Merge 2c0244cb2f into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -169,6 +169,10 @@ from .asobichannel import (
     AsobiChannelTagURLIE,
 )
 from .asobistage import AsobiStageIE
+from .atptour import (
+    ATPTourNewsIE,
+    ATPTourVideoIE,
+)
 from .atresplayer import AtresPlayerIE
 from .atscaleconf import AtScaleConfEventIE
 from .atvat import ATVAtIE

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/atptour.py

@@ -0,0 +1,127 @@
+import re
+
+from .brightcove import BrightcoveNewIE
+from .common import InfoExtractor
+from ..utils import base_url, extract_attributes, get_element_html_by_id, traverse_obj, urljoin
+
+
+class ATPTourVideoIE(InfoExtractor):
+    IE_NAME = 'atptour:video'
+    _VALID_URL = r'https?://(?:www\.)?atptour\.com/(?:en|es)/video/(?P<id>[\w-]+)'
+    _TESTS = [{
+        'url': 'https://www.atptour.com/en/video/challenger-highlights-nishikori-wins-in-como-2024',
+        'md5': '4721002227d98fe89afafa40eba3068d',
+        'info_dict': {
+            'id': '6361099221112',
+            'ext': 'mp4',
+            'description': 'md5:ef8afed21c52cbe4ad3409045d59f413',
+            'upload_date': '20240827',
+            'duration': 105.152,
+            'tags': 'count:6',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'title': 'Challenger Highlights: Nishikori wins in Como 2024',
+            'uploader_id': '6057277721001',
+            'timestamp': 1724775281,
+        },
+    }, {
+        'url': 'https://www.atptour.com/en/video/highlights-svajda-earns-highestranked-win-of-career-vs-cerundolo-winstonsalem-2024',
+        'md5': 'a3829d10bdcb1829568fd88b9e6ecb15',
+        'info_dict': {
+            'id': '6360716257112',
+            'ext': 'mp4',
+            'description': 'md5:a334aeb73eac631ffab8249b1e68194c',
+            'upload_date': '20240820',
+            'duration': 139.691,
+            'tags': 'count:5',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'title': 'Highlights: Svajda earns highest-ranked win of career vs. Cerundolo Winston-Salem 2024',
+            'uploader_id': '6057277721001',
+            'timestamp': 1724183755,
+        },
+    }, {
+        'url': 'https://www.atptour.com/es/video/highlights-michelsen-defeats-fucsovics-in-winston-salem-2024',
+        'md5': '7ba4c3aabef9eb20a1b9877f28e6f775',
+        'info_dict': {
+            'id': '6360727636112',
+            'ext': 'mp4',
+            'description': 'md5:2c5682fdfa514e508c6d947e9e9b6eeb',
+            'upload_date': '20240821',
+            'duration': 135.424,
+            'tags': 'count:6',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'title': 'Highlights: Michelsen defeats Fucsovics in Winston-Salem 2024',
+            'uploader_id': '6057277721001',
+            'timestamp': 1724205624,
+        },
+    }, {
+        'url': 'https://www.atptour.com/en/video/highlights-sonego-dominates-michelsen-for-winston-salem-open-title-2024',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id, fatal=False, impersonate=True)
+
+        hidden_inputs = self._hidden_inputs(webpage, 'class')
+        featured_videos_url = urljoin(base_url(url), hidden_inputs.get('atp_featured-videos-endpoint'))
+        json_data = self._download_json(featured_videos_url, display_id, fatal=False, impersonate=True)
+        video_data = traverse_obj(json_data, ('content', 0))
+        account_id = traverse_obj(video_data, ('videoAccountId'))
+        player_id = traverse_obj(video_data, ('videoPlayerId'))
+        video_id = traverse_obj(video_data, ('videoId'))
+        return self.url_result(
+            f'https://players.brightcove.net/{account_id}/{player_id}/index.html?videoId={video_id}', BrightcoveNewIE)
+
+
+class ATPTourNewsIE(InfoExtractor):
+    IE_NAME = 'atptour:news'
+    _VALID_URL = r'https?://(?:www\.)?atptour\.com/(?:en|es)/news/(?P<id>[\w-]+)'
+    _TESTS = [{
+        'url': 'https://www.atptour.com/en/news/sinner-zverev-cincinnati-2024-sf',
+        'playlist_mincount': 2,
+        'info_dict': {
+            'id': 'sinner-zverev-cincinnati-2024-sf',
+            'title': 'Jannik Sinner battles past Alexander Zverev to reach Cincinnati final | ATP Tour | Tennis',
+            'description': 'md5:30cd3df666c8a5d45731d1e85d8d43ae',
+        },
+    }, {
+        'url': 'https://www.atptour.com/en/news/borges-us-open-2024-this-is-tennis',
+        'playlist_mincount': 1,
+        'info_dict': {
+            'id': 'borges-us-open-2024-this-is-tennis',
+            'title': 'Nuno Borges: Building legos, facing Nadal, Cirque du Soleil & more | ATP Tour | Tennis',
+            'description': 'md5:aaef866660c4e3ced69118c0f6ed237a',
+        },
+    }, {
+        'url': 'https://www.atptour.com/es/news/popyrin-us-open-2024-feature',
+        'playlist_mincount': 1,
+        'info_dict': {
+            'id': 'popyrin-us-open-2024-feature',
+            'title': 'Alexei Popyrin: Hamilton, pollo frito y la revancha de Djokovic | ATP Tour | Tennis',
+            'description': 'md5:b62a35720a278c9ab8410847915dc581',
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id, fatal=False, impersonate=True)
+
+        title = self._html_extract_title(webpage)
+        description = self._og_search_description(webpage)
+
+        entries = []
+
+        first_video = get_element_html_by_id('articleVideoJSPlayer', webpage)
+        if first_video is not None:
+            attributes = extract_attributes(first_video)
+            account_id = traverse_obj(attributes, ('data-account'))
+            player_id = traverse_obj(attributes, ('data-player'))
+            video_id = traverse_obj(attributes, ('data-video-id'))
+            first_video_url = f'https://players.brightcove.net/{account_id}/{player_id}/index.html?videoId={video_id}'
+            entries.append(self.url_result(first_video_url, BrightcoveNewIE))
+
+        iframe_urls = re.findall(r'<iframe[^>]src="(https://players\.brightcove\.net/[^"]+)"', webpage)
+        for video_url in iframe_urls:
+            entries.append(self.url_result(video_url, BrightcoveNewIE))
+
+        return self.playlist_result(entries, display_id, title, description)

yt_dlp/extractor/common.py

@@ -1801,7 +1801,7 @@ class InfoExtractor:
         return traverse_obj(ret, traverse) or {}
 
     @staticmethod
-    def _hidden_inputs(html):
+    def _hidden_inputs(html, attr_list=('name', 'id')):
         html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)
         hidden_inputs = {}
         for input_el in re.findall(r'(?i)(<input[^>]+>)', html):
@@ -1810,7 +1810,10 @@ class InfoExtractor:
                 continue
             if attrs.get('type') not in ('hidden', 'submit'):
                 continue
-            name = attrs.get('name') or attrs.get('id')
+            for attr in variadic(attr_list):
+                name = attrs.get(attr)
+                if name is not None:
+                    break
             value = attrs.get('value')
             if name and value is not None:
                 hidden_inputs[name] = value

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