[cda] Improve extraction (closes #28709, closes #28937)

youtube_dl/extractor/cda.py

@@ -133,6 +133,8 @@ class CDAIE(InfoExtractor):
             'age_limit': 18 if need_confirm_age else 0,
         }
 
+        info = self._search_json_ld(webpage, video_id, default={})
+
         # Source: https://www.cda.pl/js/player.js?t=1606154898
         def decrypt_file(a):
             for p in ('_XDDD', '_CDA', '_ADC', '_CXD', '_QWE', '_Q5', '_IKSDE'):
@@ -197,7 +199,7 @@ class CDAIE(InfoExtractor):
                 handler = self._download_webpage
 
             webpage = handler(
-                self._BASE_URL + href, video_id,
+                urljoin(self._BASE_URL, href), video_id,
                 'Downloading %s version information' % resolution, fatal=False)
             if not webpage:
                 # Manually report warning because empty page is returned when
@@ -209,6 +211,4 @@ class CDAIE(InfoExtractor):
 
         self._sort_formats(formats)
 
-        info = self._search_json_ld(webpage, video_id, default={})
-
         return merge_dicts(info_dict, info)

youtube_dl/extractor/svt.py

@@ -146,7 +146,7 @@ class SVTPlayIE(SVTPlayBaseIE):
                         )
                         (?P<svt_id>[^/?#&]+)|
                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
-                        (?:.*?modalId=(?P<modal_id>[\da-zA-Z-]+))?
+                        (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))?
                     )
                     '''
     _TESTS = [{
@@ -177,6 +177,9 @@ class SVTPlayIE(SVTPlayBaseIE):
     }, {
         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
         'only_matching': True,
+    }, {
+        'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
+        'only_matching': True,
     }, {
         # geo restricted to Sweden
         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
@@ -259,7 +262,7 @@ class SVTPlayIE(SVTPlayBaseIE):
         if not svt_id:
             svt_id = self._search_regex(
                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
-                 r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\bmodalId=([\da-zA-Z-]+)' % re.escape(video_id),
+                 r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id),
                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',

youtube_dl/extractor/tv2dk.py

@@ -74,6 +74,12 @@ class TV2DKIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         entries = []
+
+        def add_entry(partner_id, kaltura_id):
+            entries.append(self.url_result(
+                'kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura',
+                video_id=kaltura_id))
+
         for video_el in re.findall(r'(?s)<[^>]+\bdata-entryid\s*=[^>]*>', webpage):
             video = extract_attributes(video_el)
             kaltura_id = video.get('data-entryid')
@@ -82,9 +88,14 @@ class TV2DKIE(InfoExtractor):
             partner_id = video.get('data-partnerid')
             if not partner_id:
                 continue
-            entries.append(self.url_result(
-                'kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura',
-                video_id=kaltura_id))
+            add_entry(partner_id, kaltura_id)
+        if not entries:
+            kaltura_id = self._search_regex(
+                r'entry_id\s*:\s*["\']([0-9a-z_]+)', webpage, 'kaltura id')
+            partner_id = self._search_regex(
+                (r'\\u002Fp\\u002F(\d+)\\u002F', r'/p/(\d+)/'), webpage,
+                'partner id')
+            add_entry(partner_id, kaltura_id)
         return self.playlist_result(entries)
 
 

youtube_dl/extractor/twitter.py

@@ -19,6 +19,7 @@ from ..utils import (
     strip_or_none,
     unified_timestamp,
     update_url_query,
+    url_or_none,
     xpath_text,
 )
 
@@ -52,6 +53,9 @@ class TwitterBaseIE(InfoExtractor):
             return [f]
 
     def _extract_formats_from_vmap_url(self, vmap_url, video_id):
+        vmap_url = url_or_none(vmap_url)
+        if not vmap_url:
+            return []
         vmap_data = self._download_xml(vmap_url, video_id)
         formats = []
         urls = []

youtube_dl/extractor/xtube.py

@@ -11,6 +11,7 @@ from ..utils import (
     parse_duration,
     sanitized_Request,
     str_to_int,
+    url_or_none,
 )
 
 
@@ -87,10 +88,10 @@ class XTubeIE(InfoExtractor):
                 'Cookie': 'age_verified=1; cookiesAccepted=1',
             })
 
-        title, thumbnail, duration = [None] * 3
+        title, thumbnail, duration, sources, media_definition = [None] * 5
 
         config = self._parse_json(self._search_regex(
-            r'playerConf\s*=\s*({.+?})\s*,\s*(?:\n|loaderConf)', webpage, 'config',
+            r'playerConf\s*=\s*({.+?})\s*,\s*(?:\n|loaderConf|playerWrapper)', webpage, 'config',
             default='{}'), video_id, transform_source=js_to_json, fatal=False)
         if config:
             config = config.get('mainRoll')
@@ -99,20 +100,52 @@ class XTubeIE(InfoExtractor):
                 thumbnail = config.get('poster')
                 duration = int_or_none(config.get('duration'))
                 sources = config.get('sources') or config.get('format')
+                media_definition = config.get('mediaDefinition')
 
-        if not isinstance(sources, dict):
+        if not isinstance(sources, dict) and not media_definition:
             sources = self._parse_json(self._search_regex(
                 r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),',
                 webpage, 'sources', group='sources'), video_id,
                 transform_source=js_to_json)
 
         formats = []
+        format_urls = set()
+
+        if isinstance(sources, dict):
             for format_id, format_url in sources.items():
+                format_url = url_or_none(format_url)
+                if not format_url:
+                    continue
+                if format_url in format_urls:
+                    continue
+                format_urls.add(format_url)
                 formats.append({
                     'url': format_url,
                     'format_id': format_id,
                     'height': int_or_none(format_id),
                 })
+
+        if isinstance(media_definition, list):
+            for media in media_definition:
+                video_url = url_or_none(media.get('videoUrl'))
+                if not video_url:
+                    continue
+                if video_url in format_urls:
+                    continue
+                format_urls.add(video_url)
+                format_id = media.get('format')
+                if format_id == 'hls':
+                    formats.extend(self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id='hls', fatal=False))
+                elif format_id == 'mp4':
+                    height = int_or_none(media.get('quality'))
+                    formats.append({
+                        'url': video_url,
+                        'format_id': '%s-%d' % (format_id, height) if height else format_id,
+                        'height': height,
+                    })
+
         self._remove_duplicate_formats(formats)
         self._sort_formats(formats)