mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-27 10:31:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

[youtube:tab] Extract playlist availability (#504)

Browse source 
Authored by: colethedj
This commit is contained in:
coletdjnz 2021-07-15 14:42:30 +12:00 committed by GitHub
parent 49bd8c66d3
commit 47193e0298
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 136 additions and 88 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

146
yt_dlp/extractor/youtube.py
View file

@ -645,6 +645,28 @@ def _report_alerts(self, alerts, expected=True):
def _extract_and_report_alerts(self, data, *args, **kwargs):
return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
def _extract_badges(self, renderer: dict):
badges = set()
for badge in try_get(renderer, lambda x: x['badges'], list) or []:
label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str)
if label:
badges.add(label.lower())
return badges
@staticmethod
def _join_text_entries(runs):
text = None
for run in runs:
if not isinstance(run, dict):
continue
sub_text = try_get(run, lambda x: x['text'], compat_str)
if sub_text:
if not text:
text = sub_text
continue
text += sub_text
return text
def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
default_client='WEB'):
@ -1971,20 +1993,6 @@ def parse_time_text(time_text):
if len(time_text_split) >= 3:
return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto')
@staticmethod
def _join_text_entries(runs):
text = None
for run in runs:
if not isinstance(run, dict):
continue
sub_text = try_get(run, lambda x: x['text'], compat_str)
if sub_text:
if not text:
text = sub_text
continue
text += sub_text
return text
def _extract_comment(self, comment_renderer, parent=None):
comment_id = comment_renderer.get('commentId')
if not comment_id:
@ -2959,21 +2967,20 @@ def chapter_time(mmlir):
if initial_data and is_private is not None:
is_membersonly = False
is_premium = False
contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list)
for content in contents or []:
badges = try_get(content, lambda x: x['videoPrimaryInfoRenderer']['badges'], list)
for badge in badges or []:
label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label']) or ''
if label.lower() == 'members only':
contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or []
badge_labels = set()
for content in contents:
if not isinstance(content, dict):
continue
badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer')))
for badge_label in badge_labels:
if badge_label.lower() == 'members only':
is_membersonly = True
break
elif label.lower() == 'premium':
elif badge_label.lower() == 'premium':
is_premium = True
break
if is_membersonly or is_premium:
break
elif badge_label.lower() == 'unlisted':
is_unlisted = True
# TODO: Add this for playlists
info['availability'] = self._availability(
is_private=is_private,
needs_premium=is_premium,
@ -3447,6 +3454,17 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
},
'playlist_count': 50,
}, {
'note': 'unlisted single video playlist',
'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
'info_dict': {
'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
'uploader': 'colethedj',
'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
'title': 'yt-dlp unlisted playlist test',
'availability': 'unlisted'
},
'playlist_count': 1,
}]
@classmethod
@ -3768,18 +3786,10 @@ def _extract_selected_tab(tabs):
else:
raise ExtractorError('Unable to find selected tab')
@staticmethod
def _extract_uploader(data):
@classmethod
def _extract_uploader(cls, data):
uploader = {}
sidebar_renderer = try_get(
data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
if sidebar_renderer:
for item in sidebar_renderer:
if not isinstance(item, dict):
continue
renderer = item.get('playlistSidebarSecondaryInfoRenderer')
if not isinstance(renderer, dict):
continue
renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
owner = try_get(
renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
if owner:
@ -3814,8 +3824,8 @@ def _extract_from_tabs(self, item_id, webpage, data, tabs):
thumbnails_list = (
try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
or try_get(
data,
lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer'),
lambda x: x['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
list)
or [])
@ -3839,7 +3849,6 @@ def _extract_from_tabs(self, item_id, webpage, data, tabs):
or playlist_id)
title += format_field(selected_tab, 'title', ' - %s')
title += format_field(selected_tab, 'expandedText', ' - %s')
metadata = {
'playlist_id': playlist_id,
'playlist_title': title,
@ -3850,6 +3859,9 @@ def _extract_from_tabs(self, item_id, webpage, data, tabs):
'thumbnails': thumbnails,
'tags': tags,
}
availability = self._extract_availability(data)
if availability:
metadata['availability'] = availability
if not channel_id:
metadata.update(self._extract_uploader(data))
metadata.update({
@ -3921,19 +3933,56 @@ def _extract_from_playlist(self, item_id, url, data, playlist, webpage):
self._extract_mix_playlist(playlist, playlist_id, data, webpage),
playlist_id=playlist_id, playlist_title=title)
def _extract_availability(self, data):
"""
Gets the availability of a given playlist/tab.
Note: Unless YouTube tells us explicitly, we do not assume it is public
@param data: response
"""
is_private = is_unlisted = None
renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
badge_labels = self._extract_badges(renderer)
# Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
privacy_dropdown_entries = try_get(
renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or []
for renderer_dict in privacy_dropdown_entries:
is_selected = try_get(
renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False
if not is_selected:
continue
label = self._join_text_entries(
try_get(renderer_dict, lambda x: x['privacyDropdownItemRenderer']['label']['runs'], list) or [])
if label:
badge_labels.add(label.lower())
break
for badge_label in badge_labels:
if badge_label == 'unlisted':
is_unlisted = True
elif badge_label == 'private':
is_private = True
elif badge_label == 'public':
is_unlisted = is_private = False
return self._availability(is_private, False, False, False, is_unlisted)
@staticmethod
def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
sidebar_renderer = try_get(
data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
for item in sidebar_renderer:
renderer = try_get(item, lambda x: x[info_renderer], expected_type)
if renderer:
return renderer
def _reload_with_unavailable_videos(self, item_id, data, webpage):
"""
Get playlist with unavailable videos if the 'show unavailable videos' button exists.
"""
sidebar_renderer = try_get(
data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
if not sidebar_renderer:
return
browse_id = params = None
for item in sidebar_renderer:
if not isinstance(item, dict):
continue
renderer = item.get('playlistSidebarPrimaryInfoRenderer')
renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
if not renderer:
return
menu_renderer = try_get(
renderer, lambda x: x['menu']['menuRenderer']['items'], list) or []
for menu_item in menu_renderer:
@ -4100,7 +4149,6 @@ def get_mobj(url):
if 'no-youtube-unavailable-videos' not in compat_opts:
data = self._reload_with_unavailable_videos(item_id, data, webpage) or data
self._extract_and_report_alerts(data)
tabs = try_get(
data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
if tabs: