diff options
Diffstat (limited to 'hypervideo_dl/extractor/archiveorg.py')
-rw-r--r-- | hypervideo_dl/extractor/archiveorg.py | 437 |
1 files changed, 355 insertions, 82 deletions
diff --git a/hypervideo_dl/extractor/archiveorg.py b/hypervideo_dl/extractor/archiveorg.py index 2ab3c1b..90dda9f 100644 --- a/hypervideo_dl/extractor/archiveorg.py +++ b/hypervideo_dl/extractor/archiveorg.py @@ -1,39 +1,35 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re import json +import re +import urllib.parse + from .common import InfoExtractor -from .youtube import YoutubeIE, YoutubeBaseInfoExtractor -from ..compat import ( - compat_urllib_parse_unquote, - compat_urllib_parse_unquote_plus, - compat_HTTPError -) +from .youtube import YoutubeBaseInfoExtractor, YoutubeIE +from ..compat import compat_HTTPError, compat_urllib_parse_unquote from ..utils import ( + KNOWN_EXTENSIONS, + ExtractorError, + HEADRequest, bug_reports_message, clean_html, dict_get, extract_attributes, - ExtractorError, get_element_by_id, - HEADRequest, int_or_none, join_nonempty, - KNOWN_EXTENSIONS, + js_to_json, merge_dicts, mimetype2ext, orderedSet, parse_duration, parse_qs, - str_to_int, str_or_none, + str_to_int, traverse_obj, try_get, unified_strdate, unified_timestamp, + url_or_none, urlhandle_detect_ext, - url_or_none ) @@ -54,6 +50,11 @@ class ArchiveOrgIE(InfoExtractor): 'upload_date': '20100315', 'creator': 'SRI International', 'uploader': 'laura@archive.org', + 'thumbnail': r're:https://archive\.org/download/.*\.jpg', + 'release_year': 1968, + 'display_id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.cdr', + 'track': 'XD300-23 68HighlightsAResearchCntAugHumanIntellect', + }, }, { 'url': 'https://archive.org/details/Cops1922', @@ -62,33 +63,43 @@ class ArchiveOrgIE(InfoExtractor): 'id': 'Cops1922', 'ext': 'mp4', 'title': 'Buster Keaton\'s "Cops" (1922)', - 'description': 'md5:43a603fd6c5b4b90d12a96b921212b9c', + 'description': 'md5:cd6f9910c35aedd5fc237dbc3957e2ca', 'uploader': 'yorkmba99@hotmail.com', 'timestamp': 1387699629, 'upload_date': '20131222', + 'display_id': 'Cops-v2.mp4', + 'thumbnail': r're:https://archive\.org/download/.*\.jpg', + 'duration': 1091.96, }, }, { 'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect', 'only_matching': True, }, { 'url': 'https://archive.org/details/Election_Ads', - 'md5': '284180e857160cf866358700bab668a3', + 'md5': 'eec5cddebd4793c6a653b69c3b11f2e6', 'info_dict': { 'id': 'Election_Ads/Commercial-JFK1960ElectionAdCampaignJingle.mpg', 'title': 'Commercial-JFK1960ElectionAdCampaignJingle.mpg', - 'ext': 'mp4', + 'ext': 'mpg', + 'thumbnail': r're:https://archive\.org/download/.*\.jpg', + 'duration': 59.77, + 'display_id': 'Commercial-JFK1960ElectionAdCampaignJingle.mpg', }, }, { 'url': 'https://archive.org/details/Election_Ads/Commercial-Nixon1960ElectionAdToughonDefense.mpg', - 'md5': '7915213ef02559b5501fe630e1a53f59', + 'md5': 'ea1eed8234e7d4165f38c8c769edef38', 'info_dict': { 'id': 'Election_Ads/Commercial-Nixon1960ElectionAdToughonDefense.mpg', 'title': 'Commercial-Nixon1960ElectionAdToughonDefense.mpg', - 'ext': 'mp4', + 'ext': 'mpg', 'timestamp': 1205588045, 'uploader': 'mikedavisstripmaster@yahoo.com', 'description': '1960 Presidential Campaign Election Commercials John F Kennedy, Richard M Nixon', 'upload_date': '20080315', + 'display_id': 'Commercial-Nixon1960ElectionAdToughonDefense.mpg', + 'duration': 59.51, + 'license': 'http://creativecommons.org/licenses/publicdomain/', + 'thumbnail': r're:https://archive\.org/download/.*\.jpg', }, }, { 'url': 'https://archive.org/details/gd1977-05-08.shure57.stevenson.29303.flac16', @@ -97,6 +108,12 @@ class ArchiveOrgIE(InfoExtractor): 'id': 'gd1977-05-08.shure57.stevenson.29303.flac16/gd1977-05-08d01t01.flac', 'title': 'Turning', 'ext': 'flac', + 'track': 'Turning', + 'creator': 'Grateful Dead', + 'display_id': 'gd1977-05-08d01t01.flac', + 'track_number': 1, + 'album': '1977-05-08 - Barton Hall - Cornell University', + 'duration': 39.8, }, }, { 'url': 'https://archive.org/details/gd1977-05-08.shure57.stevenson.29303.flac16/gd1977-05-08d01t07.flac', @@ -107,11 +124,20 @@ class ArchiveOrgIE(InfoExtractor): 'ext': 'flac', 'timestamp': 1205895624, 'uploader': 'mvernon54@yahoo.com', - 'description': 'md5:6a31f1996db0aa0fc9da6d6e708a1bb0', + 'description': 'md5:6c921464414814720c6593810a5c7e3d', 'upload_date': '20080319', 'location': 'Barton Hall - Cornell University', + 'duration': 438.68, + 'track': 'Deal', + 'creator': 'Grateful Dead', + 'album': '1977-05-08 - Barton Hall - Cornell University', + 'release_date': '19770508', + 'display_id': 'gd1977-05-08d01t07.flac', + 'release_year': 1977, + 'track_number': 7, }, }, { + # FIXME: give a better error message than just IndexError when all available formats are restricted 'url': 'https://archive.org/details/lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik', 'md5': '7cb019baa9b332e82ea7c10403acd180', 'info_dict': { @@ -119,6 +145,7 @@ class ArchiveOrgIE(InfoExtractor): 'title': 'Bells Of Rostov', 'ext': 'mp3', }, + 'skip': 'restricted' }, { 'url': 'https://archive.org/details/lp_the-music-of-russia_various-artists-a-askaryan-alexander-melik/disc1/02.02.+Song+And+Chorus+In+The+Polovetsian+Camp+From+%22Prince+Igor%22+(Act+2%2C+Scene+1).mp3', 'md5': '1d0aabe03edca83ca58d9ed3b493a3c3', @@ -131,6 +158,52 @@ class ArchiveOrgIE(InfoExtractor): 'description': 'md5:012b2d668ae753be36896f343d12a236', 'upload_date': '20190928', }, + 'skip': 'restricted' + }, { + # Original formats are private + 'url': 'https://archive.org/details/irelandthemakingofarepublic', + 'info_dict': { + 'id': 'irelandthemakingofarepublic', + 'title': 'Ireland: The Making of a Republic', + 'upload_date': '20160610', + 'description': 'md5:f70956a156645a658a0dc9513d9e78b7', + 'uploader': 'dimitrios@archive.org', + 'creator': ['British Broadcasting Corporation', 'Time-Life Films'], + 'timestamp': 1465594947, + }, + 'playlist': [ + { + 'md5': '0b211261b26590d49df968f71b90690d', + 'info_dict': { + 'id': 'irelandthemakingofarepublic/irelandthemakingofarepublicreel1_01.mov', + 'ext': 'mp4', + 'title': 'irelandthemakingofarepublicreel1_01.mov', + 'duration': 130.46, + 'thumbnail': 'https://archive.org/download/irelandthemakingofarepublic/irelandthemakingofarepublic.thumbs/irelandthemakingofarepublicreel1_01_000117.jpg', + 'display_id': 'irelandthemakingofarepublicreel1_01.mov', + }, + }, { + 'md5': '67335ee3b23a0da930841981c1e79b02', + 'info_dict': { + 'id': 'irelandthemakingofarepublic/irelandthemakingofarepublicreel1_02.mov', + 'ext': 'mp4', + 'duration': 1395.13, + 'title': 'irelandthemakingofarepublicreel1_02.mov', + 'display_id': 'irelandthemakingofarepublicreel1_02.mov', + 'thumbnail': 'https://archive.org/download/irelandthemakingofarepublic/irelandthemakingofarepublic.thumbs/irelandthemakingofarepublicreel1_02_001374.jpg', + }, + }, { + 'md5': 'e470e86787893603f4a341a16c281eb5', + 'info_dict': { + 'id': 'irelandthemakingofarepublic/irelandthemakingofarepublicreel2.mov', + 'ext': 'mp4', + 'duration': 1602.67, + 'title': 'irelandthemakingofarepublicreel2.mov', + 'thumbnail': 