[ie/orf:on] Improve extraction (#9677)

[yt-dlp.git] / yt_dlp / extractor / rts.py

diff --git a/yt_dlp/extractor/rts.py b/yt_dlp/extractor/rts.py
index 48f17b828c1a8c6b0494ad1edb2bc3f914b3417a..bce5cba82a289e045730b26fe1aa0b9d0eb66644 100644 (file)
--- a/yt_dlp/extractor/rts.py
+++ b/yt_dlp/extractor/rts.py
@@ -1,27 +1,26 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import re
 
 from .srgssr import SRGSSRIE
 from ..compat import compat_str
 from ..utils import (
+    determine_ext,
     int_or_none,
     parse_duration,
     parse_iso8601,
     unescapeHTML,
-    determine_ext,
+    urljoin,
 )
 
 
-class RTSIE(SRGSSRIE):
+class RTSIE(SRGSSRIE):  # XXX: Do not subclass from concrete IE
+    _WORKING = False
     IE_DESC = 'RTS.ch'
     _VALID_URL = r'rts:(?P<rts_id>\d+)|https?://(?:.+?\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html'
 
     _TESTS = [
         {
             'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
-            'md5': 'ff7f8450a90cf58dacb64e29707b4a8e',
+            'md5': '753b877968ad8afaeddccc374d4256a5',
             'info_dict': {
                 'id': '3449373',
                 'display_id': 'les-enfants-terribles',
@@ -35,6 +34,7 @@ class RTSIE(SRGSSRIE):
                 'thumbnail': r're:^https?://.*\.image',
                 'view_count': int,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
         },
         {
             'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
@@ -63,11 +63,12 @@ class RTSIE(SRGSSRIE):
                 # m3u8 download
                 'skip_download': True,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
             'skip': 'Blocked outside Switzerland',
         },
         {
             'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
-            'md5': '1bae984fe7b1f78e94abc74e802ed99f',
+            'md5': '9bb06503773c07ce83d3cbd793cebb91',
             'info_dict': {
                 'id': '5745356',
                 'display_id': 'londres-cachee-par-un-epais-smog',
@@ -81,6 +82,7 @@ class RTSIE(SRGSSRIE):
                 'thumbnail': r're:^https?://.*\.image',
                 'view_count': int,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
         },
         {
             'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html',
@@ -112,7 +114,7 @@ class RTSIE(SRGSSRIE):
     ]
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
+        m = self._match_valid_url(url)
         media_id = m.group('rts_id') or m.group('id')
         display_id = m.group('display_id') or media_id
 
@@ -135,8 +137,8 @@ def download_json(internal_id):
 
             if not entries:
                 page, urlh = self._download_webpage_handle(url, display_id)
-                if re.match(self._VALID_URL, urlh.geturl()).group('id') != media_id:
-                    return self.url_result(urlh.geturl(), 'RTS')
+                if re.match(self._VALID_URL, urlh.url).group('id') != media_id:
+                    return self.url_result(urlh.url, 'RTS')
 
                 # article with videos on rhs
                 videos = re.findall(
@@ -160,7 +162,7 @@ def download_json(internal_id):
         media_type = 'video' if 'video' in all_info else 'audio'
 
         # check for errors
-        self.get_media_data('rts', media_type, media_id)
+        self._get_media_data('rts', media_type, media_id)
 
         info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 
@@ -194,6 +196,7 @@ def extract_bitrate(url):
                     'tbr': extract_bitrate(format_url),
                 })
 
+        download_base = 'http://rtsww%s-d.rts.ch/' % ('-a' if media_type == 'audio' else '')
         for media in info.get('media', []):
             media_url = media.get('url')
             if not media_url or re.match(r'https?://', media_url):
@@ -205,12 +208,11 @@ def extract_bitrate(url):
                 format_id += '-%dk' % rate
             formats.append({
                 'format_id': format_id,
-                'url': 'http://download-video.rts.ch/' + media_url,
+                'url': urljoin(download_base, media_url),
                 'tbr': rate or extract_bitrate(media_url),
             })
 
         self._check_formats(formats, media_id)
-        self._sort_formats(formats)
 
         duration = info.get('duration') or info.get('cutout') or info.get('cutduration')
         if isinstance(duration, compat_str):