artem/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2026-08-26 13:53:45 +03:00
Code Issues Packages Projects Releases Wiki Activity

merges _extractors.py

Browse Source
This commit is contained in:
shloop
2024-08-17 12:59:44 -07:00
parent a2f4163e25 d62fef7e07
commit 2aa9e75975
1263 changed files with 67720 additions and 39232 deletions
Download Patch File Download Diff File Expand all files Collapse all files
yt_dlp/YoutubeDL.py
+800 -445
View File
File diff suppressed because it is too large Load Diff
yt_dlp/__init__.py
+180 -66
View File
@@ -1,10 +1,10 @@
try:
import contextvars # noqa: F401
except Exception:
raise Exception(
f'You are using an unsupported version of Python. Only Python versions 3.7 and above are supported by yt-dlp') # noqa: F541
import sys
__license__ = 'Public Domain'
if sys.version_info < (3, 8):
raise ImportError(
f'You are using an unsupported version of Python. Only Python versions 3.8 and above are supported by yt-dlp') # noqa: F541
__license__ = 'The Unlicense'
import collections
import getpass
@@ -12,13 +12,14 @@ import itertools
import optparse
import os
import re
import sys
import traceback
from .compat import compat_shlex_quote
from .compat import compat_os_name
from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
from .downloader.external import get_external_downloader
from .extractor import list_extractor_classes
from .extractor.adobepass import MSO_INFO
from .networking.impersonate import ImpersonateTarget
from .options import parseOpts
from .postprocessor import (
FFmpegExtractAudioPP,
@@ -48,6 +49,7 @@ from .utils import (
float_or_none,
format_field,
int_or_none,
join_nonempty,
match_filter_func,
parse_bytes,
parse_duration,
@@ -56,11 +58,13 @@ from .utils import (
read_stdin,
render_table,
setproctitle,
std_headers,
shell_quote,
traverse_obj,
variadic,
write_string,
)
from .utils.networking import std_headers
from .utils._utils import _UnsafeExtensionError
from .YoutubeDL import YoutubeDL
_IN_CLI = False
@@ -73,14 +77,16 @@ def _exit(status=0, *args):
def get_urls(urls, batchfile, verbose):
# Batch file verification
"""
@param verbose -1: quiet, 0: normal, 1: verbose
"""
batch_urls = []
if batchfile is not None:
try:
batch_urls = read_batch_urls(
read_stdin('URLs') if batchfile == '-'
read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
if verbose:
if verbose == 1:
write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
except OSError:
_exit(f'ERROR: batch file {batchfile} could not be read')
@@ -111,9 +117,9 @@ def print_extractor_information(opts, urls):
ie.description(markdown=False, search_examples=_SEARCHES)
for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
elif opts.ap_list_mso:
out = 'Supported TV Providers:\n%s\n' % render_table(
out = 'Supported TV Providers:\n{}\n'.format(render_table(
['mso', 'mso name'],
[[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
[[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]))
else:
return False
write_string(out, out=sys.stdout)
@@ -125,7 +131,7 @@ def set_compat_opts(opts):
if name not in opts.compat_opts:
return False
opts.compat_opts.discard(name)
opts.compat_opts.update(['*%s' % name])
opts.compat_opts.update([f'*{name}'])
return True
def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
@@ -187,8 +193,8 @@ def validate_options(opts):
raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
# Usernames and passwords
validate(not opts.usenetrc or (opts.username is None and opts.password is None),
'.netrc', msg='using {name} conflicts with giving username/password')
validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
msg='{name}, netrc command and username/password are mutually exclusive options')
validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
validate(opts.ap_password is None or opts.ap_username is not None,
'TV Provider account username', msg='{name} missing')
@@ -218,7 +224,7 @@ def validate_options(opts):
validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
if opts.wait_for_video is not None:
min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
min_wait, max_wait, *_ = map(parse_duration, [*opts.wait_for_video.split('-', 1), None])
validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
'time range to wait for video', opts.wait_for_video)
validate_minmax(min_wait, max_wait, 'time range to wait for video')
@@ -260,9 +266,9 @@ def validate_options(opts):
# Retry sleep function
def parse_sleep_func(expr):
NUMBER_RE = r'\d+(?:\.\d+)?'
op, start, limit, step, *_ = tuple(re.fullmatch(
op, start, limit, step, *_ = (*tuple(re.fullmatch(
rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
expr.strip()).groups()) + (None, None)
expr.strip()).groups()), None, None)
if op == 'exp':
return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
@@ -319,26 +325,49 @@ def validate_options(opts):
opts.skip_download = None
del opts.outtmpl['default']
def parse_chapters(name, value):
chapters, ranges = [], []
def parse_chapters(name, value, advanced=False):
parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
for regex in value or []:
if regex.startswith('*'):
for range_ in map(str.strip, regex[1:].split(',')):
mobj = range_ != '-' and re.fullmatch(r'([^-]+)?\s*-\s*([^-]+)?', range_)
dur = mobj and (parse_timestamp(mobj.group(1) or '0'), parse_timestamp(mobj.group(2) or 'inf'))
if None in (dur or [None]):
raise ValueError(f'invalid {name} time range "{regex}". Must be of the form "*start-end"')
ranges.append(dur)
continue
try:
chapters.append(re.compile(regex))
except re.error as err:
raise ValueError(f'invalid {name} regex "{regex}" - {err}')
return chapters, ranges
TIMESTAMP_RE = r'''(?x)(?:
(?P<start_sign>-?)(?P<start>[^-]+)
)?\s*-\s*(?:
(?P<end_sign>-?)(?P<end>[^-]+)
)?'''
opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
chapters, ranges, from_url = [], [], False
for regex in value or []:
if advanced and regex == '*from-url':
from_url = True
continue
elif not regex.startswith('*'):
try:
chapters.append(re.compile(regex))
except re.error as err:
raise ValueError(f'invalid {name} regex "{regex}" - {err}')
continue
for range_ in map(str.strip, regex[1:].split(',')):
mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
err = None
if None in (dur or [None]):
err = 'Must be of the form "*start-end"'
elif not advanced and any(signs):
err = 'Negative timestamps are not allowed'
else:
dur[0] *= -1 if signs[0] else 1
dur[1] *= -1 if signs[1] else 1
if dur[1] == float('-inf'):
err = '"-inf" is not a valid end'
if err:
raise ValueError(f'invalid {name} time range "{regex}". {err}')
ranges.append(dur)
return chapters, ranges, from_url
opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
# Cookies from browser
if opts.cookiesfrombrowser:
@@ -363,16 +392,19 @@ def validate_options(opts):
f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
if opts.impersonate is not None:
opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
# MetadataParser
def metadataparser_actions(f):
if isinstance(f, str):
cmd = '--parse-metadata %s' % compat_shlex_quote(f)
cmd = f'--parse-metadata {shell_quote(f)}'
try:
actions = [MetadataFromFieldPP.to_action(f)]
except Exception as err:
raise ValueError(f'{cmd} is invalid; {err}')
else:
cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
cmd = f'--replace-in-metadata {shell_quote(f)}'
actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
for action in actions:
@@ -383,7 +415,7 @@ def validate_options(opts):
yield action
if opts.metafromtitle is not None:
opts.parse_metadata.setdefault('pre_process', []).append('title:%s' % opts.metafromtitle)
opts.parse_metadata.setdefault('pre_process', []).append(f'title:{opts.metafromtitle}')
opts.parse_metadata = {
k: list(itertools.chain(*map(metadataparser_actions, v)))
for k, v in opts.parse_metadata.items()
@@ -396,14 +428,19 @@ def validate_options(opts):
except Exception as err:
raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
if geo_bypass_code is not None:
opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
if opts.geo_bypass.lower() not in ('default', 'never'):
try:
GeoUtils.random_ipv4(geo_bypass_code)
GeoUtils.random_ipv4(opts.geo_bypass)
except Exception:
raise ValueError('unsupported geo-bypass country or ip-block')
raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
if len(opts.geo_bypass) == 2:
opts.geo_bypass_country = opts.geo_bypass
else:
opts.geo_bypass_ip_block = opts.geo_bypass
opts.geo_bypass = opts.geo_bypass.lower() != 'never'
opts.match_filter = match_filter_func(opts.match_filter)
opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
if opts.download_archive is not None:
opts.download_archive = expand_path(opts.download_archive)
@@ -430,6 +467,10 @@ def validate_options(opts):
elif ed and proto == 'default':
default_downloader = ed.get_basename()
for policy in opts.color.values():
if policy not in ('always', 'auto', 'auto-tty', 'no_color', 'no_color-tty', 'never'):
raise ValueError(f'"{policy}" is not a valid color policy')
warnings, deprecation_warnings = [], []
# Common mistake: -f best
@@ -553,6 +594,13 @@ def validate_options(opts):
if opts.ap_username is not None and opts.ap_password is None:
opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
# compat option changes global state destructively; only allow from cli
if 'allow-unsafe-ext' in opts.compat_opts:
warnings.append(
'Using allow-unsafe-ext opens you up to potential attacks. '
'Use with great care!')
_UnsafeExtensionError.sanitize_extension = lambda x, prepend=False: x
return warnings, deprecation_warnings
@@ -563,7 +611,7 @@ def get_postprocessors(opts):
yield {
'key': 'MetadataParser',
'actions': actions,
'when': when
'when': when,
}
sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
if sponsorblock_query:
@@ -571,19 +619,19 @@ def get_postprocessors(opts):
'key': 'SponsorBlock',
'categories': sponsorblock_query,
'api': opts.sponsorblock_api,
'when': 'after_filter'
'when': 'after_filter',
}
if opts.convertsubtitles:
yield {
'key': 'FFmpegSubtitlesConvertor',
'format': opts.convertsubtitles,
'when': 'before_dl'
'when': 'before_dl',
}
if opts.convertthumbnails:
yield {
'key': 'FFmpegThumbnailsConvertor',
'format': opts.convertthumbnails,
'when': 'before_dl'
'when': 'before_dl',
}
if opts.extractaudio:
yield {
@@ -608,7 +656,7 @@ def get_postprocessors(opts):
yield {
'key': 'FFmpegEmbedSubtitle',
# already_have_subtitle = True prevents the file from being deleted after embedding
'already_have_subtitle': opts.writesubtitles and keep_subs
'already_have_subtitle': opts.writesubtitles and keep_subs,
}
if not opts.writeautomaticsub and keep_subs:
opts.writesubtitles = True
@@ -621,7 +669,7 @@ def get_postprocessors(opts):
'remove_sponsor_segments': opts.sponsorblock_remove,
'remove_ranges': opts.remove_ranges,
'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
'force_keyframes': opts.force_keyframes_at_cuts
'force_keyframes': opts.force_keyframes_at_cuts,
}
# FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
# FFmpegExtractAudioPP as containers before conversion may not support
@@ -655,7 +703,7 @@ def get_postprocessors(opts):
yield {
'key': 'EmbedThumbnail',
# already_have_thumbnail = True prevents the file from being deleted after embedding
'already_have_thumbnail': opts.writethumbnail
'already_have_thumbnail': opts.writethumbnail,
}
if not opts.writethumbnail:
opts.writethumbnail = True
@@ -689,7 +737,7 @@ ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'u
def parse_options(argv=None):
"""@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
parser, opts, urls = parseOpts(argv)
urls = get_urls(urls, opts.batchfile, opts.verbose)
urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
set_compat_opts(opts)
try:
@@ -702,9 +750,10 @@ def parse_options(argv=None):
print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
any_getting = any(getattr(opts, k) for k in (
'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl',
))
opts.quiet = opts.quiet or any_getting or opts.print_json or bool(opts.forceprint)
if opts.quiet is None:
opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
@@ -730,6 +779,7 @@ def parse_options(argv=None):
return ParsedOptions(parser, opts, urls, {
'usenetrc': opts.usenetrc,
'netrc_location': opts.netrc_location,
'netrc_cmd': opts.netrc_cmd,
'username': opts.username,
'password': opts.password,
'twofactor': opts.twofactor,
@@ -795,6 +845,7 @@ def parse_options(argv=None):
'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
'progress_with_newline': opts.progress_with_newline,
'progress_template': opts.progress_template,
'progress_delta': opts.progress_delta,
'playliststart': opts.playliststart,
'playlistend': opts.playlistend,
'playlistreverse': opts.playlist_reverse,
@@ -875,6 +926,7 @@ def parse_options(argv=None):
'postprocessors': postprocessors,
'fixup': opts.fixup,
'source_address': opts.source_address,
'impersonate': opts.impersonate,
'call_home': opts.call_home,
'sleep_interval_requests': opts.sleep_interval_requests,
'sleep_interval': opts.sleep_interval,
@@ -887,7 +939,7 @@ def parse_options(argv=None):
'playlist_items': opts.playlist_items,
'xattr_set_filesize': opts.xattr_set_filesize,
'match_filter': opts.match_filter,
'no_color': opts.no_color,
'color': opts.color,
'ffmpeg_location': opts.ffmpeg_location,
'hls_prefer_native': opts.hls_prefer_native,
'hls_use_mpegts': opts.hls_use_mpegts,
@@ -931,20 +983,80 @@ def _real_main(argv=None):
if opts.rm_cachedir:
ydl.cache.remove()
updater = Updater(ydl)
if opts.update_self and updater.update() and actual_use:
if updater.cmd:
return updater.restart()
# This code is reachable only for zip variant in py < 3.10
# It makes sense to exit here, but the old behavior is to continue
ydl.report_warning('Restart yt-dlp to use the updated version')
# return 100, 'ERROR: The program must exit for the update to complete'
try:
updater = Updater(ydl, opts.update_self)
if opts.update_self and updater.update() and actual_use:
if updater.cmd:
return updater.restart()
# This code is reachable only for zip variant in py < 3.10
# It makes sense to exit here, but the old behavior is to continue
ydl.report_warning('Restart yt-dlp to use the updated version')
# return 100, 'ERROR: The program must exit for the update to complete'
except Exception:
traceback.print_exc()
ydl._download_retcode = 100
if opts.list_impersonate_targets:
known_targets = [
# List of simplified targets we know are supported,
# to help users know what dependencies may be required.
(ImpersonateTarget('chrome'), 'curl_cffi'),
(ImpersonateTarget('edge'), 'curl_cffi'),
(ImpersonateTarget('safari'), 'curl_cffi'),
]
available_targets = ydl._get_available_impersonate_targets()
def make_row(target, handler):
return [
join_nonempty(target.client.title(), target.version, delim='-') or '-',
join_nonempty((target.os or '').title(), target.os_version, delim='-') or '-',
handler,
]
rows = [make_row(target, handler) for target, handler in available_targets]
for known_target, known_handler in known_targets:
if not any(
known_target in target and handler == known_handler
for target, handler in available_targets
):
rows.append([
ydl._format_out(text, ydl.Styles.SUPPRESS)
for text in make_row(known_target, f'{known_handler} (not available)')
])
ydl.to_screen('[info] Available impersonate targets')
ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
return
if not actual_use:
if pre_process:
return ydl._download_retcode
ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
args = sys.argv[1:] if argv is None else argv
ydl.warn_if_short_id(args)
# Show a useful error message and wait for keypress if not launched from shell on Windows
if not args and compat_os_name == 'nt' and getattr(sys, 'frozen', False):
import ctypes.wintypes
import msvcrt
kernel32 = ctypes.WinDLL('Kernel32')
buffer = (1 * ctypes.wintypes.DWORD)()
attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
# If we only have a single process attached, then the executable was double clicked
# When using `pyinstaller` with `--onefile`, two processes get attached
is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
if attached_processes == 1 or is_onefile and attached_processes == 2:
print(parser._generate_error_message(
'Do not double-click the executable, instead call it from a command line.\n'
'Please read the README for further information on how to use yt-dlp: '
'https://github.com/yt-dlp/yt-dlp#readme'))
msvcrt.getch()
_exit(2)
parser.error(
'You must provide at least one URL.\n'
'Type yt-dlp --help to see a list of all options.')
@@ -952,6 +1064,8 @@ def _real_main(argv=None):
parser.destroy()
try:
if opts.load_info_filename is not None:
if all_urls:
ydl.report_warning('URLs are ignored due to --load-info-json')
return ydl.download_with_info_file(expand_path(opts.load_info_filename))
else:
return ydl.download(all_urls)
yt_dlp/__main__.py
+1 -1
View File
@@ -1,7 +1,7 @@
#!/usr/bin/env python3
# Execute with
# $ python -m yt_dlp
# $ python3 -m yt_dlp
import sys
yt_dlp/__pyinstaller/hook-yt_dlp.py
+12 -33
View File
@@ -1,28 +1,6 @@
import ast
import os
import sys
from pathlib import Path
from PyInstaller.utils.hooks import collect_submodules
def find_attribute_accesses(node, name, path=()):
if isinstance(node, ast.Attribute):
path = [*path, node.attr]
if isinstance(node.value, ast.Name) and node.value.id == name:
yield path[::-1]
for child in ast.iter_child_nodes(node):
yield from find_attribute_accesses(child, name, path)
def collect_used_submodules(name, level):
for dirpath, _, filenames in os.walk(Path(__file__).parent.parent):
for filename in filenames:
if not filename.endswith('.py'):
continue
with open(Path(dirpath) / filename, encoding='utf8') as f:
for submodule in find_attribute_accesses(ast.parse(f.read()), name):
yield '.'.join(submodule[:level])
from PyInstaller.utils.hooks import collect_submodules, collect_data_files
def pycryptodome_module():
@@ -32,7 +10,7 @@ def pycryptodome_module():
try:
import Crypto # noqa: F401
print('WARNING: Using Crypto since Cryptodome is not available. '
'Install with: pip install pycryptodomex', file=sys.stderr)
'Install with: python3 -m pip install pycryptodomex', file=sys.stderr)
return 'Crypto'
except ImportError:
pass
@@ -40,18 +18,19 @@ def pycryptodome_module():
def get_hidden_imports():
yield 'yt_dlp.compat._legacy'
yield from collect_submodules('websockets')
crypto = pycryptodome_module()
for sm in set(collect_used_submodules('Cryptodome', 2)):
yield f'{crypto}.{sm}'
yield from ('yt_dlp.compat._legacy', 'yt_dlp.compat._deprecated')
yield from ('yt_dlp.utils._legacy', 'yt_dlp.utils._deprecated')
yield pycryptodome_module()
# Only `websockets` is required, others are collected just in case
for module in ('websockets', 'requests', 'urllib3'):
yield from collect_submodules(module)
# These are auto-detected, but explicitly add them just in case
yield from ('mutagen', 'brotli', 'certifi')
yield from ('mutagen', 'brotli', 'certifi', 'secretstorage', 'curl_cffi')
hiddenimports = list(get_hidden_imports())
print(f'Adding imports: {hiddenimports}')
excludedimports = ['youtube_dl', 'youtube_dlc', 'test', 'ytdlp_plugins', 'devscripts']
excludedimports = ['youtube_dl', 'youtube_dlc', 'test', 'ytdlp_plugins', 'devscripts', 'bundle']
datas = collect_data_files('curl_cffi', includes=['cacert.pem'])
yt_dlp/aes.py
+15 -27
View File
@@ -5,14 +5,14 @@ from .compat import compat_ord
from .dependencies import Cryptodome
from .utils import bytes_to_intlist, intlist_to_bytes
if Cryptodome:
if Cryptodome.AES:
def aes_cbc_decrypt_bytes(data, key, iv):
""" Decrypt bytes with AES-CBC using pycryptodome """
return Cryptodome.Cipher.AES.new(key, Cryptodome.Cipher.AES.MODE_CBC, iv).decrypt(data)
return Cryptodome.AES.new(key, Cryptodome.AES.MODE_CBC, iv).decrypt(data)
def aes_gcm_decrypt_and_verify_bytes(data, key, tag, nonce):
""" Decrypt bytes with AES-GCM using pycryptodome """
return Cryptodome.Cipher.AES.new(key, Cryptodome.Cipher.AES.MODE_GCM, nonce).decrypt_and_verify(data, tag)
return Cryptodome.AES.new(key, Cryptodome.AES.MODE_GCM, nonce).decrypt_and_verify(data, tag)
else:
def aes_cbc_decrypt_bytes(data, key, iv):
@@ -68,7 +68,7 @@ def pad_block(block, padding_mode):
raise NotImplementedError(f'Padding mode {padding_mode} is not implemented')
if padding_mode == 'iso7816' and padding_size:
block = block + [0x80] # NB: += mutates list
block = [*block, 0x80] # NB: += mutates list
padding_size -= 1
return block + [PADDING_BYTE[padding_mode]] * padding_size
@@ -110,9 +110,7 @@ def aes_ecb_decrypt(data, key, iv=None):
for i in range(block_count):
block = data[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES]
encrypted_data += aes_decrypt(block, expanded_key)
encrypted_data = encrypted_data[:len(data)]
return encrypted_data
return encrypted_data[:len(data)]
def aes_ctr_decrypt(data, key, iv):
@@ -148,9 +146,7 @@ def aes_ctr_encrypt(data, key, iv):
cipher_counter_block = aes_encrypt(counter_block, expanded_key)
encrypted_data += xor(block, cipher_counter_block)
encrypted_data = encrypted_data[:len(data)]
return encrypted_data
return encrypted_data[:len(data)]
def aes_cbc_decrypt(data, key, iv):
@@ -174,9 +170,7 @@ def aes_cbc_decrypt(data, key, iv):
decrypted_block = aes_decrypt(block, expanded_key)
decrypted_data += xor(decrypted_block, previous_cipher_block)
previous_cipher_block = block
decrypted_data = decrypted_data[:len(data)]
return decrypted_data
return decrypted_data[:len(data)]
def aes_cbc_encrypt(data, key, iv, *, padding_mode='pkcs7'):
@@ -224,7 +218,7 @@ def aes_gcm_decrypt_and_verify(data, key, tag, nonce):
hash_subkey = aes_encrypt([0] * BLOCK_SIZE_BYTES, key_expansion(key))
if len(nonce) == 12:
j0 = nonce + [0, 0, 0, 1]
j0 = [*nonce, 0, 0, 0, 1]
else:
fill = (BLOCK_SIZE_BYTES - (len(nonce) % BLOCK_SIZE_BYTES)) % BLOCK_SIZE_BYTES + 8
ghash_in = nonce + [0] * fill + bytes_to_intlist((8 * len(nonce)).to_bytes(8, 'big'))
@@ -242,11 +236,11 @@ def aes_gcm_decrypt_and_verify(data, key, tag, nonce):
data
+ [0] * (BLOCK_SIZE_BYTES - len(data) + pad_len) # pad
+ bytes_to_intlist((0 * 8).to_bytes(8, 'big') # length of associated data
+ ((len(data) * 8).to_bytes(8, 'big'))) # length of data
+ ((len(data) * 8).to_bytes(8, 'big'))), # length of data
)
if tag != aes_ctr_encrypt(s_tag, key, j0):
raise ValueError("Mismatching authentication tag")
raise ValueError('Mismatching authentication tag')
return decrypted_data
@@ -288,9 +282,7 @@ def aes_decrypt(data, expanded_key):
data = list(iter_mix_columns(data, MIX_COLUMN_MATRIX_INV))
data = shift_rows_inv(data)
data = sub_bytes_inv(data)
data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
return data
return xor(data, expanded_key[:BLOCK_SIZE_BYTES])
def aes_decrypt_text(data, password, key_size_bytes):
@@ -318,9 +310,7 @@ def aes_decrypt_text(data, password, key_size_bytes):
cipher = data[NONCE_LENGTH_BYTES:]
decrypted_data = aes_ctr_decrypt(cipher, key, nonce + [0] * (BLOCK_SIZE_BYTES - NONCE_LENGTH_BYTES))
plaintext = intlist_to_bytes(decrypted_data)
return plaintext
return intlist_to_bytes(decrypted_data)
RCON = (0x8d, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36)
@@ -428,9 +418,7 @@ def key_expansion(data):
for _ in range(3 if key_size_bytes == 32 else 2 if key_size_bytes == 24 else 0):
temp = data[-4:]
data += xor(temp, data[-key_size_bytes: 4 - key_size_bytes])
data = data[:expanded_key_size_bytes]
return data
return data[:expanded_key_size_bytes]
def iter_vector(iv):
@@ -511,7 +499,7 @@ def block_product(block_x, block_y):
# NIST SP 800-38D, Algorithm 1
if len(block_x) != BLOCK_SIZE_BYTES or len(block_y) != BLOCK_SIZE_BYTES:
raise ValueError("Length of blocks need to be %d bytes" % BLOCK_SIZE_BYTES)
raise ValueError(f'Length of blocks need to be {BLOCK_SIZE_BYTES} bytes')
block_r = [0xE1] + [0] * (BLOCK_SIZE_BYTES - 1)
block_v = block_y[:]
@@ -534,7 +522,7 @@ def ghash(subkey, data):
# NIST SP 800-38D, Algorithm 2
if len(data) % BLOCK_SIZE_BYTES:
raise ValueError("Length of data should be %d bytes" % BLOCK_SIZE_BYTES)
raise ValueError(f'Length of data should be {BLOCK_SIZE_BYTES} bytes')
last_y = [0] * BLOCK_SIZE_BYTES
for i in range(0, len(data), BLOCK_SIZE_BYTES):