'https://archive.org/download/irelandthemakingofarepublic/irelandthemakingofarepublic.thumbs/irelandthemakingofarepublicreel2_001554.jpg', + 'display_id': 'irelandthemakingofarepublicreel2.mov', + }, + } + ] }] @staticmethod @@ -146,7 +219,7 @@ class ArchiveOrgIE(InfoExtractor): return json.loads(extract_attributes(element)['value']) def _real_extract(self, url): - video_id = compat_urllib_parse_unquote_plus(self._match_id(url)) + video_id = urllib.parse.unquote_plus(self._match_id(url)) identifier, entry_id = (video_id.split('/', 1) + [None])[:2] # Archive.org metadata API doesn't clearly demarcate playlist entries @@ -221,17 +294,25 @@ class ArchiveOrgIE(InfoExtractor): 'filesize': int_or_none(f.get('size'))}) extension = (f['name'].rsplit('.', 1) + [None])[1] - if extension in KNOWN_EXTENSIONS: + + # We don't want to skip private formats if the user has access to them, + # however without access to an account with such privileges we can't implement/test this. + # For now to be safe, we will only skip them if there is no user logged in. + is_logged_in = bool(self._get_cookies('https://archive.org').get('logged-in-sig')) + if extension in KNOWN_EXTENSIONS and (not f.get('private') or is_logged_in): entry['formats'].append({ 'url': 'https://archive.org/download/' + identifier + '/' + f['name'], 'format': f.get('format'), 'width': int_or_none(f.get('width')), 'height': int_or_none(f.get('height')), 'filesize': int_or_none(f.get('size')), - 'protocol': 'https'}) + 'protocol': 'https', + 'source_preference': 0 if f.get('source') == 'original' else -1, + 'format_note': f.get('source') + }) for entry in entries.values(): - self._sort_formats(entry['formats']) + entry['_format_sort_fields'] = ('source', ) if len(entries) == 1: # If there's only one item, use it as the main info dict @@ -287,7 +368,9 @@ class YoutubeWebArchiveIE(InfoExtractor): 'channel_id': 'UCukCyHaD-bK3in_pKpfH9Eg', 'duration': 32, 'uploader_id': 'Zeurel', - 'uploader_url': 'http://www.youtube.com/user/Zeurel' + 'uploader_url': 'https://www.youtube.com/user/Zeurel', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'channel_url': 'https://www.youtube.com/channel/UCukCyHaD-bK3in_pKpfH9Eg', } }, { # Internal link @@ -302,7 +385,9 @@ class YoutubeWebArchiveIE(InfoExtractor): 'channel_id': 'UCHnyfMqiRRG1u-2MsSQLbXA', 'duration': 771, 'uploader_id': '1veritasium', - 'uploader_url': 'http://www.youtube.com/user/1veritasium' + 'uploader_url': 'https://www.youtube.com/user/1veritasium', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'channel_url': 'https://www.youtube.com/channel/UCHnyfMqiRRG1u-2MsSQLbXA', } }, { # Video from 2012, webm format itag 45. Newest capture is deleted video, with an invalid description. @@ -316,7 +401,9 @@ class YoutubeWebArchiveIE(InfoExtractor): 'duration': 398, 'description': 'md5:ff4de6a7980cb65d951c2f6966a4f2f3', 'uploader_id': 'machinima', - 'uploader_url': 'http://www.youtube.com/user/machinima' + 'uploader_url': 'https://www.youtube.com/user/machinima', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'uploader': 'machinima' } }, { # FLV video. Video file URL does not provide itag information @@ -330,7 +417,10 @@ class YoutubeWebArchiveIE(InfoExtractor): 'duration': 19, 'description': 'md5:10436b12e07ac43ff8df65287a56efb4', 'uploader_id': 'jawed', - 'uploader_url': 'http://www.youtube.com/user/jawed' + 'uploader_url': 'https://www.youtube.com/user/jawed', + 'channel_url': 'https://www.youtube.com/channel/UC4QobU6STFB0P71PMvOGN5A', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'uploader': 'jawed', } }, { 'url': 'https://web.archive.org/web/20110712231407/http://www.youtube.com/watch?v=lTx3G6h2xyA', @@ -344,7 +434,9 @@ class YoutubeWebArchiveIE(InfoExtractor): 'duration': 204, 'description': 'md5:f7535343b6eda34a314eff8b85444680', 'uploader_id': 'itsmadeon', - 'uploader_url': 'http://www.youtube.com/user/itsmadeon' + 'uploader_url': 'https://www.youtube.com/user/itsmadeon', + 'channel_url': 'https://www.youtube.com/channel/UCqMDNf3Pn5L7pcNkuSEeO3w', + 'thumbnail': r're:https?://.