[ie/youtube] Extract upload timestamp if available (#9856)

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

diff --git a/yt_dlp/extractor/arte.py b/yt_dlp/extractor/arte.py
index e3cc5afb05682835da1a4da58f43c656dd6bdc3a..46fe006cc93b61dabf3c3cd287b3aa54c5fefa7d 100644 (file)
--- a/yt_dlp/extractor/arte.py
+++ b/yt_dlp/extractor/arte.py
@@ -5,6 +5,7 @@
     ExtractorError,
     GeoRestrictedError,
     int_or_none,
+    join_nonempty,
     parse_iso8601,
     parse_qs,
     strip_or_none,
@@ -31,34 +32,10 @@ class ArteTVIE(ArteTVBaseIE):
     _TESTS = [{
         'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
         'only_matching': True,
-    }, {
-        'url': 'https://www.arte.tv/pl/videos/100103-000-A/usa-dyskryminacja-na-porodowce/',
-        'info_dict': {
-            'id': '100103-000-A',
-            'title': 'USA: Dyskryminacja na porodówce',
-            'description': 'md5:242017b7cce59ffae340a54baefcafb1',
-            'alt_title': 'ARTE Reportage',
-            'upload_date': '20201103',
-            'duration': 554,
-            'thumbnail': r're:https://api-cdn\.arte\.tv/.+940x530',
-            'timestamp': 1604417980,
-            'ext': 'mp4',
-        },
-        'params': {'skip_download': 'm3u8'}
     }, {
         'note': 'No alt_title',
         'url': 'https://www.arte.tv/fr/videos/110371-000-A/la-chaleur-supplice-des-arbres-de-rue/',
-        'info_dict': {
-            'id': '110371-000-A',
-            'ext': 'mp4',
-            'upload_date': '20220718',
-            'duration': 154,
-            'timestamp': 1658162460,
-            'description': 'md5:5890f36fe7dccfadb8b7c0891de54786',
-            'title': 'La chaleur, supplice des arbres de rue',
-            'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/CPE2sQDtD8GLQgt8DuYHLf/940x530',
-        },
-        'params': {'skip_download': 'm3u8'}
+        'only_matching': True,
     }, {
         'url': 'https://api.arte.tv/api/player/v2/config/de/100605-013-A',
         'only_matching': True,
@@ -67,19 +44,38 @@ class ArteTVIE(ArteTVBaseIE):
         'only_matching': True,
     }, {
         'url': 'https://www.arte.tv/de/videos/110203-006-A/zaz/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.arte.tv/fr/videos/109067-000-A/la-loi-de-teheran/',
         'info_dict': {
-            'id': '110203-006-A',
-            'chapters': 'count:16',
-            'description': 'md5:cf592f1df52fe52007e3f8eac813c084',
-            'alt_title': 'Zaz',
-            'title': 'Baloise Session 2022',
-            'timestamp': 1668445200,
-            'duration': 4054,
-            'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/ubQjmVCGyRx3hmBuZEK9QZ/940x530',
-            'upload_date': '20221114',
+            'id': '109067-000-A',
             'ext': 'mp4',
+            'description': 'md5:d2ca367b8ecee028dddaa8bd1aebc739',
+            'timestamp': 1713927600,
+            'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/3rR6PLzfbigSkkeHtkCZNF/940x530',
+            'duration': 7599,
+            'title': 'La loi de Téhéran',
+            'upload_date': '20240424',
+            'subtitles': {
+                'fr': 'mincount:1',
+                'fr-acc': 'mincount:1',
+                'fr-forced': 'mincount:1',
+            },
         },
-        'expected_warnings': ['geo restricted']
+    }, {
+        'note': 'age-restricted',
+        'url': 'https://www.arte.tv/de/videos/006785-000-A/the-element-of-crime/',
+        'info_dict': {
+            'id': '006785-000-A',
+            'description': 'md5:c2f94fdfefc8a280e4dab68ab96ab0ba',
+            'title': 'The Element of Crime',
+            'timestamp': 1696111200,
+            'duration': 5849,
+            'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/q82dTTfyuCXupPsGxXsd7B/940x530',
+            'upload_date': '20230930',
+            'ext': 'mp4',
+        },
+        'skip': '404 Not Found',
     }]
 
     _GEO_BYPASS = True
@@ -130,13 +126,27 @@ class ArteTVIE(ArteTVBaseIE):
         ),
     }
 
+    @staticmethod
+    def _fix_accessible_subs_locale(subs):
+        updated_subs = {}
+        for lang, sub_formats in subs.items():
+            for fmt in sub_formats:
+                url = fmt.get('url') or ''
+                suffix = ('acc' if url.endswith('-MAL.m3u8')
+                          else 'forced' if '_VO' not in url
+                          else None)
+                updated_subs.setdefault(join_nonempty(lang, suffix), []).append(fmt)
+        return updated_subs
+
     def _real_extract(self, url):
         mobj = self._match_valid_url(url)
         video_id = mobj.group('id')
         lang = mobj.group('lang') or mobj.group('lang_2')
-        langauge_code = self._LANG_MAP.get(lang)
+        language_code = self._LANG_MAP.get(lang)
 
-        config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id)
+        config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id, headers={
+            'x-validated-age': '18'
+        })
 
         geoblocking = traverse_obj(config, ('data', 'attributes', 'restriction', 'geoblocking')) or {}
         if geoblocking.get('restrictedArea'):
@@ -160,16 +170,16 @@ def _real_extract(self, url):
             m = self._VERSION_CODE_RE.match(stream_version_code)
             if m:
                 lang_pref = int(''.join('01'[x] for x in (
-                    m.group('vlang') == langauge_code,      # we prefer voice in the requested language
+                    m.group('vlang') == language_code,      # we prefer voice in the requested language
                     not m.group('audio_desc'),              # and not the audio description version
                     bool(m.group('original_voice')),        # but if voice is not in the requested language, at least choose the original voice
-                    m.group('sub_lang') == langauge_code,   # if subtitles are present, we prefer them in the requested language
+                    m.group('sub_lang') == language_code,   # if subtitles are present, we prefer them in the requested language
                     not m.group('has_sub'),                 # but we prefer no subtitles otherwise
                     not m.group('sdh_sub'),                 # and we prefer not the hard-of-hearing subtitles if there are subtitles
                 )))
 
             short_label = traverse_obj(stream_version, 'shortLabel', expected_type=str, default='?')
-            if stream['protocol'].startswith('HLS'):
+            if 'HLS' in stream['protocol']:
                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
                     stream['url'], video_id=video_id, ext='mp4', m3u8_id=stream_version_code, fatal=False)
                 for fmt in fmts:
@@ -181,6 +191,7 @@ def _real_extract(self, url):
                     secondary_formats.extend(fmts)
                 else:
                     formats.extend(fmts)
+                subs = self._fix_accessible_subs_locale(subs)
                 self._merge_subtitles(subs, target=subtitles)
 
             elif stream['protocol'] in ('HTTPS', 'RTMP'):