yt_dlp/cache.py
+2 -2
View File
@@ -81,10 +81,10 @@ class Cache:
cachedir = self._get_root_dir()
if not any((term in cachedir) for term in ('cache', 'tmp')):
raise Exception('Not removing directory %s - this does not look like a cache dir' % cachedir)
raise Exception(f'Not removing directory {cachedir} - this does not look like a cache dir')
self._ydl.to_screen(
'Removing cache dir %s .' % cachedir, skip_eol=True)
f'Removing cache dir {cachedir} .', skip_eol=True)
if os.path.exists(cachedir):
self._ydl.to_screen('.', skip_eol=True)
shutil.rmtree(cachedir)
yt_dlp/compat/__init__.py
+15 -11
View File
@@ -1,14 +1,11 @@
import os
import sys
import warnings
import xml.etree.ElementTree as etree
from ._deprecated import * # noqa: F401, F403
from .compat_utils import passthrough_module
# XXX: Implement this the same way as other DeprecationWarnings without circular import
passthrough_module(__name__, '._legacy', callback=lambda attr: warnings.warn(
DeprecationWarning(f'{__name__}.{attr} is deprecated'), stacklevel=5))
passthrough_module(__name__, '._deprecated')
del passthrough_module
# HTMLParseError has been deprecated in Python 3.3 and removed in
@@ -30,12 +27,9 @@ def compat_etree_fromstring(text):
compat_os_name = os._name if os.name == 'java' else os.name
if compat_os_name == 'nt':
def compat_shlex_quote(s):
import re
return s if re.match(r'^[-_\w./]+$', s) else '"%s"' % s.replace('"', '\\"')
else:
from shlex import quote as compat_shlex_quote # noqa: F401
def compat_shlex_quote(s):
from ..utils import shell_quote
return shell_quote(s)
def compat_ord(c):
@@ -70,3 +64,13 @@ if compat_os_name in ('nt', 'ce'):
return userhome + path[i:]
else:
compat_expanduser = os.path.expanduser
def urllib_req_to_req(urllib_request):
"""Convert urllib Request to a networking Request"""
from ..networking import Request
from ..utils.networking import HTTPHeaderDict
return Request(
urllib_request.get_full_url(), data=urllib_request.data, method=urllib_request.get_method(),
headers=HTTPHeaderDict(urllib_request.headers, urllib_request.unredirected_hdrs),
extensions={'timeout': urllib_request.timeout} if hasattr(urllib_request, 'timeout') else None)
yt_dlp/compat/_deprecated.py
+8 -1
View File
@@ -1,4 +1,12 @@
"""Deprecated - New code should avoid these"""
import warnings
from .compat_utils import passthrough_module
# XXX: Implement this the same way as other DeprecationWarnings without circular import
passthrough_module(__name__, '.._legacy', callback=lambda attr: warnings.warn(
DeprecationWarning(f'{__name__}.{attr} is deprecated'), stacklevel=6))
del passthrough_module
import base64
import urllib.error
@@ -8,7 +16,6 @@ compat_str = str
compat_b64decode = base64.b64decode
compat_HTTPError = urllib.error.HTTPError
compat_urlparse = urllib.parse
compat_parse_qs = urllib.parse.parse_qs
compat_urllib_parse_unquote = urllib.parse.unquote
yt_dlp/compat/_legacy.py
+5 -4
View File
@@ -16,12 +16,12 @@ import shlex
import shutil
import socket
import struct
import subprocess
import tokenize
import urllib.error
import urllib.parse
import urllib.request
import xml.etree.ElementTree as etree
from subprocess import DEVNULL
# isort: split
import asyncio # noqa: F401
@@ -32,9 +32,10 @@ from re import match as compat_Match # noqa: F401
from . import compat_expanduser, compat_HTMLParseError, compat_realpath
from .compat_utils import passthrough_module
from ..dependencies import Cryptodome_AES as compat_pycrypto_AES # noqa: F401
from ..dependencies import brotli as compat_brotli # noqa: F401
from ..dependencies import websockets as compat_websockets # noqa: F401
from ..dependencies.Cryptodome import AES as compat_pycrypto_AES # noqa: F401
from ..networking.exceptions import HTTPError as compat_HTTPError
passthrough_module(__name__, '...utils', ('WINDOWS_VT_MODE', 'windows_enable_vt_mode'))
@@ -84,10 +85,10 @@ compat_socket_create_connection = socket.create_connection
compat_Struct = struct.Struct
compat_struct_pack = struct.pack
compat_struct_unpack = struct.unpack
compat_subprocess_get_DEVNULL = lambda: DEVNULL
compat_subprocess_get_DEVNULL = lambda: subprocess.DEVNULL
compat_tokenize_tokenize = tokenize.tokenize
compat_urllib_error = urllib.error
compat_urllib_HTTPError = urllib.error.HTTPError
compat_urllib_HTTPError = compat_HTTPError
compat_urllib_parse = urllib.parse
compat_urllib_parse_parse_qs = urllib.parse.parse_qs
compat_urllib_parse_quote = urllib.parse.quote
yt_dlp/compat/compat_utils.py
+2 -2
View File
@@ -15,7 +15,7 @@ def get_package_info(module):
name=getattr(module, '_yt_dlp__identifier', module.__name__),
version=str(next(filter(None, (
getattr(module, attr, None)
for attr in ('__version__', 'version_string', 'version')
for attr in ('_yt_dlp__version', '__version__', 'version_string', 'version')
)), None)))
@@ -48,7 +48,7 @@ def passthrough_module(parent, child, allowed_attributes=(..., ), *, callback=la
"""Passthrough parent module into a child module, creating the parent if necessary"""
def __getattr__(attr):
if _is_package(parent):
with contextlib.suppress(ImportError):
with contextlib.suppress(ModuleNotFoundError):
return importlib.import_module(f'.{attr}', parent.__name__)
ret = from_child(attr)
yt_dlp/compat/functools.py
+1 -15
View File
@@ -7,20 +7,6 @@ passthrough_module(__name__, 'functools')
del passthrough_module
try:
cache # >= 3.9
_ = cache # >= 3.9
except NameError:
cache = lru_cache(maxsize=None)
try:
cached_property # >= 3.8
except NameError:
class cached_property:
def __init__(self, func):
update_wrapper(self, func)
self.func = func
def __get__(self, instance, _):
if instance is None:
return self
setattr(instance, self.func.__name__, self.func(instance))
return getattr(instance, self.func.__name__)
yt_dlp/compat/imghdr.py
+16 -10
View File
@@ -1,16 +1,22 @@
tests = {
'webp': lambda h: h[0:4] == b'RIFF' and h[8:] == b'WEBP',
'png': lambda h: h[:8] == b'\211PNG\r\n\032\n',
'jpeg': lambda h: h[6:10] in (b'JFIF', b'Exif'),
'gif': lambda h: h[:6] in (b'GIF87a', b'GIF89a'),
}
def what(file=None, h=None):
"""Detect format of image (Currently supports jpeg, png, webp, gif only)
Ref: https://github.com/python/cpython/blob/3.10/Lib/imghdr.py
Ref: https://github.com/python/cpython/blob/3.11/Lib/imghdr.py
Ref: https://www.w3.org/Graphics/JPEG/itu-t81.pdf
"""
if h is None:
with open(file, 'rb') as f:
h = f.read(12)
return next((type_ for type_, test in tests.items() if test(h)), None)
if h.startswith(b'RIFF') and h.startswith(b'WEBP', 8):
return 'webp'
if h.startswith(b'\x89PNG'):
return 'png'
if h.startswith(b'\xFF\xD8\xFF'):
return 'jpeg'
if h.startswith(b'GIF'):
return 'gif'
return None
yt_dlp/compat/types.py
+13
View File
@@ -0,0 +1,13 @@
# flake8: noqa: F405
from types import * # noqa: F403
from .compat_utils import passthrough_module
passthrough_module(__name__, 'types')
del passthrough_module
try:
# NB: pypy has builtin NoneType, so checking NameError won't work
from types import NoneType # >= 3.10
except ImportError:
NoneType = type(None)
yt_dlp/compat/urllib/__init__.py
+10
View File
@@ -0,0 +1,10 @@
# flake8: noqa: F405
from urllib import * # noqa: F403
del request # noqa: F821
from . import request # noqa: F401
from ..compat_utils import passthrough_module
passthrough_module(__name__, 'urllib')
del passthrough_module
yt_dlp/compat/urllib/request.py
+40
View File
@@ -0,0 +1,40 @@
# flake8: noqa: F405
from urllib.request import * # noqa: F403
from ..compat_utils import passthrough_module
passthrough_module(__name__, 'urllib.request')
del passthrough_module
from .. import compat_os_name
if compat_os_name == 'nt':
# On older Python versions, proxies are extracted from Windows registry erroneously. [1]
# If the https proxy in the registry does not have a scheme, urllib will incorrectly add https:// to it. [2]
# It is unlikely that the user has actually set it to be https, so we should be fine to safely downgrade
# it to http on these older Python versions to avoid issues
# This also applies for ftp proxy type, as ftp:// proxy scheme is not supported.
# 1: https://github.com/python/cpython/issues/86793
# 2: https://github.com/python/cpython/blob/51f1ae5ceb0673316c4e4b0175384e892e33cc6e/Lib/urllib/request.py#L2683-L2698
import sys
from urllib.request import getproxies_environment, getproxies_registry
def getproxies_registry_patched():
proxies = getproxies_registry()
if (
sys.version_info >= (3, 10, 5) # https://docs.python.org/3.10/whatsnew/changelog.html#python-3-10-5-final
or (3, 9, 13) <= sys.version_info < (3, 10) # https://docs.python.org/3.9/whatsnew/changelog.html#python-3-9-13-final
):
return proxies
for scheme in ('https', 'ftp'):
if scheme in proxies and proxies[scheme].startswith(f'{scheme}://'):
proxies[scheme] = 'http' + proxies[scheme][len(scheme):]
return proxies
def getproxies():
return getproxies_environment() or getproxies_registry_patched()
del compat_os_name
yt_dlp/cookies.py
+403 -134
View File
@@ -1,7 +1,13 @@
import base64
import collections
import contextlib
import datetime as dt
import functools
import glob
import hashlib
import http.cookiejar
import http.cookies
import io
import json
import os
import re
@@ -11,15 +17,15 @@ import subprocess
import sys
import tempfile
import time
from datetime import datetime, timedelta, timezone
import urllib.request
from enum import Enum, auto
from hashlib import pbkdf2_hmac
from .aes import (
aes_cbc_decrypt_bytes,
aes_gcm_decrypt_and_verify_bytes,
unpad_pkcs7,
)
from .compat import compat_os_name
from .dependencies import (
_SECRETSTORAGE_UNAVAILABLE_REASON,
secretstorage,
@@ -27,37 +33,26 @@ from .dependencies import (
)
from .minicurses import MultilinePrinter, QuietMultilinePrinter
from .utils import (
DownloadError,
Popen,
YoutubeDLCookieJar,
error_to_str,
expand_path,
is_path_like,
sanitize_url,
str_or_none,
try_call,
write_string,
)
from .utils._utils import _YDLLogger
from .utils.networking import normalize_url
CHROMIUM_BASED_BROWSERS = {'brave', 'chrome', 'chromium', 'edge', 'opera', 'vivaldi'}
CHROMIUM_BASED_BROWSERS = {'brave', 'chrome', 'chromium', 'edge', 'opera', 'vivaldi', 'whale'}
SUPPORTED_BROWSERS = CHROMIUM_BASED_BROWSERS | {'firefox', 'safari'}
class YDLLogger:
def __init__(self, ydl=None):
self._ydl = ydl
def debug(self, message):
if self._ydl:
self._ydl.write_debug(message)
def info(self, message):
if self._ydl:
self._ydl.to_screen(f'[Cookies] {message}')
def warning(self, message, only_once=False):
if self._ydl:
self._ydl.report_warning(message, only_once)
def error(self, message):
if self._ydl:
self._ydl.report_error(message)
class YDLLogger(_YDLLogger):
def warning(self, message, only_once=False): # compat
return super().warning(message, once=only_once)
class ProgressBar(MultilinePrinter):
_DELAY, _timer = 0.1, 0
@@ -105,7 +100,7 @@ def load_cookies(cookie_file, browser_specification, ydl):
jar = YoutubeDLCookieJar(cookie_file)
if not is_filename or os.access(cookie_file, os.R_OK):
jar.load(ignore_discard=True, ignore_expires=True)
jar.load()
cookie_jars.append(jar)
return _merge_cookie_jars(cookie_jars)
@@ -126,17 +121,18 @@ def _extract_firefox_cookies(profile, container, logger):
logger.info('Extracting cookies from firefox')
if not sqlite3:
logger.warning('Cannot extract cookies from firefox without sqlite3 support. '
'Please use a python interpreter compiled with sqlite3 support')
'Please use a Python interpreter compiled with sqlite3 support')
return YoutubeDLCookieJar()
if profile is None:
search_root = _firefox_browser_dir()
search_roots = list(_firefox_browser_dirs())
elif _is_path(profile):
search_root = profile
search_roots = [profile]
else:
search_root = os.path.join(_firefox_browser_dir(), profile)
search_roots = [os.path.join(path, profile) for path in _firefox_browser_dirs()]
search_root = ', '.join(map(repr, search_roots))
cookie_database_path = _find_most_recently_used_file(search_root, 'cookies.sqlite', logger)
cookie_database_path = _newest(_firefox_cookie_dbs(search_roots))
if cookie_database_path is None:
raise FileNotFoundError(f'could not find firefox cookies database in {search_root}')
logger.debug(f'Extracting cookies from: "{cookie_database_path}"')
@@ -146,11 +142,11 @@ def _extract_firefox_cookies(profile, container, logger):
containers_path = os.path.join(os.path.dirname(cookie_database_path), 'containers.json')
if not os.path.isfile(containers_path) or not os.access(containers_path, os.R_OK):
raise FileNotFoundError(f'could not read containers.json in {search_root}')
with open(containers_path) as containers:
with open(containers_path, encoding='utf8') as containers:
identities = json.load(containers).get('identities', [])
container_id = next((context.get('userContextId') for context in identities if container in (
context.get('name'),
try_call(lambda: re.fullmatch(r'userContext([^\.]+)\.label', context['l10nID']).group())
try_call(lambda: re.fullmatch(r'userContext([^\.]+)\.label', context['l10nID']).group()),
)), None)
if not isinstance(container_id, int):
raise ValueError(f'could not find firefox container "{container}" in containers.json')
@@ -190,12 +186,25 @@ def _extract_firefox_cookies(profile, container, logger):
cursor.connection.close()
def _firefox_browser_dir():
def _firefox_browser_dirs():
if sys.platform in ('cygwin', 'win32'):
return os.path.expandvars(R'%APPDATA%\Mozilla\Firefox\Profiles')
yield os.path.expandvars(R'%APPDATA%\Mozilla\Firefox\Profiles')
elif sys.platform == 'darwin':
return os.path.expanduser('~/Library/Application Support/Firefox')
return os.path.expanduser('~/.mozilla/firefox')
yield os.path.expanduser('~/Library/Application Support/Firefox/Profiles')
else:
yield from map(os.path.expanduser, (
'~/.mozilla/firefox',
'~/snap/firefox/common/.mozilla/firefox',
'~/.var/app/org.mozilla.firefox/.mozilla/firefox',
))
def _firefox_cookie_dbs(roots):
for root in map(os.path.abspath, roots):
for pattern in ('', '*/', 'Profiles/*/'):
yield from glob.iglob(os.path.join(root, pattern, 'cookies.sqlite'))
def _get_chromium_based_browser_settings(browser_name):
@@ -210,6 +219,7 @@ def _get_chromium_based_browser_settings(browser_name):
'edge': os.path.join(appdata_local, R'Microsoft\Edge\User Data'),
'opera': os.path.join(appdata_roaming, R'Opera Software\Opera Stable'),
'vivaldi': os.path.join(appdata_local, R'Vivaldi\User Data'),
'whale': os.path.join(appdata_local, R'Naver\Naver Whale\User Data'),
}[browser_name]
elif sys.platform == 'darwin':
@@ -221,6 +231,7 @@ def _get_chromium_based_browser_settings(browser_name):
'edge': os.path.join(appdata, 'Microsoft Edge'),
'opera': os.path.join(appdata, 'com.operasoftware.Opera'),
'vivaldi': os.path.join(appdata, 'Vivaldi'),
'whale': os.path.join(appdata, 'Naver/Whale'),
}[browser_name]
else:
@@ -232,6 +243,7 @@ def _get_chromium_based_browser_settings(browser_name):
'edge': os.path.join(config, 'microsoft-edge'),
'opera': os.path.join(config, 'opera'),
'vivaldi': os.path.join(config, 'vivaldi'),
'whale': os.path.join(config, 'naver-whale'),
}[browser_name]
# Linux keyring names can be determined by snooping on dbus while opening the browser in KDE:
@@ -243,6 +255,7 @@ def _get_chromium_based_browser_settings(browser_name):
'edge': 'Microsoft Edge' if sys.platform == 'darwin' else 'Chromium',
'opera': 'Opera' if sys.platform == 'darwin' else 'Chromium',
'vivaldi': 'Vivaldi' if sys.platform == 'darwin' else 'Chrome',
'whale': 'Whale',
}[browser_name]
browsers_without_profiles = {'opera'}
@@ -250,7 +263,7 @@ def _get_chromium_based_browser_settings(browser_name):
return {
'browser_dir': browser_dir,
'keyring_name': keyring_name,
