Merge 8f5a765e25 into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -1518,8 +1518,12 @@ from .pgatour import PGATourIE
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pialive import PiaLiveIE
 from .piapro import PiaproIE
-from .piaulizaportal import PIAULIZAPortalIE
+from .piaulizaportal import (
+    PIAULIZAPortalAPIIE,
+    PIAULIZAPortalIE,
+)
 from .picarto import (
     PicartoIE,
     PicartoVodIE,

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/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

yt_dlp/extractor/pialive.py

@@ -0,0 +1,126 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    extract_attributes,
+    get_element_by_class,
+    get_element_html_by_class,
+    multipart_encode,
+    unified_timestamp,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class PiaLiveIE(InfoExtractor):
+    PLAYER_ROOT_URL = 'https://player.pia-live.jp/'
+    PIA_LIVE_API_URL = 'https://api.pia-live.jp'
+    API_KEY = 'kfds)FKFps-dms9e'
+    _VALID_URL = r'https?://player\.pia-live\.jp/stream/(?P<id>[\w-]+)'
+
+    _TESTS = [
+        {
+            'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krUDqGOwN4d61dCWQYOd6CTxl4hjya9dsfEZGsM4uGOUdax60lEI4twsXGXf7crmz8Gk__GhupTrWxA7RFRVt76',
+            'info_dict': {
+                'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'display_id': '2431867_001',
+                'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+                'live_status': 'was_live',
+                'comment_count': int,
+            },
+            'params': {
+                'getcomments': True,
+                'skip_download': True,
+                'ignore_no_formats_error': True,
+            },
+            'skip': 'The video is no longer available',
+        },
+        {
+            'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krJdu0GVBVbVy01IwpJ6J3qBEm3d9TCTt1d0eWpsZGj7DrOjVOmS7GAWGwyscMgiThopJvzgWC4H5b-7XQjAfRZ',
+            'info_dict': {
+                'id': '9ce8b8ba-f6d1-4d1f-83a0-18c3148ded93',
+                'display_id': '2431867_002',
+                'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+                'live_status': 'was_live',
+                'comment_count': int,
+            },
+            'params': {
+                'getcomments': True,
+                'skip_download': True,
+                'ignore_no_formats_error': True,
+            },
+            'skip': 'The video is no longer available',
+        },
+    ]
+
+    def _extract_vars(self, variable, html):
+        return self._search_regex(
+            rf'(?:var|const|let)\s+{variable}\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+            html, f'variable {variable}', group='value')
+
+    def _real_extract(self, url):
+        video_key = self._match_id(url)
+        webpage = self._download_webpage(url, video_key)
+
+        program_code = self._extract_vars('programCode', webpage)
+        article_code = self._extract_vars('articleCode', webpage)
+        title = self._html_extract_title(webpage)
+
+        if get_element_html_by_class('play-end', webpage):
+            raise ExtractorError('The video is no longer available', expected=True, video_id=program_code)
+
+        if start_info := clean_html(get_element_by_class('play-waiting__date', webpage)):
+            date, time = self._search_regex(
+                r'(?P<date>\d{4}/\d{1,2}/\d{1,2})\([月火水木金土日]\)(?P<time>\d{2}:\d{2})',
+                start_info, 'start_info', fatal=False, group=('date', 'time'))
+            if date and time:
+                release_timestamp_str = f'{date} {time} +09:00'
+                release_timestamp = unified_timestamp(release_timestamp_str)
+                self.raise_no_formats(f'The video will be available after {release_timestamp_str}', expected=True)
+                return {
+                    'id': program_code,
+                    'title': title,
+                    'live_status': 'is_upcoming',
+                    'release_timestamp': release_timestamp,
+                }
+
+        payload, content_type = multipart_encode({
+            'play_url': video_key,
+            'api_key': self.API_KEY,
+        })
+        api_kwargs = {
+            'video_id': program_code,
+            'data': payload,
+            'headers': {'Content-Type': content_type, 'Referer': self.PLAYER_ROOT_URL},
+        }
+
+        player_tag_list = self._download_json(
+            f'{self.PIA_LIVE_API_URL}/perf/player-tag-list/{program_code}', **api_kwargs,
+            note='Fetching player tag list', errnote='Unable to fetch player tag list')
+        chat_room_url = None
+        if self.get_param('getcomments'):
+            chat_room_url = traverse_obj(self._download_json(
+                f'{self.PIA_LIVE_API_URL}/perf/chat-tag-list/{program_code}/{article_code}', **api_kwargs,
+                note='Fetching chat info', errnote='Unable to fetch chat info', fatal=False),
+                ('data', 'chat_one_tag', {extract_attributes}, 'src', {url_or_none}))
+
+        return self.url_result(
+            extract_attributes(player_tag_list['data']['movie_one_tag'])['src'], url_transparent=True,
+            video_title=title, display_id=program_code, __post_extractor=self.extract_comments(
+                program_code, chat_room_url))
+
+    def _get_comments(self, video_id, chat_room_url):
+        if not chat_room_url:
+            return
+        if comment_page := self._download_webpage(
+                chat_room_url, video_id, headers={'Referer': self.PLAYER_ROOT_URL},
+                note='Fetching comment page', errnote='Unable to fetch comment page', fatal=False):
+            yield from traverse_obj(self._search_json(
+                r'var\s+_history\s*=', comment_page, 'comment list',
+                video_id, contains_pattern=r'\[(?s:.+)\]', fatal=False), (..., {
+                    'timestamp': 0,
+                    'author_is_uploader': (1, {lambda x: x == 2}),
+                    'author': 2,
+                    'text': 3,
+                    'id': 4,
+                }))

yt_dlp/extractor/piaulizaportal.py

@@ -1,11 +1,62 @@
 from .common import InfoExtractor
-from ..utils import (
+from ..utils import ExtractorError, int_or_none, parse_qs, time_seconds
-    ExtractorError,
+from ..utils.traversal import traverse_obj
-    int_or_none,
+
-    parse_qs,
+
-    time_seconds,
+class PIAULIZAPortalAPIIE(InfoExtractor):
-    traverse_obj,
+    _VALID_URL = r'https://player-api\.p\.uliza\.jp/v1/players/[^?#]+\?(?:[^#]*&)?name=(?P<id>[^#&]+)'
-)
+    _TESTS = [
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/timeshift-disabled/pia/admin?type=normal&playerobjectname=ulizaPlayer&name=livestream01_dvr&repeatable=true',
+            'info_dict': {
+                'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'ext': 'mp4',
+                'title': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'live_status': 'was_live',
+            },
+        },
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/uliza_jp_gallery_normal/promotion/admin?type=presentation&name=cookings&targetid=player1',
+            'info_dict': {
+                'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+                'ext': 'mp4',
+                'title': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+                'live_status': 'not_live',
+            },
+        },
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/default-player/pia/admin?type=normal&name=pia_movie_uliza_fix&targetid=ulizahtml5&repeatable=true',
+            'info_dict': {
+                'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+                'ext': 'mp4',
+                'title': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+                'live_status': 'not_live',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        player_data = self._download_webpage(
+            url, display_id, headers={'Referer': 'https://player-api.p.uliza.jp/'},
+            note='Fetching player data', errnote='Unable to fetch player data')
+
+        m3u8_url = self._search_regex(
+            r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data, 'm3u8 url')
+        video_id = parse_qs(m3u8_url).get('ss', [display_id])[0]
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id)
+        m3u8_type = self._search_regex(
+            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
+        return {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+            'live_status': {
+                'video': 'is_live',
+                'dvr': 'was_live',  # short-term archives
+            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
+        }
 
 
 class PIAULIZAPortalIE(InfoExtractor):
@@ -14,7 +65,8 @@ class PIAULIZAPortalIE(InfoExtractor):
     _TESTS = [{
         'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
         'info_dict': {
-            'id': '005f18b7-e810-5618-cb82-0987c5755d44',
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'display_id': '005f18b7-e810-5618-cb82-0987c5755d44',
             'title': 'プレゼンテーションプレイヤーのサンプル',
             'live_status': 'not_live',
         },
@@ -25,7 +77,8 @@ class PIAULIZAPortalIE(InfoExtractor):
     }, {
         'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
         'info_dict': {
-            'id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'display_id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
             'title': '【確認用】視聴サンプルページ（ULIZA）',
             'live_status': 'not_live',
         },
@@ -44,27 +97,9 @@ class PIAULIZAPortalIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        player_data = self._download_webpage(
+        player_data_url = self._search_regex(
-            self._search_regex(
+            r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
-                r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
+            webpage, 'player data url')
-                webpage, 'player data url'),
+        return self.url_result(
-            video_id, headers={'Referer': 'https://ulizaportal.jp/'},
+            player_data_url, PIAULIZAPortalAPIIE, url_transparent=True,
-            note='Fetching player data', errnote='Unable to fetch player data')
+            display_id=video_id, video_title=self._html_extract_title(webpage))
-
-        formats = self._extract_m3u8_formats(
-            self._search_regex(
-                r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data,
-                'm3u8 url', default=None),
-            video_id, fatal=False)
-        m3u8_type = self._search_regex(
-            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
-
-        return {
-            'id': video_id,
-            'title': self._html_extract_title(webpage),
-            'formats': formats,
-            'live_status': {
-                'video': 'is_live',
-                'dvr': 'was_live',  # short-term archives
-            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
-        }