From f549b18512570d0c000179df9147415e4eba1649 Mon Sep 17 00:00:00 2001 From: bashonly <88596187+bashonly@users.noreply.github.com> Date: Fri, 9 Dec 2022 23:46:04 +0000 Subject: [PATCH] [extractor/pinterest] Fix extractor (#5739) Closes #1772 Authored by: bashonly --- yt_dlp/extractor/pinterest.py | 153 ++++++++++++++++++++++------------ 1 file changed, 102 insertions(+), 51 deletions(-) diff --git a/yt_dlp/extractor/pinterest.py b/yt_dlp/extractor/pinterest.py index 2c6cd6d4bb..8361fbbc5f 100644 --- a/yt_dlp/extractor/pinterest.py +++ b/yt_dlp/extractor/pinterest.py @@ -1,19 +1,24 @@ import json from .common import InfoExtractor -from ..compat import compat_str from ..utils import ( determine_ext, float_or_none, int_or_none, - try_get, + str_or_none, + strip_or_none, + traverse_obj, unified_timestamp, url_or_none, ) class PinterestBaseIE(InfoExtractor): - _VALID_URL_BASE = r'https?://(?:[^/]+\.)?pinterest\.(?:com|fr|de|ch|jp|cl|ca|it|co\.uk|nz|ru|com\.au|at|pt|co\.kr|es|com\.mx|dk|ph|th|com\.uy|co|nl|info|kr|ie|vn|com\.vn|ec|mx|in|pe|co\.at|hu|co\.in|co\.nz|id|com\.ec|com\.py|tw|be|uk|com\.bo|com\.pe)' + _VALID_URL_BASE = r'''(?x) + https?://(?:[^/]+\.)?pinterest\.(?: + com|fr|de|ch|jp|cl|ca|it|co\.uk|nz|ru|com\.au|at|pt|co\.kr|es|com\.mx| + dk|ph|th|com\.uy|co|nl|info|kr|ie|vn|com\.vn|ec|mx|in|pe|co\.at|hu| + co\.in|co\.nz|id|com\.ec|com\.py|tw|be|uk|com\.bo|com\.pe)''' def _call_api(self, resource, video_id, options): return self._download_json( @@ -24,14 +29,53 @@ class PinterestBaseIE(InfoExtractor): def _extract_video(self, data, extract_formats=True): video_id = data['id'] + thumbnails = [] + images = data.get('images') + if isinstance(images, dict): + for thumbnail_id, thumbnail in images.items(): + if not isinstance(thumbnail, dict): + continue + thumbnail_url = url_or_none(thumbnail.get('url')) + if not thumbnail_url: + continue + thumbnails.append({ + 'url': thumbnail_url, + 'width': int_or_none(thumbnail.get('width')), + 'height': int_or_none(thumbnail.get('height')), + }) - title = (data.get('title') or data.get('grid_title') or video_id).strip() + info = { + 'title': strip_or_none(traverse_obj(data, 'title', 'grid_title', default='')), + 'description': traverse_obj(data, 'seo_description', 'description'), + 'timestamp': unified_timestamp(data.get('created_at')), + 'thumbnails': thumbnails, + 'uploader': traverse_obj(data, ('closeup_attribution', 'full_name')), + 'uploader_id': str_or_none(traverse_obj(data, ('closeup_attribution', 'id'))), + 'repost_count': int_or_none(data.get('repin_count')), + 'comment_count': int_or_none(data.get('comment_count')), + 'categories': traverse_obj(data, ('pin_join', 'visual_annotation'), expected_type=list), + 'tags': traverse_obj(data, 'hashtags', expected_type=list), + } urls = [] formats = [] duration = None - if extract_formats: - for format_id, format_dict in data['videos']['video_list'].items(): + domain = data.get('domain', '') + if domain.lower() != 'uploaded by user' and traverse_obj(data, ('embed', 'src')): + if not info['title']: + info['title'] = None + return { + '_type': 'url_transparent', + 'url': data['embed']['src'], + **info, + } + + elif extract_formats: + video_list = traverse_obj( + data, ('videos', 'video_list'), + ('story_pin_data', 'pages', ..., 'blocks', ..., 'video', 'video_list'), + expected_type=dict, get_all=False, default={}) + for format_id, format_dict in video_list.items(): if not isinstance(format_dict, dict): continue format_url = url_or_none(format_dict.get('url')) @@ -53,72 +97,79 @@ class PinterestBaseIE(InfoExtractor): 