Merge 2358117d39 into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -446,6 +446,11 @@ from .cspan import (
     CSpanCongressIE,
     CSpanIE,
 )
+from .ctc import (
+    CTCIE,
+    CTCSeasonIE,
+    CTCSeriesIE,
+)
 from .ctsnews import CtsNewsIE
 from .ctv import CTVIE
 from .ctvnews import CTVNewsIE

yt_dlp/extractor/archiveorg.py

@@ -205,6 +205,26 @@ class ArchiveOrgIE(InfoExtractor):
                 },
             },
         ],
+    }, {
+        # The reviewbody is None for one of the reviews; just need to extract data without crashing
+        'url': 'https://archive.org/details/gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+        'info_dict': {
+            'id': 'gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+            'ext': 'mp3',
+            'title': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'creators': ['Grateful Dead'],
+            'duration': 338.31,
+            'track': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'description': 'md5:764348a470b986f1217ffd38d6ac7b72',
+            'display_id': 'gd95-04-02d1t04.shn',
+            'location': 'Pyramid Arena',
+            'uploader': 'jon@archive.org',
+            'album': '1995-04-02 - Pyramid Arena',
+            'upload_date': '20040519',
+            'track_number': 4,
+            'release_date': '19950402',
+            'timestamp': 1084927901,
+        },
     }]
 
     @staticmethod
@@ -335,7 +355,7 @@ class ArchiveOrgIE(InfoExtractor):
                 info['comments'].append({
                     'id': review.get('review_id'),
                     'author': review.get('reviewer'),
-                    'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
+                    'text': join_nonempty('reviewtitle', 'reviewbody', from_dict=review, delim='\n\n'),
                     'timestamp': unified_timestamp(review.get('createdate')),
                     'parent': 'root'})
 

