[acast] Extract more metadata

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

diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py
index 10cf808857e231cee482434010161a93eee85027..b544da414e1113a88eef68e9b4652c3164033e89 100644 (file)
--- a/youtube_dl/extractor/svt.py
+++ b/youtube_dl/extractor/svt.py
@@ -4,15 +4,23 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
     determine_ext,
     dict_get,
     int_or_none,
     try_get,
+    urljoin,
+    compat_str,
 )
 
 
 class SVTBaseIE(InfoExtractor):
+    _GEO_COUNTRIES = ['SE']
+
     def _extract_video(self, video_info, video_id):
         formats = []
         for vr in video_info['videoReferences']:
@@ -38,7 +46,9 @@ class SVTBaseIE(InfoExtractor):
                     'url': vurl,
                 })
         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
-            self.raise_geo_restricted('This video is only available in Sweden')
+            self.raise_geo_restricted(
+                'This video is only available in Sweden',
+                countries=self._GEO_COUNTRIES)
         self._sort_formats(formats)
 
         subtitles = {}
@@ -118,7 +128,11 @@ class SVTIE(SVTBaseIE):
         return info_dict
 
 
-class SVTPlayIE(SVTBaseIE):
+class SVTPlayBaseIE(SVTBaseIE):
+    _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
+
+
+class SVTPlayIE(SVTPlayBaseIE):
     IE_DESC = 'SVT Play and Öppet arkiv'
     _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)'
     _TESTS = [{
@@ -153,8 +167,8 @@ class SVTPlayIE(SVTBaseIE):
 
         data = self._parse_json(
             self._search_regex(
-                r'root\["__svtplay"\]\s*=\s*([^;]+);',
-                webpage, 'embedded data', default='{}'),
+                self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
+                group='json'),
             video_id, fatal=False)
 
         thumbnail = self._og_search_thumbnail(webpage)
@@ -177,10 +191,92 @@ class SVTPlayIE(SVTBaseIE):
 
         if video_id:
             data = self._download_json(
-                'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id)
+                'https://api.svt.se/videoplayer-api/video/%s' % video_id,
+                video_id, headers=self.geo_verification_headers())
             info_dict = self._extract_video(data, video_id)
             if not info_dict.get('title'):
                 info_dict['title'] = re.sub(
                     r'\s*\|\s*.+?$', '',
                     info_dict.get('episode') or self._og_search_title(webpage))
             return info_dict
+
+
+class SVTSeriesIE(SVTPlayBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)'
+    _TESTS = [{
+        'url': 'https://www.svtplay.se/rederiet',
+        'info_dict': {
+            'id': 'rederiet',
+            'title': 'Rederiet',
+            'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
+        },
+        'playlist_mincount': 318,
+    }, {
+        'url': 'https://www.svtplay.se/rederiet?tab=sasong2',
+        'info_dict': {
+            'id': 'rederiet-sasong2',
+            'title': 'Rederiet - Säsong 2',
+            'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
+        },
+        'playlist_count': 12,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        series_id = self._match_id(url)
+
+        qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+        season_slug = qs.get('tab', [None])[0]
+
+        if season_slug:
+            series_id += '-%s' % season_slug
+
+        webpage = self._download_webpage(
+            url, series_id, 'Downloading series page')
+
+        root = self._parse_json(
+            self._search_regex(
+                self._SVTPLAY_RE, webpage, 'content', group='json'),
+            series_id)
+
+        season_name = None
+
+        entries = []
+        for season in root['relatedVideoContent']['relatedVideosAccordion']:
+            if not isinstance(season, dict):
+                continue
+            if season_slug:
+                if season.get('slug') != season_slug:
+                    continue
+                season_name = season.get('name')
+            videos = season.get('videos')
+            if not isinstance(videos, list):
+                continue
+            for video in videos:
+                content_url = video.get('contentUrl')
+                if not content_url or not isinstance(content_url, compat_str):
+                    continue
+                entries.append(
+                    self.url_result(
+                        urljoin(url, content_url),
+                        ie=SVTPlayIE.ie_key(),
+                        video_title=video.get('title')
+                    ))
+
+        metadata = root.get('metaData')
+        if not isinstance(metadata, dict):
+            metadata = {}
+
+        title = metadata.get('title')
+        season_name = season_name or season_slug
+
+        if title and season_name:
+            title = '%s - %s' % (title, season_name)
+        elif season_slug:
+            title = season_slug
+
+        return self.playlist_result(
+            entries, series_id, title, metadata.get('description'))