From 9cd90447907a59c8a2727583f4a755fb23ed8cd3 Mon Sep 17 00:00:00 2001 From: chtk Date: Mon, 22 Jan 2024 06:57:52 +0100 Subject: [PATCH] [ie/Floatplane] Improve metadata extraction (#8934) Authored by: chtk --- yt_dlp/extractor/floatplane.py | 103 +++++++++++++++++++++++++++------ 1 file changed, 84 insertions(+), 19 deletions(-) diff --git a/yt_dlp/extractor/floatplane.py b/yt_dlp/extractor/floatplane.py index 2cf4d4e64..8676d73f6 100644 --- a/yt_dlp/extractor/floatplane.py +++ b/yt_dlp/extractor/floatplane.py @@ -11,6 +11,7 @@ join_nonempty, parse_codecs, parse_iso8601, + url_or_none, urljoin, ) from ..utils.traversal import traverse_obj @@ -108,6 +109,64 @@ class FloatplaneIE(InfoExtractor): 'availability': 'subscriber_only', }, 'params': {'skip_download': 'm3u8'}, + }, { + 'url': 'https://www.floatplane.com/post/65B5PNoBtf', + 'info_dict': { + 'id': '65B5PNoBtf', + 'description': 'I recorded the inbuilt demo mode for your 90\'s enjoyment, thanks for being Floaties!', + 'display_id': '65B5PNoBtf', + 'like_count': int, + 'release_timestamp': 1701249480, + 'uploader': 'The Trash Network', + 'availability': 'subscriber_only', + 'uploader_id': '61bc20c9a131fb692bf2a513', + 'uploader_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home', + 'channel_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home/thedrumthing', + 'comment_count': int, + 'title': 'The $50 electronic drum kit.', + 'channel_id': '64424fe73cd58cbcf8d8e131', + 'thumbnail': 'https://pbs.floatplane.com/blogPost_thumbnails/65B5PNoBtf/725555379422705_1701247052743.jpeg', + 'dislike_count': int, + 'channel': 'The Drum Thing', + 'release_date': '20231129', + }, + 'playlist_count': 2, + 'playlist': [{ + 'info_dict': { + 'id': 'ISPJjexylS', + 'ext': 'mp4', + 'release_date': '20231129', + 'release_timestamp': 1701249480, + 'title': 'The $50 electronic drum kit. .mov', + 'channel_id': '64424fe73cd58cbcf8d8e131', + 'thumbnail': 'https://pbs.floatplane.com/video_thumbnails/ISPJjexylS/335202812134041_1701249383392.jpeg', + 'availability': 'subscriber_only', + 'uploader': 'The Trash Network', + 'duration': 622, + 'channel': 'The Drum Thing', + 'uploader_id': '61bc20c9a131fb692bf2a513', + 'channel_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home/thedrumthing', + 'uploader_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home', + }, + }, { + 'info_dict': { + 'id': 'qKfxu6fEpu', + 'ext': 'aac', + 'release_date': '20231129', + 'release_timestamp': 1701249480, + 'title': 'Roland TD-7 Demo.m4a', + 'channel_id': '64424fe73cd58cbcf8d8e131', + 'availability': 'subscriber_only', + 'uploader': 'The Trash Network', + 'duration': 114, + 'channel': 'The Drum Thing', + 'uploader_id': '61bc20c9a131fb692bf2a513', + 'channel_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home/thedrumthing', + 'uploader_url': 'https://www.floatplane.com/channel/TheTrashNetwork/home', + }, + }], + 'skip': 'requires subscription: "The Trash Network"', + 'params': {'skip_download': 'm3u8'}, }] def _real_initialize(self): @@ -124,6 +183,22 @@ def _real_extract(self, url): if not any(traverse_obj(post_data, ('metadata', ('hasVideo', 'hasAudio')))): raise ExtractorError('Post does not contain a video or audio track', expected=True) + uploader_url = format_field( + post_data, [('creator', 'urlname')], 'https://www.floatplane.com/channel/%s/home') or None + + common_info = { + 'uploader_url': uploader_url, + 'channel_url': urljoin(f'{uploader_url}/', traverse_obj(post_data, ('channel', 'urlname'))), + 'availability': self._availability(needs_subscription=True), + **traverse_obj(post_data, { + 'uploader': ('creator', 'title', {str}), + 'uploader_id': ('creator', 'id', {str}), + 'channel': ('channel', 'title', {str}), + 'channel_id': ('channel', 'id', {str}), + 'release_timestamp': ('releaseDate', {parse_iso8601}), + }), + } + items = [] for media in traverse_obj(post_data, (('videoAttachments', 'audioAttachments'), ...)): media_id = media['id'] @@ -150,11 +225,11 @@ def format_path(params): formats = [] for quality in traverse_obj(stream, ('resource', 'data', 'qualityLevels', ...)): url = urljoin(stream['cdn'], format_path(traverse_obj( - stream, ('resource', 'data', 'qualityLevelParams', quality['name'])))) + stream, ('resource', 'data', 'qualityLevelParams', quality['name'], {dict})))) formats.append({ **traverse_obj(quality, { - 'format_id': 'name', - 'format_note': 'label', + 'format_id': ('name', {str}), + 'format_note': ('label', {str}), 'width': ('width', {int}), 'height': ('height', {int}), }), @@ -164,38 +239,28 @@ def format_path(params): }) items.append({ + **common_info, 'id': media_id, **traverse_obj(metadata, { - 'title': 'title', + 'title': ('title', {str}), 'duration': ('duration', {int_or_none}), - 'thumbnail': ('thumbnail', 'path'), + 'thumbnail': ('thumbnail', 'path', {url_or_none}), }), 'formats': formats, }) - uploader_url = format_field( - post_data, [('creator', 'urlname')], 'https://www.floatplane.com/channel/%s/home') or None - channel_url = urljoin(f'{uploader_url}/', traverse_obj(post_data, ('channel', 'urlname'))) - post_info = { + **common_info, 'id': post_id, 'display_id': post_id, **traverse_obj(post_data, { - 'title': 'title', + 'title': ('title', {str}), 'description': ('text', {clean_html}), - 'uploader': ('creator', 'title'), - 'uploader_id': ('creator', 'id'), - 'channel': ('channel', 'title'), - 'channel_id': ('channel', 'id'), 'like_count': ('likes', {int_or_none}), 'dislike_count': ('dislikes', {int_or_none}), 'comment_count': ('comments', {int_or_none}), - 'release_timestamp': ('releaseDate', {parse_iso8601}), - 'thumbnail': ('thumbnail', 'path'), + 'thumbnail': ('thumbnail', 'path', {url_or_none}), }), - 'uploader_url': uploader_url, - 'channel_url': channel_url, - 'availability': self._availability(needs_subscription=True), } if len(items) > 1: