+(?P<id>[^/?#&]+)
                     '''
     _TESTS = [{
-        'url': 'http://www.viafree.se/program/livsstil/husraddarna/sasong-2/avsnitt-2',
+        'url': 'http://www.viafree.no/programmer/underholdning/det-beste-vorspielet/sesong-2/episode-1',
         'info_dict': {
-            'id': '395375',
+            'id': '757786',
             'ext': 'mp4',
-            'title': 'Husräddarna S02E02',
+            'title': 'Det beste vorspielet - Sesong 2 - Episode 1',
-            'description': 'md5:4db5c933e37db629b5a2f75dfb34829e',
+            'description': 'md5:b632cb848331404ccacd8cd03e83b4c3',
-            'series': 'Husräddarna',
+            'series': 'Det beste vorspielet',
-            'season': 'Säsong 2',
             'season_number': 2,
-            'duration': 2576,
+            'duration': 1116,
-            'timestamp': 1400596321,
+            'timestamp': 1471200600,
-            'upload_date': '20140520',
+            'upload_date': '20160814',
         },
         'params': {
             'skip_download': True,
         },
-        'add_ie': [TVPlayIE.ie_key()],
     }, {
         # with relatedClips
         'url': 'http://www.viafree.se/program/reality/sommaren-med-youtube-stjarnorna/sasong-1/avsnitt-1',
-        'info_dict': {
+        'only_matching': True,
-            'id': '758770',
-            'ext': 'mp4',
-            'title': 'Sommaren med YouTube-stjärnorna S01E01',
-            'description': 'md5:2bc69dce2c4bb48391e858539bbb0e3f',
-            'series': 'Sommaren med YouTube-stjärnorna',
-            'season': 'Säsong 1',
-            'season_number': 1,
-            'duration': 1326,
-            'timestamp': 1470905572,
-            'upload_date': '20160811',
-        },
-        'params': {
-            'skip_download': True,
-        },
-        'add_ie': [TVPlayIE.ie_key()],
     }, {
         # Different og:image URL schema
         'url': 'http://www.viafree.se/program/reality/sommaren-med-youtube-stjarnorna/sasong-1/avsnitt-2',
         'only_matching': True,
     }, {
-        'url': 'http://www.viafree.no/programmer/underholdning/det-beste-vorspielet/sesong-2/episode-1',
+        'url': 'http://www.viafree.se/program/livsstil/husraddarna/sasong-2/avsnitt-2',
         'only_matching': True,
     }, {
         'url': 'http://www.viafree.dk/programmer/reality/paradise-hotel/saeson-7/episode-5',
         'only_matching': True,
     }]
+    _GEO_BYPASS = False
 
     @classmethod
     def suitable(cls, url):
         return False if TVPlayIE.suitable(url) else super(ViafreeIE, cls).suitable(url)
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        country, path = re.match(self._VALID_URL, url).groups()
+        content = self._download_json(
+            'https://viafree-content.mtg-api.com/viafree-content/v1/%s/path/%s' % (country, path), path)
+        program = content['_embedded']['viafreeBlocks'][0]['_embedded']['program']
+        guid = program['guid']
+        meta = content['meta']
+        title = meta['title']
 
-        webpage = self._download_webpage(url, video_id)
+        try:
+            stream_href = self._download_json(
+                program['_links']['streamLink']['href'], guid,
+                headers=self.geo_verification_headers())['embedded']['prioritizedStreams'][0]['links']['stream']['href']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                self.raise_geo_restricted(countries=[country])
+            raise
 
-        data = self._parse_json(
+        formats = self._extract_m3u8_formats(stream_href, guid, 'mp4')
-            self._search_regex(
+        self._sort_formats(formats)
-                r'(?s)window\.App\s*=\s*({.+?})\s*;\s*</script',
+        episode = program.get('episode') or {}
-                webpage, 'data', default='{}'),
-            video_id, transform_source=lambda x: re.sub(
-                r'(?s)function\s+[a-zA-Z_][\da-zA-Z_]*\s*\([^)]*\)\s*{[^}]*}\s*',
-                'null', x), fatal=False)
 
-        video_id = None
+        return {
-
+            'id': guid,
-        if data:
+            'title': title,
-            video_id = try_get(
+            'thumbnail': meta.get('image'),
-                data, lambda x: x['context']['dispatcher']['stores'][
+            'description': meta.get('description'),
-                    'ContentPageProgramStore']['currentVideo']['id'],
+            'series': episode.get('seriesTitle'),
-                compat_str)
+            'episode_number': int_or_none(episode.get('episodeNumber')),
-
+            'season_number': int_or_none(episode.get('seasonNumber')),
-        # Fallback #1 (extract from og:image URL schema)
+            'duration': int_or_none(try_get(program, lambda x: x['video']['duration']['milliseconds']), 1000),
-        if not video_id:
+            'timestamp': parse_iso8601(try_get(program, lambda x: x['availability']['start'])),
-            thumbnail = self._og_search_thumbnail(webpage, default=None)
+            'formats': formats,
-            if thumbnail:
+        }
-                video_id = self._search_regex(
-                    # Patterns seen:
-                    #  http://cdn.playapi.mtgx.tv/imagecache/600x315/cloud/content-images/inbox/765166/a2e95e5f1d735bab9f309fa345cc3f25.jpg
-                    #  http://cdn.playapi.mtgx.tv/imagecache/600x315/cloud/content-images/seasons/15204/758770/4a5ba509ca8bc043e1ebd1a76131cdf2.jpg
-                    r'https?://[^/]+/imagecache/(?:[^/]+/)+(\d{6,})/',
-                    thumbnail, 'video id', default=None)
-
-        # Fallback #2. Extract from raw JSON string.
-        # May extract wrong video id if relatedClips is present.
-        if not video_id:
-            video_id = self._search_regex(
-                r'currentVideo["\']\s*:\s*.+?["\']id["\']\s*:\s*["\'](\d{6,})',
-                webpage, 'video id')
-
-        return self.url_result(
-            smuggle_url(
-                'mtg:%s' % video_id,
-                {
-                    'geo_countries': [
-                        compat_urlparse.urlparse(url).netloc.rsplit('.', 1)[-1]],
-                    # rtmp host mtgfs.fplive.net for viafree is unresolvable
-                    'skip_rtmp': True,
-                }),
-            ie=TVPlayIE.ie_key(), video_id=video_id)
 
 
 class TVPlayHomeIE(InfoExtractor):