'supports_profiles': browser_name not in browsers_without_profiles
'supports_profiles': browser_name not in browsers_without_profiles,
}
@@ -259,7 +272,7 @@ def _extract_chrome_cookies(browser_name, profile, keyring, logger):
if not sqlite3:
logger.warning(f'Cannot extract cookies from {browser_name} without sqlite3 support. '
'Please use a python interpreter compiled with sqlite3 support')
'Please use a Python interpreter compiled with sqlite3 support')
return YoutubeDLCookieJar()
config = _get_chromium_based_browser_settings(browser_name)
@@ -276,7 +289,7 @@ def _extract_chrome_cookies(browser_name, profile, keyring, logger):
logger.error(f'{browser_name} does not support profiles')
search_root = config['browser_dir']
cookie_database_path = _find_most_recently_used_file(search_root, 'Cookies', logger)
cookie_database_path = _newest(_find_files(search_root, 'Cookies', logger))
if cookie_database_path is None:
raise FileNotFoundError(f'could not find {browser_name} cookies database in "{search_root}"')
logger.debug(f'Extracting cookies from: "{cookie_database_path}"')
@@ -315,6 +328,12 @@ def _extract_chrome_cookies(browser_name, profile, keyring, logger):
counts['unencrypted'] = unencrypted_cookies
logger.debug(f'cookie version breakdown: {counts}')
return jar
except PermissionError as error:
if compat_os_name == 'nt' and error.errno == 13:
message = 'Could not copy Chrome cookie database. See https://github.com/yt-dlp/yt-dlp/issues/7271 for more info'
logger.error(message)
raise DownloadError(message) # force exit
raise
finally:
if cursor is not None:
cursor.connection.close()
@@ -332,6 +351,11 @@ def _process_chrome_cookie(decryptor, host_key, name, value, encrypted_value, pa
if value is None:
return is_encrypted, None
# In chrome, session cookies have expires_utc set to 0
# In our cookie-store, cookies that do not expire should have expires set to None
if not expires_utc:
expires_utc = None
return is_encrypted, http.cookiejar.Cookie(
version=0, name=name, value=value, port=None, port_specified=False,
domain=host_key, domain_specified=bool(host_key), domain_initial_dot=host_key.startswith('.'),
@@ -346,7 +370,9 @@ class ChromeCookieDecryptor:
Linux:
- cookies are either v10 or v11
- v10: AES-CBC encrypted with a fixed key
- also attempts empty password if decryption fails
- v11: AES-CBC encrypted with an OS protected key (keyring)
- also attempts empty password if decryption fails
- v11 keys can be stored in various places depending on the activate desktop environment [2]
Mac:
@@ -361,7 +387,7 @@ class ChromeCookieDecryptor:
Sources:
- [1] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/
- [2] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/key_storage_linux.cc
- [2] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/key_storage_linux.cc
- KeyStorageLinux::CreateService
"""
@@ -383,32 +409,49 @@ class LinuxChromeCookieDecryptor(ChromeCookieDecryptor):
def __init__(self, browser_keyring_name, logger, *, keyring=None):
self._logger = logger
self._v10_key = self.derive_key(b'peanuts')
password = _get_linux_keyring_password(browser_keyring_name, keyring, logger)
self._v11_key = None if password is None else self.derive_key(password)
self._empty_key = self.derive_key(b'')
self._cookie_counts = {'v10': 0, 'v11': 0, 'other': 0}
self._browser_keyring_name = browser_keyring_name
self._keyring = keyring
@functools.cached_property
def _v11_key(self):
password = _get_linux_keyring_password(self._browser_keyring_name, self._keyring, self._logger)
return None if password is None else self.derive_key(password)
@staticmethod
def derive_key(password):
# values from
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_linux.cc
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_linux.cc
return pbkdf2_sha1(password, salt=b'saltysalt', iterations=1, key_length=16)
def decrypt(self, encrypted_value):
"""
following the same approach as the fix in [1]: if cookies fail to decrypt then attempt to decrypt
with an empty password. The failure detection is not the same as what chromium uses so the
results won't be perfect
References:
- [1] https://chromium.googlesource.com/chromium/src/+/bbd54702284caca1f92d656fdcadf2ccca6f4165%5E%21/
- a bugfix to try an empty password as a fallback
"""
version = encrypted_value[:3]
ciphertext = encrypted_value[3:]
if version == b'v10':
self._cookie_counts['v10'] += 1
return _decrypt_aes_cbc(ciphertext, self._v10_key, self._logger)
return _decrypt_aes_cbc_multi(ciphertext, (self._v10_key, self._empty_key), self._logger)
elif version == b'v11':
self._cookie_counts['v11'] += 1
if self._v11_key is None:
self._logger.warning('cannot decrypt v11 cookies: no key found', only_once=True)
return None
return _decrypt_aes_cbc(ciphertext, self._v11_key, self._logger)
return _decrypt_aes_cbc_multi(ciphertext, (self._v11_key, self._empty_key), self._logger)
else:
self._logger.warning(f'unknown cookie version: "{version}"', only_once=True)
self._cookie_counts['other'] += 1
return None
@@ -423,7 +466,7 @@ class MacChromeCookieDecryptor(ChromeCookieDecryptor):
@staticmethod
def derive_key(password):
# values from
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_mac.mm
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_mac.mm
return pbkdf2_sha1(password, salt=b'saltysalt', iterations=1003, key_length=16)
def decrypt(self, encrypted_value):
@@ -436,12 +479,12 @@ class MacChromeCookieDecryptor(ChromeCookieDecryptor):
self._logger.warning('cannot decrypt v10 cookies: no key found', only_once=True)
return None
return _decrypt_aes_cbc(ciphertext, self._v10_key, self._logger)
return _decrypt_aes_cbc_multi(ciphertext, (self._v10_key,), self._logger)
else:
self._cookie_counts['other'] += 1
# other prefixes are considered 'old data' which were stored as plaintext
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_mac.mm
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_mac.mm
return encrypted_value
@@ -461,7 +504,7 @@ class WindowsChromeCookieDecryptor(ChromeCookieDecryptor):
self._logger.warning('cannot decrypt v10 cookies: no key found', only_once=True)
return None
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_win.cc
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_win.cc
# kNonceLength
nonce_length = 96 // 8
# boringssl
@@ -478,23 +521,27 @@ class WindowsChromeCookieDecryptor(ChromeCookieDecryptor):
else:
self._cookie_counts['other'] += 1
# any other prefix means the data is DPAPI encrypted
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_win.cc
# https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_win.cc
return _decrypt_windows_dpapi(encrypted_value, self._logger).decode()
def _extract_safari_cookies(profile, logger):
if profile is not None:
logger.error('safari does not support profiles')
if sys.platform != 'darwin':
raise ValueError(f'unsupported platform: {sys.platform}')
cookies_path = os.path.expanduser('~/Library/Cookies/Cookies.binarycookies')
if not os.path.isfile(cookies_path):
logger.debug('Trying secondary cookie location')
cookies_path = os.path.expanduser('~/Library/Containers/com.apple.Safari/Data/Library/Cookies/Cookies.binarycookies')
if profile:
cookies_path = os.path.expanduser(profile)
if not os.path.isfile(cookies_path):
raise FileNotFoundError('could not find safari cookies database')
raise FileNotFoundError('custom safari cookies database not found')
else:
cookies_path = os.path.expanduser('~/Library/Cookies/Cookies.binarycookies')
if not os.path.isfile(cookies_path):
logger.debug('Trying secondary cookie location')
cookies_path = os.path.expanduser('~/Library/Containers/com.apple.Safari/Data/Library/Cookies/Cookies.binarycookies')
if not os.path.isfile(cookies_path):
raise FileNotFoundError('could not find safari cookies database')
with open(cookies_path, 'rb') as f:
cookies_data = f.read()
@@ -560,7 +607,7 @@ class DataParser:
def _mac_absolute_time_to_posix(timestamp):
return int((datetime(2001, 1, 1, 0, 0, tzinfo=timezone.utc) + timedelta(seconds=timestamp)).timestamp())