*\.(jpg|webp)', } }, { # First capture is of dead video, second is the oldest from CDX response. @@ -355,10 +447,13 @@ class YoutubeWebArchiveIE(InfoExtractor): 'title': 'Fake Teen Doctor Strikes AGAIN! - Weekly Weird News', 'upload_date': '20160218', 'channel_id': 'UCdIaNUarhzLSXGoItz7BHVA', - 'duration': 1236, + 'duration': 1235, 'description': 'md5:21032bae736421e89c2edf36d1936947', 'uploader_id': 'MachinimaETC', - 'uploader_url': 'http://www.youtube.com/user/MachinimaETC' + 'uploader_url': 'https://www.youtube.com/user/MachinimaETC', + 'channel_url': 'https://www.youtube.com/channel/UCdIaNUarhzLSXGoItz7BHVA', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'uploader': 'ETC News', } }, { # First capture of dead video, capture date in link links to dead capture. @@ -369,10 +464,13 @@ class YoutubeWebArchiveIE(InfoExtractor): 'title': 'WTF: Video Games Still Launch BROKEN?! - T.U.G.S.', 'upload_date': '20160219', 'channel_id': 'UCdIaNUarhzLSXGoItz7BHVA', - 'duration': 798, + 'duration': 797, 'description': 'md5:a1dbf12d9a3bd7cb4c5e33b27d77ffe7', 'uploader_id': 'MachinimaETC', - 'uploader_url': 'http://www.youtube.com/user/MachinimaETC' + 'uploader_url': 'https://www.youtube.com/user/MachinimaETC', + 'channel_url': 'https://www.youtube.com/channel/UCdIaNUarhzLSXGoItz7BHVA', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'uploader': 'ETC News', }, 'expected_warnings': [ r'unable to download capture webpage \(it may not be archived\)' @@ -392,12 +490,11 @@ class YoutubeWebArchiveIE(InfoExtractor): 'title': 'It\'s Bootleg AirPods Time.', 'upload_date': '20211021', 'channel_id': 'UC7Jwj9fkrf1adN4fMmTkpug', - 'channel_url': 'http://www.youtube.com/channel/UC7Jwj9fkrf1adN4fMmTkpug', + 'channel_url': 'https://www.youtube.com/channel/UC7Jwj9fkrf1adN4fMmTkpug', 'duration': 810, 'description': 'md5:7b567f898d8237b256f36c1a07d6d7bc', + 'thumbnail': r're:https?://.*\.(jpg|webp)', 'uploader': 'DankPods', - 'uploader_id': 'UC7Jwj9fkrf1adN4fMmTkpug', - 'uploader_url': 'http://www.youtube.com/channel/UC7Jwj9fkrf1adN4fMmTkpug' } }, { # player response contains '};' See: https://github.com/ytdl-org/youtube-dl/issues/27093 @@ -408,12 +505,135 @@ class YoutubeWebArchiveIE(InfoExtractor): 'title': 'bitch lasagna', 'upload_date': '20181005', 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw', - 'channel_url': 'http://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw', + 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw', 'duration': 135, 'description': 'md5:2dbe4051feeff2dab5f41f82bb6d11d0', 'uploader': 'PewDiePie', 'uploader_id': 'PewDiePie', - 'uploader_url': 'http://www.youtube.com/user/PewDiePie' + 'uploader_url': 'https://www.youtube.com/user/PewDiePie', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + } + }, { + # ~June 2010 Capture. swfconfig + 'url': 'https://web.archive.org/web/0/https://www.youtube.com/watch?v=8XeW5ilk-9Y', + 'info_dict': { + 'id': '8XeW5ilk-9Y', + 'ext': 'flv', + 'title': 'Story of Stuff, The Critique Part 4 of 4', + 'duration': 541, + 'description': 'md5:28157da06f2c5e94c97f7f3072509972', + 'uploader': 'HowTheWorldWorks', + 