'duration': duration, }) - description = data.get('description') or data.get('description_html') or data.get('seo_description') - timestamp = unified_timestamp(data.get('created_at')) - - def _u(field): - return try_get(data, lambda x: x['closeup_attribution'][field], compat_str) - - uploader = _u('full_name') - uploader_id = _u('id') - - repost_count = int_or_none(data.get('repin_count')) - comment_count = int_or_none(data.get('comment_count')) - categories = try_get(data, lambda x: x['pin_join']['visual_annotation'], list) - tags = data.get('hashtags') - - thumbnails = [] - images = data.get('images') - if isinstance(images, dict): - for thumbnail_id, thumbnail in images.items(): - if not isinstance(thumbnail, dict): - continue - thumbnail_url = url_or_none(thumbnail.get('url')) - if not thumbnail_url: - continue - thumbnails.append({ - 'url': thumbnail_url, - 'width': int_or_none(thumbnail.get('width')), - 'height': int_or_none(thumbnail.get('height')), - }) - return { 'id': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'timestamp': timestamp, - 'thumbnails': thumbnails, - 'uploader': uploader, - 'uploader_id': uploader_id, - 'repost_count': repost_count, - 'comment_count': comment_count, - 'categories': categories, - 'tags': tags, 'formats': formats, + 'duration': duration, + 'webpage_url': f'https://www.pinterest.com/pin/{video_id}/', 'extractor_key': PinterestIE.ie_key(), + 'extractor': PinterestIE.IE_NAME, + **info, } class PinterestIE(PinterestBaseIE): _VALID_URL = r'%s/pin/(?P\d+)' % PinterestBaseIE._VALID_URL_BASE _TESTS = [{ + # formats found in data['videos'] 'url': 'https://www.pinterest.com/pin/664281013778109217/', 'md5': '6550c2af85d6d9f3fe3b88954d1577fc', 'info_dict': { 'id': '664281013778109217', 'ext': 'mp4', 'title': 'Origami', - 'description': 'md5:b9d90ddf7848e897882de9e73344f7dd', + 'description': 'md5:e29801cab7d741ea8c741bc50c8d00ab', 'duration': 57.7, 'timestamp': 1593073622, 'upload_date': '20200625', - 'uploader': 'Love origami -I am Dafei', - 'uploader_id': '586523688879454212', - 'repost_count': 50, - 'comment_count': 0, + 'repost_count': int, + 'comment_count': int, 'categories': list, 'tags': list, + 'thumbnail': r're:^https?://.*\.(?:jpg|png)$', + }, + }, { + # formats found in data['story_pin_data'] + 'url': 'https://www.pinterest.com/pin/1084663891475263837/', + 'md5': '069ac19919ab9e1e13fa60de46290b03', + 'info_dict': { + 'id': '1084663891475263837', + 'ext': 'mp4', + 'title': 'Gadget, Cool products, Amazon product, technology, Kitchen gadgets', + 'description': 'md5:d0a4b6ae996ff0c6eed83bc869598d13', + 'uploader': 'CoolCrazyGadgets', + 'uploader_id': '1084664028912989237', + 'upload_date': '20211003', + 'timestamp': 1633246654.0, + 'duration': 14.9, + 'comment_count': int, + 'repost_count': int, + 'categories': 'count:9', + 'tags': list, + 'thumbnail': r're:^https?://.*\.(?:jpg|png)$', + }, + }, { + # vimeo.com embed + 'url': 'https://www.pinterest.ca/pin/441282463481903715/', + 'info_dict': { + 'id': '111691128', + 'ext': 'mp4', + 'title': 'Tonite Let\'s All Make Love In London (1967)', + 'description': 'md5:8190f37b3926807809ec57ec21aa77b2', + 'uploader': 'Vimeo', + 'uploader_id': '473792960706651251', + 'upload_date': '20180120', + 'timestamp': 1516409040, + 'duration': 3404, + 'comment_count': int, + 'repost_count': int, + 'categories': 'count:9', + 'tags': [], + 'thumbnail': r're:^https?://.*\.(?:jpg|png)$', + 'uploader_url': 'https://vimeo.com/willardandrade', + }, + 'params': { + 'skip_download': 'm3u8', }, }, { 'url': 'https://co.pinterest.com/pin/824721750502199491/',