yt_dlp/extractor/ctc.py

@@ -0,0 +1,164 @@
+from .common import InfoExtractor
+from ..utils import (
+    traverse_obj, 
+    int_or_none
+)
+
+class CTCIE(InfoExtractor):
+    IE_NAME = 'ctc'
+    _VALID_URL = (
+        r'https?://ctc\.ru/projects/filmi/(?P<project_name>[^/]+)/?(?:video/?)?$'
+        r'|https?://ctc\.ru/projects/(?P<category>show|multiki|serials)/(?P<project_name2>[^/]+)/video/'
+        r'(?:$'
+        r'|(?P<season_number>\d+)-sezon/(?P<episode_number>\d+)-(?:vypusk|serija)/?$'
+        r'|promo/[^/]+/?$'
+        r')'
+    )
+    _GEO_COUNTRIES = ['RU']
+
+    def _real_extract(self, url):
+        url_slug = url.split("https://ctc.ru/")[1]
+
+        item_response = self._download_json(
+            f'https://ctc.ru/api/page/v1/{url_slug}', url_slug,
+            note='Downloading item data', headers={
+                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:97.0) Gecko/20100101 Firefox/97.0"
+            }
+        )
+
+        track_hub_id = str(traverse_obj(item_response, ('content', 0, 'trackHubId'), expected_type=int))
+        if not track_hub_id:
+            self.raise_no_formats('trackHubId not found')
+
+        video_url = traverse_obj(
+                item_response, ('content', 0, 'videoUrl'), get_all=False
+            ) or traverse_obj(
+                item_response, ('content', 0, 'trackUrl'), get_all=False
+        )
+        stream_response = self._download_json(
+            video_url.replace("/player/", "/playlist/"), 
+            track_hub_id,
+            note='Downloading stream data', headers={
+                'X-Referer': 'https://ctc.ru',
+                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:97.0) Gecko/20100101 Firefox/97.0',
+            }
+        )
+
+        if traverse_obj(stream_response, ('playlist', 'items', 0, 'errors', 0, 'code')) == 102:
+            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
+        elif traverse_obj(stream_response, ('playlist', 'items', 0, 'errors', 0, 'code')) == 103:
+            self.raise_login_required(msg='This video is only available for registered users with the required subscription')
+
+        formats = []
+        for stream in traverse_obj(stream_response, ('playlist', 'items', 0, 'streams')):
+            protocol = stream.get('protocol')
+            video_id = traverse_obj(stream_response, ('playlist', 'items', 0, 'track_id'))
+
+            fmts = []
+            if protocol == 'HLS':
+                fmts, _ = self._extract_m3u8_formats_and_subtitles(
+                    stream.get('url'), video_id, ext='mp4', preference=1, m3u8_id="video/hls", fatal=False)
+            formats.extend(fmts)
+
+        return {
+            'formats': formats,
+            'id': track_hub_id,
+            'title': ": ".join(filter(None, [
+                traverse_obj(stream_response, ('playlist', 'items', 0, 'project_name')),
+                traverse_obj(stream_response, ('playlist', 'items', 0, 'episode_name'))
+            ])),
+            **traverse_obj(item_response, {
+                'description': ('content', 0, 'description'),
+            }),
+            **traverse_obj(stream_response, {
+                'episode': ('playlist', 'items', 0, 'episode_name'),
+                'duration': ('playlist', 'items', 0, 'duration'),
+                'thumbnail': ('playlist', 'items', 0, 'thumbnail_url'),
+                'season': ('playlist', 'items', 0, 'season_name'),
+                'age_limit': ('playlist', 'items', 0, 'min_age'),
+            }),
+            'season_number': int_or_none(self._match_valid_url(url).group('season_number')),
+            'episode_number': int_or_none(self._match_valid_url(url).group('episode_number')),
+        }
+
+
+class CTCSeasonIE(InfoExtractor):
+    IE_NAME = 'ctc:season'
+    _VALID_URL = (
+        r'https?://ctc\.ru/projects/(?P<category>show|multiki|serials)/'
+        r'(?P<project_name>[^/]+)/video/(?P<season_number>\d+)-sezon/?$'
+    )
+    _GEO_COUNTRIES = ['RU']
+
+    def _real_extract(self, url):
+        url_slug = url.split("https://ctc.ru/")[1]
+        
+        season_data = self._download_json(f'https://ctc.ru/api/page/v1/{url_slug}', url_slug)
+
+        entries = [{
+            '_type': 'url',
+            'title': episode.get('title'),
+            'url': f"https://ctc.ru{episode.get('popupUrl')}",
+            'ie_key': CTCIE.ie_key(),
+            'season_number': self._match_valid_url(url).group('season_number'),
+            'episode_number': self._search_regex(
+                r'/(?P<episode>\d+)-(vypusk|serija)/', 
+                episode.get('popupUrl'), 
+                'episode number', 
+            )
+        } for episode in traverse_obj(season_data, ('content', 1, 'widgets')) if episode.get('popupUrl')]
+
+        return {
+            '_type': 'playlist',    
+            'entries': entries,
+            **traverse_obj(season_data, {
+                'id': ('content', 9, 'entityId'),
+                'title': ('content', 0, 'widgets', 1, 'title'),
+                'season_number': self._match_valid_url(url).group('season_number'),
+                'series': ('content', 0, 'widgets', 1, 'title'),
+                'description': ('content', 0, 'widgets', 1, 'description'),
+            }),
+        }
+
+
+class CTCSeriesIE(InfoExtractor):
+    IE_NAME = 'ctc:series'
+    _VALID_URL = (
+        r'https?://ctc\.ru/projects/(?P<category>show|multiki|serials)/'
+        r'(?P<slug>[^/]+)/?$'
+    )
+    _GEO_COUNTRIES = ['RU']
+
+    def _real_extract(self, url):
+        url_slug = url.split("https://ctc.ru/")[1]
+
+        series_data = self._download_json(f'https://ctc.ru/api/page/v1/{url_slug}', url_slug)
+
+        # cartoons doesn't indicate in the url what type it is, so
+        # if it's a movie, then redirect it to CTCIE
+        if traverse_obj(series_data, ('content', 5, 'tabs')) == []:
+            return self.url_result(f'https://ctc.ru/{url_slug}/video')
+
+        entries = [{
+            '_type': 'url',
+            'title': season.get('title'),
+            'url': f'https://ctc.ru{season.get("url")}',
+            'ie_key': CTCSeasonIE.ie_key(),
+            'season_number': self._search_regex(
+                r'/(?P<episode>\d+)-sezon/', 
+                season.get("url"), 
+                'season number', 
+            ),
+            'series': season.get('title'),
+        } for season in traverse_obj(series_data, ('content', 1, 'tabs')) if season.get("url").endswith("sezon/")]
+
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            **traverse_obj(series_data, {
+                'id': ('content', 0, 'projectId'),
+                'title': ('content', 0, 'widgets', 1, 'title'),
+                'series': ('content', 0, 'widgets', 1, 'title'),
+                'description': ('content', 0, 'widgets', 1, 'description')
+            }),
+        }

yt_dlp/extractor/facebook.py

@@ -563,13 +563,13 @@ class FacebookIE(InfoExtractor):
                 return extract_video_data(try_get(
                     js_data, lambda x: x['jsmods']['instances'], list) or [])
 
-        def extract_dash_manifest(video, formats):
+        def extract_dash_manifest(vid_data, formats, mpd_url=None):
             dash_manifest = traverse_obj(
-                video, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', expected_type=str)
+                vid_data, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 'manifest_xml', expected_type=str)
             if dash_manifest:
                 formats.extend(self._parse_mpd_formats(
                     compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
-                    mpd_url=url_or_none(video.get('dash_manifest_url'))))
+                    mpd_url=url_or_none(video.get('dash_manifest_url')) or mpd_url))
 
         def process_formats(info):
             # Downloads with browser's User-Agent are rate limited. Working around
@@ -619,9 +619,12 @@ class FacebookIE(InfoExtractor):
                         video = video['creation_story']
                         video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
                         video.update(reel_info)
-                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
+
                     formats = []
                     q = qualities(['sd', 'hd'])
+
+                    # Legacy formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
                     for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
                                            ('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
                                            ('browser_native_sd_url', 'sd')):
@@ -629,7 +632,7 @@ class FacebookIE(InfoExtractor):
                         if not playable_url:
                             continue
                         if determine_ext(playable_url) == 'mpd':
-                            formats.extend(self._extract_mpd_formats(playable_url, video_id))
+                            formats.extend(self._extract_mpd_formats(playable_url, video_id, fatal=False))
                         else:
                             formats.append({
                                 'format_id': format_id,
@@ -638,6 +641,28 @@ class FacebookIE(InfoExtractor):
                                 'url': playable_url,
                             })
                     extract_dash_manifest(fmt_data, formats)
+
+                    # New videoDeliveryResponse formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryResponseFragment', 'videoDeliveryResponseResult'))
+                    mpd_urls = traverse_obj(fmt_data, ('dash_manifest_urls', ..., 'manifest_url', {url_or_none}))
+                    dash_manifests = traverse_obj(fmt_data, ('dash_manifests', lambda _, v: v['manifest_xml']))
+                    for idx, dash_manifest in enumerate(dash_manifests):
+                        extract_dash_manifest(dash_manifest, formats, mpd_url=traverse_obj(mpd_urls, idx))
+                    if not dash_manifests:
+                        # Only extract from MPD URLs if the manifests are not already provided
+                        for mpd_url in mpd_urls:
+                            formats.extend(self._extract_mpd_formats(mpd_url, video_id, fatal=False))
+                    for prog_fmt in traverse_obj(fmt_data, ('progressive_urls', lambda _, v: v['progressive_url'])):
+                        format_id = traverse_obj(prog_fmt, ('metadata', 'quality', {str.lower}))
+                        formats.append({
+                            'format_id': format_id,
+                            # sd, hd formats w/o resolution info should be deprioritized below DASH
+                            'quality': q(format_id) - 3,
+                            'url': prog_fmt['progressive_url'],
+                        })
+                    for m3u8_url in traverse_obj(fmt_data, ('hls_playlist_urls', ..., 'hls_playlist_url', {url_or_none})):
+                        formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', fatal=False, m3u8_id='hls'))
+
                     if not formats:
                         # Do not append false positive entry w/o any formats
                         return