'uploader_id': 'HowTheWorldWorks', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'uploader_url': 'https://www.youtube.com/user/HowTheWorldWorks', + 'upload_date': '20090520', + } + }, { + # Jan 2011: watch-video-date/eow-date surrounded by whitespace + 'url': 'https://web.archive.org/web/20110126141719/http://www.youtube.com/watch?v=Q_yjX80U7Yc', + 'info_dict': { + 'id': 'Q_yjX80U7Yc', + 'ext': 'flv', + 'title': 'Spray Paint Art by Clay Butler: Purple Fantasy Forest', + 'uploader_id': 'claybutlermusic', + 'description': 'md5:4595264559e3d0a0ceb3f011f6334543', + 'upload_date': '20090803', + 'uploader': 'claybutlermusic', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'duration': 132, + 'uploader_url': 'https://www.youtube.com/user/claybutlermusic', + } + }, { + # ~May 2009 swfArgs. ytcfg is spread out over various vars + 'url': 'https://web.archive.org/web/0/https://www.youtube.com/watch?v=c5uJgG05xUY', + 'info_dict': { + 'id': 'c5uJgG05xUY', + 'ext': 'webm', + 'title': 'Story of Stuff, The Critique Part 1 of 4', + 'uploader_id': 'HowTheWorldWorks', + 'uploader': 'HowTheWorldWorks', + 'uploader_url': 'https://www.youtube.com/user/HowTheWorldWorks', + 'upload_date': '20090513', + 'description': 'md5:4ca77d79538064e41e4cc464e93f44f0', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'duration': 754, + } + }, { + # ~June 2012. Upload date is in another lang so cannot extract. + 'url': 'https://web.archive.org/web/20120607174520/http://www.youtube.com/watch?v=xWTLLl-dQaA', + 'info_dict': { + 'id': 'xWTLLl-dQaA', + 'ext': 'mp4', + 'title': 'Black Nerd eHarmony Video Bio Parody (SPOOF)', + 'uploader_url': 'https://www.youtube.com/user/BlackNerdComedy', + 'description': 'md5:e25f0133aaf9e6793fb81c18021d193e', + 'uploader_id': 'BlackNerdComedy', + 'uploader': 'BlackNerdComedy', + 'duration': 182, + 'thumbnail': r're:https?://.*\.(jpg|webp)', + } + }, { + # ~July 2013 + 'url': 'https://web.archive.org/web/*/https://www.youtube.com/watch?v=9eO1aasHyTM', + 'info_dict': { + 'id': '9eO1aasHyTM', + 'ext': 'mp4', + 'title': 'Polar-oid', + 'description': 'Cameras and bears are dangerous!', + 'uploader_url': 'https://www.youtube.com/user/punkybird', + 'uploader_id': 'punkybird', + 'duration': 202, + 'channel_id': 'UC62R2cBezNBOqxSerfb1nMQ', + 'channel_url': 'https://www.youtube.com/channel/UC62R2cBezNBOqxSerfb1nMQ', + 'upload_date': '20060428', + 'uploader': 'punkybird', + } + }, { + # April 2020: Player response in player config + 'url': 'https://web.archive.org/web/20200416034815/https://www.youtube.com/watch?v=Cf7vS8jc7dY&gl=US&hl=en', + 'info_dict': { + 'id': 'Cf7vS8jc7dY', + 'ext': 'mp4', + 'title': 'A Dramatic Pool Story (by Jamie Spicer-Lewis) - Game Grumps Animated', + 'duration': 64, + 'upload_date': '20200408', + 'uploader_id': 'GameGrumps', + 'uploader': 'GameGrumps', + 'channel_url': 'https://www.youtube.com/channel/UC9CuvdOVfMPvKCiwdGKL3cQ', + 'channel_id': 'UC9CuvdOVfMPvKCiwdGKL3cQ', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'description': 'md5:c625bb3c02c4f5fb4205971e468fa341', + 'uploader_url': 'https://www.youtube.com/user/GameGrumps', + } + }, { + # watch7-user-header with yt-user-info + 'url': 'ytarchive:kbh4T_b4Ixw:20160307085057', + 'info_dict': { + 'id': 'kbh4T_b4Ixw', + 'ext': 'mp4', + 'title': 'Shovel Knight OST - Strike the Earth! Plains of Passage 16 bit SNES style remake / remix', + 'channel_url': 'https://www.youtube.com/channel/UCnTaGvsHmMy792DWeT6HbGA', + 'uploader': 'Nelward music', + 'duration': 213, + 'description': 'md5:804b4a9ce37b050a5fefdbb23aeba54d', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'upload_date': '20150503', + 'channel_id': 'UCnTaGvsHmMy792DWeT6HbGA', + } + }, { + # April 2012 + 'url': 'https://web.archive.org/web/0/https://www.youtube.com/watch?v=SOm7mPoPskU', + 'info_dict': { + 'id': 'SOm7mPoPskU', + 'ext': 'mp4', + 'title': 'Boyfriend - Justin Bieber Parody', + 'uploader_url': 'https://www.youtube.com/user/thecomputernerd01', + 'uploader': 'thecomputernerd01', + 'thumbnail': r're:https?://.*\.(jpg|webp)', + 'description': 'md5:dd7fa635519c2a5b4d566beaecad7491', + 'duration': 200, + 'upload_date': '20120407', + 'uploader_id': 'thecomputernerd01', } }, { 'url': 'https://web.archive.org/web/http://www.youtube.com/watch?v=kH-G_aIBlFw', @@ -445,9 +665,11 @@ class YoutubeWebArchiveIE(InfoExtractor): 'only_matching': True }, ] - _YT_INITIAL_DATA_RE = r'(?:(?:(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_DATA_RE - _YT_INITIAL_PLAYER_RESPONSE_RE = r'(?:(?:(?:window\s*\[\s*["\']ytInitialPlayerResponse["\']\s*\]|ytInitialPlayerResponse)\s*=[(\s]*({.+?})[)\s]*;)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_PLAYER_RESPONSE_RE - _YT_INITIAL_BOUNDARY_RE = r'(?:(?:var\s+meta|</script|\n)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_BOUNDARY_RE + _YT_INITIAL_DATA_RE = YoutubeBaseInfoExtractor._YT_INITIAL_DATA_RE + _YT_INITIAL_PLAYER_RESPONSE_RE = fr'''(?x: + (?:window\s*\[\s*["\']ytInitialPlayerResponse["\']\s*\]|ytInitialPlayerResponse)\s*=[(\s]*| + {YoutubeBaseInfoExtractor._YT_INITIAL_PLAYER_RESPONSE_RE} + )''' _YT_DEFAULT_THUMB_SERVERS = ['i.ytimg.com'] # thumbnails most likely archived on these servers _YT_ALL_THUMB_SERVERS = orderedSet( @@ -477,11 +699,6 @@ class YoutubeWebArchiveIE(InfoExtractor): elif not isinstance(res, list) or len(res) != 0: self.report_warning('Error while parsing CDX API response' + bug_reports_message()) - def _extract_yt_initial_variable(self, webpage, regex, video_id, name): - return self._parse_json(self._search_regex( - (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE), - regex), webpage, name, default='{}'), video_id, fatal=False) - def _extract_webpage_title(self, webpage): page_title = self._html_extract_title(webpage, default='') # YouTube video pages appear to always have either 'YouTube -' as prefix or '- YouTube' as suffix. @@ -491,10 +708,32 @@ class YoutubeWebArchiveIE(InfoExtractor): def _extract_metadata(self, video_id, webpage): search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None)) if webpage else (lambda x: None)) - player_response = self._extract_yt_initial_variable( - webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, video_id, 'initial player response') or {} - initial_data = self._extract_yt_initial_variable( - webpage, self._YT_INITIAL_DATA_RE, video_id, 'initial player response') or {} + player_response = self._search_json( + self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', + video_id, default={}) + initial_data = self._search_json( + self._YT_INITIAL_DATA_RE, webpage, 'initial data', video_id, default={}) + + ytcfg = {} + for j in re.findall(r'yt\.setConfig\(\s*(?P<json>{\s*(?s:.+?)\s*})\s*\);', webpage): # ~June 2010 + ytcfg.update(self._parse_json(j, video_id, fatal=False, ignore_extra=True, transform_source=js_to_json, errnote='') or {}) + + # XXX: this also may contain a 'ptchn' key + player_config = ( + self._search_json( + r'(?:yt\.playerConfig|ytplayer\.config|swfConfig)\s*=', + webpage, 'player config', video_id, default=None) + or ytcfg.get('PLAYER_CONFIG') or {}) + + # XXX: this may also contain a 'creator' key. + swf_args = self._search_json(r'swfArgs\s*=', webpage, 'swf config', video_id, default={}) + if swf_args and not traverse_obj(player_config, ('args',)): + player_config['args'] = swf_args + + if not player_response: + # April 2020 + player_response = self._parse_json( + traverse_obj(player_config, ('args', 'player_response')) or '{}', video_id, fatal=False) initial_data_video = traverse_obj( initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'videoPrimaryInfoRenderer'), @@ -510,21 +749,64 @@ class YoutubeWebArchiveIE(InfoExtractor): video_details.get('title') or YoutubeBaseInfoExtractor._get_text(microformats, 'title') or YoutubeBaseInfoExtractor._get_text(initial_data_video, 'title') + or traverse_obj(player_config, ('args', 'title')) or self._extract_webpage_title(webpage) or search_meta(['og:title', 'twitter:title', 'title'])) + def id_from_url(url, type_): + return self._search_regex( + rf'(?:{type_})/([^/#&?]+)', url or '', f'{type_} id', default=None) + + # XXX: would the get_elements_by_... functions be better suited here? + _CHANNEL_URL_HREF_RE = r'href="[^"]*(?P<url>https?://www\.youtube\.com/(?:user|channel)/[^"]+)"' + uploader_or_channel_url = self._search_regex( + [fr'<(?:link\s*itemprop=\"url\"|a\s*id=\"watch-username\").*?\b{_CHANNEL_URL_HREF_RE}>', # @fd05024 + fr'<div\s*id=\"(?:watch-channel-stats|watch-headline-user-info)\"[^>]*>\s*<a[^>]*\b{_CHANNEL_URL_HREF_RE}'], # ~ May 2009, ~June 2012 + webpage, 'uploader or channel url', default=None) + + owner_profile_url = url_or_none(microformats.get('ownerProfileUrl')) # @a6211d2 + + # Uploader refers to the /user/ id ONLY + uploader_id = ( + id_from_url(owner_profile_url, 'user') + or id_from_url(uploader_or_channel_url, 'user') + or ytcfg.get('VIDEO_USERNAME')) + uploader_url = f'https://www.youtube.com/user/{uploader_id}' if uploader_id else None + + # XXX: do we want to differentiate uploader and channel? + uploader = ( + self._search_regex( + [r'<a\s*id="watch-username"[^>]*>\s*<strong>([^<]+)</strong>', # June 2010 + r'var\s*watchUsername\s*=\s*\'(.+?)\';', # ~May 2009 + r'<div\s*\bid=\"watch-channel-stats"[^>]*>\s*<a[^>]*>\s*(.+?)\s*</a', # ~May 2009 + r'<a\s*id="watch-userbanner"[^>]*title="\s*(.+?)\s*"'], # ~June 2012 + webpage, 'uploader', default=None) + or self._html_search_regex( + [r'(?s)<div\s*class="yt-user-info".*?<a[^>]*[^>]*>\s*(.*?)\s*</a', # March 2016 + r'(?s)<a[^>]*yt-user-name[^>]*>\s*(.*?)\s*</a'], # july 2013 + get_element_by_id('watch7-user-header', webpage), 'uploader', default=None) + or self._html_search_regex( + r'<button\s*href="/user/[^>]*>\s*<span[^>]*>\s*(.+?)\s*<', # April 2012 + get_element_by_id('watch-headline-user-info', webpage), 'uploader', default=None) + or traverse_obj(player_config, ('args', 'creator')) + or video_details.get('author')) + channel_id = str_or_none( video_details.get('channelId') or microformats.get('externalChannelId') or search_meta('channelId') or self._search_regex( r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1', # @b45a9e6 - webpage, 'channel id', default=None, group='id')) - channel_url = f'http://www.youtube.com/channel/{channel_id}' if channel_id else None + webpage, 'channel id', default=None, group='id') + or id_from_url(owner_profile_url, 'channel') + or id_from_url(uploader_or_channel_url, 'channel') + or traverse_obj(player_config, ('args', 'ucid'))) + channel_url = f'https://www.youtube.com/channel/{channel_id}' if channel_id else None duration = int_or_none( video_details.get('lengthSeconds') or microformats.get('lengthSeconds') + or traverse_obj(player_config, ('args', ('length_seconds', 