Backups/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-10-04 15:34:58 -04:00
Code Issues Projects Releases Wiki Activity

[compat] Remove deprecated functions from core code

Browse Source
This commit is contained in:
pukkandan
2022-06-24 16:24:43 +05:30
parent 54007a45f1
commit 14f25df2b6
30 changed files with 203 additions and 245 deletions
Download Patch File Download Diff File Expand all files Collapse all files

99
yt_dlp/extractor/youtube.py
View File

@@ -13,18 +13,11 @@ import sys
import threading
import time
import traceback
import urllib.error
import urllib.parse
from .common import InfoExtractor, SearchInfoExtractor
from ..compat import functools # isort: split
from ..compat import (
compat_HTTPError,
compat_parse_qs,
compat_str,
compat_urllib_parse_urlencode,
compat_urllib_parse_urlparse,
compat_urlparse,
)
from ..compat import functools
from ..jsinterp import JSInterpreter
from ..utils import (
NO_DEFAULT,
@@ -381,11 +374,11 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
pref = {}
if pref_cookie:
try:
pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
except ValueError:
self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
pref.update({'hl': 'en', 'tz': 'UTC'})
self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
def _real_initialize(self):
self._initialize_pref()
@@ -413,19 +406,19 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
def _extract_client_name(self, ytcfg, default_client='web'):
return self._ytcfg_get_safe(
ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client)
lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
def _extract_client_version(self, ytcfg, default_client='web'):
return self._ytcfg_get_safe(
ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client)
lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
def _select_api_hostname(self, req_api_hostname, default_client=None):
return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
or req_api_hostname or self._get_innertube_host(default_client or 'web'))
def _extract_api_key(self, ytcfg=None, default_client='web'):
return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
def _extract_context(self, ytcfg=None, default_client='web'):
context = get_first(
@@ -497,7 +490,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
# Deprecated?
def _extract_identity_token(self, ytcfg=None, webpage=None):
if ytcfg:
token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
if token:
return token
if webpage:
@@ -513,12 +506,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
"""
for data in args:
# ytcfg includes channel_syncid if on secondary channel
delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str)
delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
if delegated_sid:
return delegated_sid
sync_ids = (try_get(
data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
lambda x: x['DATASYNC_ID']), compat_str) or '').split('||')
lambda x: x['DATASYNC_ID']), str) or '').split('||')
if len(sync_ids) >= 2 and sync_ids[1]:
# datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
# and just "user_syncid||" for primary channel. We only want the channel_syncid
@@ -552,7 +545,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
headers = {
'X-YouTube-Client-Name': compat_str(
'X-YouTube-Client-Name': str(
self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
'Origin': origin,
@@ -612,7 +605,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
def _extract_continuation_ep_data(cls, continuation_ep: dict):
if isinstance(continuation_ep, dict):
continuation = try_get(
continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
continuation_ep, lambda x: x['continuationCommand']['token'], str)
if not continuation:
return
ctp = continuation_ep.get('clickTrackingParams')
@@ -672,7 +665,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
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)
label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], str)
if label:
badges.add(label.lower())
return badges
@@ -687,7 +680,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
obj = [obj]
for item in obj:
text = try_get(item, lambda x: x['simpleText'], compat_str)
text = try_get(item, lambda x: x['simpleText'], str)
if text:
return text
runs = try_get(item, lambda x: x['runs'], list) or []
@@ -789,20 +782,20 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
except ExtractorError as e:
if isinstance(e.cause, network_exceptions):
if isinstance(e.cause, compat_HTTPError):
if isinstance(e.cause, urllib.error.HTTPError):
first_bytes = e.cause.read(512)
if not is_html(first_bytes):
yt_error = try_get(
self._parse_json(
self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
lambda x: x['error']['message'], compat_str)
lambda x: x['error']['message'], str)
if yt_error:
self._report_alerts([('ERROR', yt_error)], fatal=False)
# Downloading page may result in intermittent 5xx HTTP error
# Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
# We also want to catch all other network exceptions since errors in later pages can be troublesome
# See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
last_error = error_to_compat_str(e.cause or e.msg)
if count < retries:
continue
@@ -2345,7 +2338,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Obtain from MPD's maximum seq value
old_mpd_url = mpd_url
last_error = ctx.pop('last_error', None)
expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
or (mpd_url, stream_number, False))
if not refresh_sequence:
@@ -2427,7 +2420,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _extract_player_url(self, *ytcfgs, webpage=None):
player_url = traverse_obj(
ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
get_all=False, expected_type=compat_str)
get_all=False, expected_type=str)
if not player_url:
return
return urljoin('https://www.youtube.com', player_url)
@@ -2444,7 +2437,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _signature_cache_id(self, example_sig):
""" Return a string representation of a signature """
return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
return '.'.join(str(len(part)) for part in example_sig.split('.'))
@classmethod
def _extract_player_info(cls, player_url):
@@ -2526,7 +2519,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
cache_spec = [ord(c) for c in cache_res]
expr_code = ' + '.join(gen_sig_code(cache_spec))
signature_id_tuple = '(%s)' % (
', '.join(compat_str(len(p)) for p in example_sig.split('.')))
', '.join(str(len(p)) for p in example_sig.split('.')))
code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
' return %s\n') % (signature_id_tuple, expr_code)
self.to_screen('Extracted signature function:\n' + code)
@@ -2649,8 +2642,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if not url:
self.report_warning(f'Unable to mark {label}watched')
return
parsed_url = compat_urlparse.urlparse(url)
qs = compat_urlparse.parse_qs(parsed_url.query)
parsed_url = urllib.parse.urlparse(url)
qs = urllib.parse.parse_qs(parsed_url.query)
# cpn generation algorithm is reverse engineered from base.js.
# In fact it works even with dummy cpn.
@@ -2675,8 +2668,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'et': video_length,
})
url = compat_urlparse.urlunparse(
parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
url = urllib.parse.urlunparse(
parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
self._download_webpage(
url, video_id, f'Marking {label}watched',
@@ -2793,12 +2786,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
author = self._get_text(comment_renderer, 'authorText')
author_id = try_get(comment_renderer,
lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], str)
votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
lambda x: x['likeCount']), compat_str)) or 0
lambda x: x['likeCount']), str)) or 0
author_thumbnail = try_get(comment_renderer,
lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], str)
author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
is_favorited = 'creatorHeart' in (try_get(
@@ -3178,7 +3171,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
fmt_url = fmt.get('url')
if not fmt_url:
sc = compat_parse_qs(fmt.get('signatureCipher'))
sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
encrypted_sig = try_get(sc, lambda x: x['s'][0])
if not all((sc, fmt_url, player_url, encrypted_sig)):
@@ -3419,12 +3412,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Unquote should take place before split on comma (,) since textual
# fields may contain comma as well (see
# https://github.com/ytdl-org/youtube-dl/issues/8536)
feed_data = compat_parse_qs(
feed_data = urllib.parse.parse_qs(
urllib.parse.unquote_plus(feed))
def feed_entry(name):
return try_get(
feed_data, lambda x: x[name][0], compat_str)
feed_data, lambda x: x[name][0], str)
feed_id = feed_entry('id')
if not feed_id:
@@ -3651,9 +3644,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
info['automatic_captions'] = automatic_captions
info['subtitles'] = subtitles
parsed_url = compat_urllib_parse_urlparse(url)
parsed_url = urllib.parse.urlparse(url)
for component in [parsed_url.fragment, parsed_url.query]:
query = compat_parse_qs(component)
query = urllib.parse.parse_qs(component)
for k, v in query.items():
for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
d_k += '_time'
@@ -3946,7 +3939,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
# generic endpoint URL support
ep_url = urljoin('https://www.youtube.com/', try_get(
renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
compat_str))
str))
if ep_url:
for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
if ie.suitable(ep_url):
@@ -3990,7 +3983,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
def _shelf_entries(self, shelf_renderer, skip_channels=False):
ep = try_get(
shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
compat_str)
str)
shelf_url = urljoin('https://www.youtube.com', ep)
if shelf_url:
# Skipping links to another channels, note that checking for
@@ -4050,7 +4043,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
yield entry
# playlist attachment
playlist_id = try_get(
post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str)
post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
if playlist_id:
yield self.url_result(
'https://www.youtube.com/playlist?list=%s' % playlist_id,
@@ -4061,7 +4054,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
if not isinstance(run, dict):
continue
ep_url = try_get(
run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
if not ep_url:
continue
if not YoutubeIE.suitable(ep_url):
@@ -4238,10 +4231,10 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
uploader['uploader'] = self._search_regex(
r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
uploader['uploader_id'] = try_get(
owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], str)
uploader['uploader_url'] = urljoin(
'https://www.youtube.com/',
try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], str))
return {k: v for k, v in uploader.items() if v is not None}
def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
@@ -4369,13 +4362,13 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
title = playlist.get('title') or try_get(
data, lambda x: x['titleText']['simpleText'], compat_str)
data, lambda x: x['titleText']['simpleText'], str)
playlist_id = playlist.get('playlistId') or item_id
# Delegating everything except mix playlists to regular tab-based playlist URL
playlist_url = urljoin(url, try_get(
playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
compat_str))
str))
# Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
# [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
@@ -4446,7 +4439,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
continue
nav_item_renderer = menu_item.get('menuNavigationItemRenderer')
text = try_get(
nav_item_renderer, lambda x: x['text']['simpleText'], compat_str)
nav_item_renderer, lambda x: x['text']['simpleText'], str)
if not text or text.lower() != 'show unavailable videos':
continue
browse_endpoint = try_get(
@@ -4488,7 +4481,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
except ExtractorError as e:
if isinstance(e.cause, network_exceptions):
if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429):
if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
last_error = error_to_compat_str(e.cause or e.msg)
if count < retries:
continue
@@ -5301,8 +5294,8 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
@YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
def _real_extract(self, url, smuggled_data):
item_id = self._match_id(url)
url = compat_urlparse.urlunparse(
compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
url = urllib.parse.urlunparse(
urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
compat_opts = self.get_param('compat_opts', [])
def get_mobj(url):
@@ -5322,7 +5315,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
mdata = self._extract_tab_endpoint(
f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
get_all=False, expected_type=compat_str)
get_all=False, expected_type=str)
if not murl:
raise ExtractorError('Failed to resolve album to playlist')
return self.url_result(murl, ie=YoutubeTabIE.ie_key())