Upstream/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-12 14:26:49 +01:00
Code Issues Packages Projects Releases Wiki Activity

[web.archive:youtube] Add ytarchive: prefix

Browse Source 
and misc cleanup
This commit is contained in:
pukkandan 2022-01-23 18:32:15 +05:30
parent e612f66c7c
commit 1f13021eca
No known key found for this signature in database
GPG Key ID: 0F00D95A001F4698
1 changed files with 37 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

64
yt_dlp/extractor/archiveorg.py
View File

@ -19,6 +19,7 @@
get_element_by_id, get_element_by_id,
HEADRequest, HEADRequest,
int_or_none, int_or_none,
join_nonempty,
KNOWN_EXTENSIONS, KNOWN_EXTENSIONS,
merge_dicts, merge_dicts,
mimetype2ext, mimetype2ext,
@ -64,7 +65,7 @@ class ArchiveOrgIE(InfoExtractor):
'description': 'md5:43a603fd6c5b4b90d12a96b921212b9c', 'description': 'md5:43a603fd6c5b4b90d12a96b921212b9c',
'uploader': 'yorkmba99@hotmail.com', 'uploader': 'yorkmba99@hotmail.com',
'timestamp': 1387699629, 'timestamp': 1387699629,
'upload_date': "20131222", 'upload_date': '20131222',
}, },
}, { }, {
'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect', 'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
@ -150,8 +151,7 @@ def _real_extract(self, url):
# Archive.org metadata API doesn't clearly demarcate playlist entries # Archive.org metadata API doesn't clearly demarcate playlist entries
# or subtitle tracks, so we get them from the embeddable player. # or subtitle tracks, so we get them from the embeddable player.
embed_page = self._download_webpage( embed_page = self._download_webpage(f'https://archive.org/embed/{identifier}', identifier)
'https://archive.org/embed/' + identifier, identifier)
playlist = self._playlist_data(embed_page) playlist = self._playlist_data(embed_page)
entries = {} entries = {}
@ -166,17 +166,17 @@ def _real_extract(self, url):
'thumbnails': [], 'thumbnails': [],
'artist': p.get('artist'), 'artist': p.get('artist'),
'track': p.get('title'), 'track': p.get('title'),
'subtitles': {}} 'subtitles': {},
}
for track in p.get('tracks', []): for track in p.get('tracks', []):
if track['kind'] != 'subtitles': if track['kind'] != 'subtitles':
continue continue
entries[p['orig']][track['label']] = { entries[p['orig']][track['label']] = {
'url': 'https://archive.org/' + track['file'].lstrip('/')} 'url': 'https://archive.org/' + track['file'].lstrip('/')
}
metadata = self._download_json( metadata = self._download_json('http://archive.org/metadata/' + identifier, identifier)
'http://archive.org/metadata/' + identifier, identifier)
m = metadata['metadata'] m = metadata['metadata']
identifier = m['identifier'] identifier = m['identifier']
@ -189,7 +189,7 @@ def _real_extract(self, url):
'license': m.get('licenseurl'), 'license': m.get('licenseurl'),
'release_date': unified_strdate(m.get('date')), 'release_date': unified_strdate(m.get('date')),
'timestamp': unified_timestamp(dict_get(m, ['publicdate', 'addeddate'])), 'timestamp': unified_timestamp(dict_get(m, ['publicdate', 'addeddate'])),
'webpage_url': 'https://archive.org/details/' + identifier, 'webpage_url': f'https://archive.org/details/{identifier}',
'location': m.get('venue'), 'location': m.get('venue'),
'release_year': int_or_none(m.get('year'))} 'release_year': int_or_none(m.get('year'))}
@ -230,13 +230,12 @@ def _real_extract(self, url):
'filesize': int_or_none(f.get('size')), 'filesize': int_or_none(f.get('size')),
'protocol': 'https'}) 'protocol': 'https'})
# Sort available formats by filesize
for entry in entries.values(): for entry in entries.values():
entry['formats'] = list(sorted(entry['formats'], key=lambda x: x.get('filesize', -1))) self._sort_formats(entry['formats'])
if len(entries) == 1: if len(entries) == 1:
# If there's only one item, use it as the main info dict # If there's only one item, use it as the main info dict
only_video = entries[list(entries.keys())[0]] only_video = next(iter(entries.values()))
if entry_id: if entry_id:
info = merge_dicts(only_video, info) info = merge_dicts(only_video, info)
else: else:
@ -261,19 +260,19 @@ def _real_extract(self, url):
class YoutubeWebArchiveIE(InfoExtractor): class YoutubeWebArchiveIE(InfoExtractor):
IE_NAME = 'web.archive:youtube' IE_NAME = 'web.archive:youtube'
IE_DESC = 'web.archive.org saved youtube videos' IE_DESC = 'web.archive.org saved youtube videos, "ytarchive:" prefix'
_VALID_URL = r"""(?x)^ _VALID_URL = r'''(?x)(?:(?P<prefix>ytarchive:)|
(?:https?://)?web\.archive\.org/ (?:https?://)?web\.archive\.org/
(?:web/)? (?:web/)?(?:(?P<date>[0-9]{14})?[0-9A-Za-z_*]*/)? # /web and the version index is optional
(?:(?P<date>[0-9]{14})?[0-9A-Za-z_*]*/)? # /web and the version index is optional (?:https?(?::|%3[Aa])//)?(?:
(?:https?(?::|%3[Aa])//)?
(?:
(?:\w+\.)?youtube\.com(?::(?:80|443))?/watch(?:\.php)?(?:\?|%3[fF])(?:[^\#]+(?:&|%26))?v(?:=|%3[dD]) # Youtube URL (?:\w+\.)?youtube\.com(?::(?:80|443))?/watch(?:\.php)?(?:\?|%3[fF])(?:[^\#]+(?:&|%26))?v(?:=|%3[dD]) # Youtube URL
|(?:wayback-fakeurl\.archive\.org/yt/) # Or the internal fake url |(?:wayback-fakeurl\.archive\.org/yt/) # Or the internal fake url
) )
(?P<id>[0-9A-Za-z_-]{11})(?:%26|\#|&|$) )(?P<id>[0-9A-Za-z_-]{11})
""" (?(prefix)
(?::(?P<date2>[0-9]{14}))?$|
(?:%26|[#&]|$)
)'''
_TESTS = [ _TESTS = [
{ {
@ -438,7 +437,13 @@ class YoutubeWebArchiveIE(InfoExtractor):
}, { }, {
'url': 'https://web.archive.org/http://www.youtube.com:80/watch?v=-05VVye-ffg', 'url': 'https://web.archive.org/http://www.youtube.com:80/watch?v=-05VVye-ffg',
'only_matching': True 'only_matching': True
} }, {
'url': 'ytarchive:BaW_jenozKc:20050214000000',
'only_matching': True
},{
'url': 'ytarchive:BaW_jenozKc',
'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_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_PLAYER_RESPONSE_RE = r'(?:(?:(?:window\s*\[\s*["\']ytInitialPlayerResponse["\']\s*\]|ytInitialPlayerResponse)\s*=[(\s]*({.+?})[)\s]*;)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_PLAYER_RESPONSE_RE
@ -484,7 +489,6 @@ def _extract_webpage_title(self, webpage):
page_title, 'title', default='') page_title, 'title', default='')
def _extract_metadata(self, video_id, webpage): 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)) 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( player_response = self._extract_yt_initial_variable(
webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, video_id, 'initial player response') or {} webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, video_id, 'initial player response') or {}
@ -596,7 +600,7 @@ def _get_capture_dates(self, video_id, url_date):
# Prefer the new polymer UI captures as we support extracting more metadata from them # 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 # WBM captures seem to all switch to this layout ~July 2020
modern_captures = list(filter(lambda x: x >= 20200701000000, all_captures)) modern_captures = [x for x in all_captures if x >= 20200701000000]
if modern_captures: if modern_captures:
capture_dates.append(modern_captures[0]) capture_dates.append(modern_captures[0])
capture_dates.append(url_date) capture_dates.append(url_date)
@ -608,11 +612,11 @@ def _get_capture_dates(self, video_id, url_date):
# Fallbacks if any of the above fail # Fallbacks if any of the above fail
capture_dates.extend([self._OLDEST_CAPTURE_DATE, self._NEWEST_CAPTURE_DATE]) capture_dates.extend([self._OLDEST_CAPTURE_DATE, self._NEWEST_CAPTURE_DATE])
return orderedSet(capture_dates) return orderedSet(filter(None, capture_dates))
def _real_extract(self, url): def _real_extract(self, url):
video_id, url_date, url_date_2 = self._match_valid_url(url).group('id', 'date', 'date2')
url_date, video_id = self._match_valid_url(url).groups() url_date = url_date or url_date_2
urlh = None urlh = None
try: try:
@ -629,11 +633,9 @@ def _real_extract(self, url):
raise raise
capture_dates = self._get_capture_dates(video_id, int_or_none(url_date)) capture_dates = self._get_capture_dates(video_id, int_or_none(url_date))
self.write_debug('Captures to try: ' + ', '.join(str(i) for i in capture_dates if i is not None)) self.write_debug('Captures to try: ' + join_nonempty(*capture_dates, delim=', '))
info = {'id': video_id} info = {'id': video_id}
for capture in capture_dates: for capture in capture_dates:
if not capture:
continue
webpage = self._download_webpage( webpage = self._download_webpage(
(self._WAYBACK_BASE_URL + 'http://www.youtube.com/watch?v=%s') % (capture, video_id), (self._WAYBACK_BASE_URL + 'http://www.youtube.com/watch?v=%s') % (capture, video_id),
video_id=video_id, fatal=False, errnote='unable to download capture webpage (it may not be archived)', video_id=video_id, fatal=False, errnote='unable to download capture webpage (it may not be archived)',
@ -648,7 +650,7 @@ def _real_extract(self, url):
info['thumbnails'] = self._extract_thumbnails(video_id) info['thumbnails'] = self._extract_thumbnails(video_id)
if urlh: if urlh:
url = compat_urllib_parse_unquote(urlh.url) url = compat_urllib_parse_unquote(urlh.geturl())
video_file_url_qs = parse_qs(url) video_file_url_qs = parse_qs(url)
# Attempt to recover any ext & format info from playback url & response headers # Attempt to recover any ext & format info from playback url & response headers
format = {'url': url, 'filesize': int_or_none(urlh.headers.get('x-archive-orig-content-length'))} format = {'url': url, 'filesize': int_or_none(urlh.headers.get('x-archive-orig-content-length'))}