'l')), get_all=False) or parse_duration(search_meta('duration'))) description = ( video_details.get('shortDescription') @@ -532,26 +814,13 @@ class YoutubeWebArchiveIE(InfoExtractor): or clean_html(get_element_by_id('eow-description', webpage)) # @9e6dd23 or search_meta(['description', 'og:description', 'twitter:description'])) - uploader = video_details.get('author') - - # Uploader ID and URL - uploader_mobj = re.search( - r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">', # @fd05024 - webpage) - if uploader_mobj is not None: - uploader_id, uploader_url = uploader_mobj.group('uploader_id'), uploader_mobj.group('uploader_url') - else: - # @a6211d2 - uploader_url = url_or_none(microformats.get('ownerProfileUrl')) - uploader_id = self._search_regex( - r'(?:user|channel)/([^/]+)', uploader_url or '', 'uploader id', default=None) - upload_date = unified_strdate( dict_get(microformats, ('uploadDate', 'publishDate')) or search_meta(['uploadDate', 'datePublished']) or self._search_regex( - [r'(?s)id="eow-date.*?>(.*?)</span>', - r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'], # @7998520 + [r'(?s)id="eow-date.*?>\s*(.*?)\s*</span>', + r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']', # @7998520 + r'class\s*=\s*"(?:watch-video-date|watch-video-added post-date)"[^>]*>\s*([^<]+?)\s*<'], # ~June 2010, ~Jan 2009 (respectively) webpage, 'upload date', default=None)) return { @@ -597,7 +866,7 @@ class YoutubeWebArchiveIE(InfoExtractor): response = self._call_cdx_api( video_id, f'https://www.youtube.com/watch?v={video_id}', filters=['mimetype:text/html'], collapse=['timestamp:6', 'digest'], query={'matchType': 'prefix'}) or [] - all_captures = sorted([int_or_none(r['timestamp']) for r in response if int_or_none(r['timestamp']) is not None]) + all_captures = sorted(int_or_none(r['timestamp']) for r in response if int_or_none(r['timestamp']) is not None) # Prefer the new polymer UI captures as we support extracting more metadata from them # WBM captures seem to all switch to this layout ~July 2020 @@ -620,18 +889,22 @@ class YoutubeWebArchiveIE(InfoExtractor): url_date = url_date or url_date_2 urlh = None - try: - urlh = self._request_webpage( - HEADRequest('https://web.archive.org/web/2oe_/http://wayback-fakeurl.archive.org/yt/%s' % video_id), - video_id, note='Fetching archived video file url', expected_status=True) - except ExtractorError as e: - # HTTP Error 404 is expected if the video is not saved. - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: - self.raise_no_formats( - 'The requested video is not archived, indexed, or there is an issue with web.archive.org', - expected=True) - else: - raise + retry_manager = self.RetryManager(fatal=False) + for retry in retry_manager: + try: + urlh = self._request_webpage( + HEADRequest('https://web.archive.org/web/2oe_/http://wayback-fakeurl.archive.org/yt/%s' % video_id), + video_id, note='Fetching archived video file url', expected_status=True) + except ExtractorError as e: + # HTTP Error 404 is expected if the video is not saved. + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + self.raise_no_formats( + 'The requested video is not archived, indexed, or there is an issue with web.archive.org (try again later)', expected=True) + else: + retry.error = e + + if retry_manager.error: + self.raise_no_formats(retry_manager.error, expected=True, video_id=video_id) capture_dates = self._get_capture_dates(video_id, int_or_none(url_date)) self.write_debug('Captures to try: ' + join_nonempty(*capture_dates, delim=', ')) |