return int((dt.datetime(2001, 1, 1, 0, 0, tzinfo=dt.timezone.utc) + dt.timedelta(seconds=timestamp)).timestamp())
def _parse_safari_cookies_header(data, logger):
@@ -657,19 +704,27 @@ class _LinuxDesktopEnvironment(Enum):
"""
OTHER = auto()
CINNAMON = auto()
DEEPIN = auto()
GNOME = auto()
KDE = auto()
KDE3 = auto()
KDE4 = auto()
KDE5 = auto()
KDE6 = auto()
PANTHEON = auto()
UKUI = auto()
UNITY = auto()
XFCE = auto()
LXQT = auto()
class _LinuxKeyring(Enum):
"""
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/key_storage_util_linux.h
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/key_storage_util_linux.h
SelectedLinuxBackend
"""
KWALLET = auto()
KWALLET = auto() # KDE4
KWALLET5 = auto()
KWALLET6 = auto()
GNOMEKEYRING = auto()
BASICTEXT = auto()
@@ -677,7 +732,7 @@ class _LinuxKeyring(Enum):
SUPPORTED_KEYRINGS = _LinuxKeyring.__members__.keys()
def _get_linux_desktop_environment(env):
def _get_linux_desktop_environment(env, logger):
"""
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/base/nix/xdg_util.cc
GetDesktopEnvironment
@@ -685,58 +740,102 @@ def _get_linux_desktop_environment(env):
xdg_current_desktop = env.get('XDG_CURRENT_DESKTOP', None)
desktop_session = env.get('DESKTOP_SESSION', None)
if xdg_current_desktop is not None:
xdg_current_desktop = xdg_current_desktop.split(':')[0].strip()
if xdg_current_desktop == 'Unity':
if desktop_session is not None and 'gnome-fallback' in desktop_session:
for part in map(str.strip, xdg_current_desktop.split(':')):
if part == 'Unity':
if desktop_session is not None and 'gnome-fallback' in desktop_session:
return _LinuxDesktopEnvironment.GNOME
else:
return _LinuxDesktopEnvironment.UNITY
elif part == 'Deepin':
return _LinuxDesktopEnvironment.DEEPIN
elif part == 'GNOME':
return _LinuxDesktopEnvironment.GNOME
else:
return _LinuxDesktopEnvironment.UNITY
elif xdg_current_desktop == 'GNOME':
return _LinuxDesktopEnvironment.GNOME
elif xdg_current_desktop == 'X-Cinnamon':
return _LinuxDesktopEnvironment.CINNAMON
elif xdg_current_desktop == 'KDE':
return _LinuxDesktopEnvironment.KDE
elif xdg_current_desktop == 'Pantheon':
return _LinuxDesktopEnvironment.PANTHEON
elif xdg_current_desktop == 'XFCE':
return _LinuxDesktopEnvironment.XFCE
elif part == 'X-Cinnamon':
return _LinuxDesktopEnvironment.CINNAMON
elif part == 'KDE':
kde_version = env.get('KDE_SESSION_VERSION', None)
if kde_version == '5':
return _LinuxDesktopEnvironment.KDE5
elif kde_version == '6':
return _LinuxDesktopEnvironment.KDE6
elif kde_version == '4':
return _LinuxDesktopEnvironment.KDE4
else:
logger.info(f'unknown KDE version: "{kde_version}". Assuming KDE4')
return _LinuxDesktopEnvironment.KDE4
elif part == 'Pantheon':
return _LinuxDesktopEnvironment.PANTHEON
elif part == 'XFCE':
return _LinuxDesktopEnvironment.XFCE
elif part == 'UKUI':
return _LinuxDesktopEnvironment.UKUI
elif part == 'LXQt':
return _LinuxDesktopEnvironment.LXQT
logger.info(f'XDG_CURRENT_DESKTOP is set to an unknown value: "{xdg_current_desktop}"')
elif desktop_session is not None:
if desktop_session in ('mate', 'gnome'):
if desktop_session == 'deepin':
return _LinuxDesktopEnvironment.DEEPIN
elif desktop_session in ('mate', 'gnome'):
return _LinuxDesktopEnvironment.GNOME
elif 'kde' in desktop_session:
return _LinuxDesktopEnvironment.KDE
elif 'xfce' in desktop_session:
elif desktop_session in ('kde4', 'kde-plasma'):
return _LinuxDesktopEnvironment.KDE4
elif desktop_session == 'kde':
if 'KDE_SESSION_VERSION' in env:
return _LinuxDesktopEnvironment.KDE4
else:
return _LinuxDesktopEnvironment.KDE3
elif 'xfce' in desktop_session or desktop_session == 'xubuntu':
return _LinuxDesktopEnvironment.XFCE
elif desktop_session == 'ukui':
return _LinuxDesktopEnvironment.UKUI
else:
logger.info(f'DESKTOP_SESSION is set to an unknown value: "{desktop_session}"')
else:
if 'GNOME_DESKTOP_SESSION_ID' in env:
return _LinuxDesktopEnvironment.GNOME
elif 'KDE_FULL_SESSION' in env:
return _LinuxDesktopEnvironment.KDE
if 'KDE_SESSION_VERSION' in env:
return _LinuxDesktopEnvironment.KDE4
else:
return _LinuxDesktopEnvironment.KDE3
return _LinuxDesktopEnvironment.OTHER
def _choose_linux_keyring(logger):
"""
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/key_storage_util_linux.cc
SelectBackend
SelectBackend in [1]
There is currently support for forcing chromium to use BASIC_TEXT by creating a file called
`Disable Local Encryption` [1] in the user data dir. The function to write this file (`WriteBackendUse()` [1])
does not appear to be called anywhere other than in tests, so the user would have to create this file manually
and so would be aware enough to tell yt-dlp to use the BASIC_TEXT keyring.
References:
- [1] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/key_storage_util_linux.cc
"""
desktop_environment = _get_linux_desktop_environment(os.environ)
desktop_environment = _get_linux_desktop_environment(os.environ, logger)
logger.debug(f'detected desktop environment: {desktop_environment.name}')
if desktop_environment == _LinuxDesktopEnvironment.KDE:
if desktop_environment == _LinuxDesktopEnvironment.KDE4:
linux_keyring = _LinuxKeyring.KWALLET
elif desktop_environment == _LinuxDesktopEnvironment.OTHER:
elif desktop_environment == _LinuxDesktopEnvironment.KDE5:
linux_keyring = _LinuxKeyring.KWALLET5
elif desktop_environment == _LinuxDesktopEnvironment.KDE6:
linux_keyring = _LinuxKeyring.KWALLET6
elif desktop_environment in (
_LinuxDesktopEnvironment.KDE3, _LinuxDesktopEnvironment.LXQT, _LinuxDesktopEnvironment.OTHER,
):
linux_keyring = _LinuxKeyring.BASICTEXT
else:
linux_keyring = _LinuxKeyring.GNOMEKEYRING
return linux_keyring
def _get_kwallet_network_wallet(logger):
def _get_kwallet_network_wallet(keyring, logger):
""" The name of the wallet used to store network passwords.
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/kwallet_dbus.cc
https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/kwallet_dbus.cc
KWalletDBus::NetworkWallet
which does a dbus call to the following function:
https://api.kde.org/frameworks/kwallet/html/classKWallet_1_1Wallet.html
@@ -744,11 +843,23 @@ def _get_kwallet_network_wallet(logger):
"""
default_wallet = 'kdewallet'
try:
if keyring == _LinuxKeyring.KWALLET:
service_name = 'org.kde.kwalletd'
wallet_path = '/modules/kwalletd'
elif keyring == _LinuxKeyring.KWALLET5:
service_name = 'org.kde.kwalletd5'
wallet_path = '/modules/kwalletd5'
elif keyring == _LinuxKeyring.KWALLET6:
service_name = 'org.kde.kwalletd6'
wallet_path = '/modules/kwalletd6'
else:
raise ValueError(keyring)
stdout, _, returncode = Popen.run([
'dbus-send', '--session', '--print-reply=literal',
'--dest=org.kde.kwalletd5',
'/modules/kwalletd5',
'org.kde.KWallet.networkWallet'
f'--dest={service_name}',
wallet_path,
'org.kde.KWallet.networkWallet',
], text=True, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
if returncode:
@@ -762,8 +873,8 @@ def _get_kwallet_network_wallet(logger):
return default_wallet
def _get_kwallet_password(browser_keyring_name, logger):
logger.debug('using kwallet-query to obtain password from kwallet')
def _get_kwallet_password(browser_keyring_name, keyring, logger):
logger.debug(f'using kwallet-query to obtain password from {keyring.name}')
if shutil.which('kwallet-query') is None:
logger.error('kwallet-query command not found. KWallet and kwallet-query '
@@ -771,14 +882,14 @@ def _get_kwallet_password(browser_keyring_name, logger):
'included in the kwallet package for your distribution')
return b''
network_wallet = _get_kwallet_network_wallet(logger)
network_wallet = _get_kwallet_network_wallet(keyring, logger)
try:
stdout, _, returncode = Popen.run([
'kwallet-query',
'--read-password', f'{browser_keyring_name} Safe Storage',
'--folder', f'{browser_keyring_name} Keys',
network_wallet
network_wallet,
], stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
if returncode:
@@ -793,8 +904,9 @@ def _get_kwallet_password(browser_keyring_name, logger):
# checks hasEntry. To verify this:
# dbus-monitor "interface='org.kde.KWallet'" "type=method_return"
# while starting chrome.
# this may be a bug as the intended behaviour is to generate a random password and store
# it, but that doesn't matter here.
# this was identified as a bug later and fixed in
# https://chromium.googlesource.com/chromium/src/+/bbd54702284caca1f92d656fdcadf2ccca6f4165%5E%21/#F0
# https://chromium.googlesource.com/chromium/src/+/5463af3c39d7f5b6d11db7fbd51e38cc1974d764
return b''
else:
logger.debug('password found')
@@ -817,9 +929,8 @@ def _get_gnome_keyring_password(browser_keyring_name, logger):
for item in col.get_all_items():
if item.get_label() == f'{browser_keyring_name} Safe Storage':
return item.get_secret()
else:
logger.error('failed to read from keyring')
return b''
logger.error('failed to read from keyring')
return b''
def _get_linux_keyring_password(browser_keyring_name, keyring, logger):
@@ -832,8 +943,8 @@ def _get_linux_keyring_password(browser_keyring_name, keyring, logger):
keyring = _LinuxKeyring[keyring] if keyring else _choose_linux_keyring(logger)
logger.debug(f'Chosen keyring: {keyring.name}')
if keyring == _LinuxKeyring.KWALLET:
return _get_kwallet_password(browser_keyring_name, logger)
if keyring in (_LinuxKeyring.KWALLET, _LinuxKeyring.KWALLET5, _LinuxKeyring.KWALLET6):
return _get_kwallet_password(browser_keyring_name, keyring, logger)
elif keyring == _LinuxKeyring.GNOMEKEYRING:
return _get_gnome_keyring_password(browser_keyring_name, logger)
elif keyring == _LinuxKeyring.BASICTEXT:
@@ -861,7 +972,11 @@ def _get_mac_keyring_password(browser_keyring_name, logger):
def _get_windows_v10_key(browser_root, logger):
path = _find_most_recently_used_file(browser_root, 'Local State', logger)
"""
References:
- [1] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/sync/os_crypt_win.cc
"""
path = _newest(_find_files(browser_root, 'Local State', logger))
if path is None:
logger.error('could not find local state file')
return None
@@ -869,11 +984,13 @@ def _get_windows_v10_key(browser_root, logger):
with open(path, encoding='utf8') as f:
data = json.load(f)
try:
# kOsCryptEncryptedKeyPrefName in [1]
base64_key = data['os_crypt']['encrypted_key']
except KeyError:
logger.error('no encrypted key in Local State')
return None
encrypted_key = base64.b64decode(base64_key)
# kDPAPIKeyPrefix in [1]
prefix = b'DPAPI'
if not encrypted_key.startswith(prefix):
logger.error('invalid key')
@@ -882,16 +999,18 @@ def _get_windows_v10_key(browser_root, logger):
def pbkdf2_sha1(password, salt, iterations, key_length):
return pbkdf2_hmac('sha1', password, salt, iterations, key_length)
return hashlib.pbkdf2_hmac('sha1', password, salt, iterations, key_length)
def _decrypt_aes_cbc(ciphertext, key, logger, initialization_vector=b' ' * 16):
plaintext = unpad_pkcs7(aes_cbc_decrypt_bytes(ciphertext, key, initialization_vector))
try:
return plaintext.decode()
except UnicodeDecodeError:
logger.warning('failed to decrypt cookie (AES-CBC) because UTF-8 decoding failed. Possibly the key is wrong?', only_once=True)
return None
def _decrypt_aes_cbc_multi(ciphertext, keys, logger, initialization_vector=b' ' * 16):
for key in keys:
plaintext = unpad_pkcs7(aes_cbc_decrypt_bytes(ciphertext, key, initialization_vector))
try:
return plaintext.decode()
except UnicodeDecodeError:
pass
logger.warning('failed to decrypt cookie (AES-CBC) because UTF-8 decoding failed. Possibly the key is wrong?', only_once=True)
return None
def _decrypt_aes_gcm(ciphertext, key, nonce, authentication_tag, logger):
@@ -931,7 +1050,7 @@ def _decrypt_windows_dpapi(ciphertext, logger):
None, # pvReserved: must be NULL
None, # pPromptStruct: information about prompts to display
0, # dwFlags
ctypes.byref(blob_out) # pDataOut
ctypes.byref(blob_out), # pDataOut
)
if not ret:
logger.warning('failed to decrypt with DPAPI', only_once=True)
@@ -959,17 +1078,20 @@ def _get_column_names(cursor, table_name):
return [row[1].decode() for row in table_info]
def _find_most_recently_used_file(root, filename, logger):
def _newest(files):
return max(files, key=lambda path: os.lstat(path).st_mtime, default=None)
def _find_files(root, filename, logger):
# if there are multiple browser profiles, take the most recently used one
i, paths = 0, []
i = 0
with _create_progress_bar(logger) as progress_bar:
for curr_root, dirs, files in os.walk(root):
for curr_root, _, files in os.walk(root):
for file in files:
i += 1
progress_bar.print(f'Searching for "{filename}": {i: 6d} files searched')
if file == filename:
paths.append(os.path.join(curr_root, file))
return None if not paths else max(paths, key=lambda path: os.lstat(path).st_mtime)
yield os.path.join(curr_root, file)
def _merge_cookie_jars(jars):
@@ -983,7 +1105,7 @@ def _merge_cookie_jars(jars):
def _is_path(value):
return os.path.sep in value
return any(sep in value for sep in (os.path.sep, os.path.altsep) if sep)
def _parse_browser_specification(browser_name, profile=None, keyring=None, container=None):
@@ -1004,24 +1126,24 @@ class LenientSimpleCookie(http.cookies.SimpleCookie):
_LEGAL_VALUE_CHARS = _LEGAL_KEY_CHARS + re.escape('(),/<=>?@[]{}')
_RESERVED = {
"expires",
"path",
"comment",
"domain",
"max-age",
"secure",
"httponly",
"version",
"samesite",
'expires',
'path',
'comment',
'domain',
'max-age',
'secure',
'httponly',
'version',
'samesite',
}
_FLAGS = {"secure", "httponly"}
_FLAGS = {'secure', 'httponly'}
# Added 'bad' group to catch the remaining value
_COOKIE_PATTERN = re.compile(r"""
_COOKIE_PATTERN = re.compile(r'''
\s* # Optional whitespace at start of cookie
(?P<key> # Start of group 'key'
[""" + _LEGAL_KEY_CHARS + r"""]+?# Any word of at least one letter
[''' + _LEGAL_KEY_CHARS + r''']+?# Any word of at least one letter
) # End of group 'key'
( # Optional group: there may not be a value.
\s*=\s* # Equal Sign
@@ -1031,7 +1153,7 @@ class LenientSimpleCookie(http.cookies.SimpleCookie):
| # or
\w{3},\s[\w\d\s-]{9,11}\s[\d:]{8}\sGMT # Special case for "expires" attr
| # or
[""" + _LEGAL_VALUE_CHARS + r"""]* # Any word or empty string
[''' + _LEGAL_VALUE_CHARS + r''']* # Any word or empty string
) # End of group 'val'
| # or
(?P<bad>(?:\\;|[^;])*?) # 'bad' group fallback for invalid values
@@ -1039,7 +1161,7 @@ class LenientSimpleCookie(http.cookies.SimpleCookie):
)? # End of optional value group
\s* # Any number of spaces.
(\s+|;|$) # Ending either at space, semicolon, or EOS.
""", re.ASCII | re.VERBOSE)
''', re.ASCII | re.VERBOSE)
def load(self, data):
# Workaround for https://github.com/yt-dlp/yt-dlp/issues/4776
@@ -1085,3 +1207,150 @@ class LenientSimpleCookie(http.cookies.SimpleCookie):
else:
morsel = None
class YoutubeDLCookieJar(http.cookiejar.MozillaCookieJar):
"""
See [1] for cookie file format.
1. https://curl.haxx.se/docs/http-cookies.html
"""
_HTTPONLY_PREFIX = '#HttpOnly_'
_ENTRY_LEN = 7
_HEADER = '''# Netscape HTTP Cookie File
# This file is generated by yt-dlp. Do not edit.
'''
_CookieFileEntry = collections.namedtuple(
'CookieFileEntry',
('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
def __init__(self, filename=None, *args, **kwargs):
super().__init__(None, *args, **kwargs)
if is_path_like(filename):
filename = os.fspath(filename)
self.filename = filename
@staticmethod
def _true_or_false(cndn):
return 'TRUE' if cndn else 'FALSE'
@contextlib.contextmanager
def open(self, file, *, write=False):
if is_path_like(file):
with open(file, 'w' if write else 'r', encoding='utf-8') as f:
yield f
else:
if write:
file.truncate(0)
yield file
def _really_save(self, f, ignore_discard, ignore_expires):
now = time.time()
for cookie in self:
if (not ignore_discard and cookie.discard
or not ignore_expires and cookie.is_expired(now)):
continue
name, value = cookie.name, cookie.value
if value is None:
# cookies.txt regards 'Set-Cookie: foo' as a cookie
# with no name, whereas http.cookiejar regards it as a
# cookie with no value.
name, value = '', name
f.write('{}\n'.format('\t'.join((
cookie.domain,
self._true_or_false(cookie.domain.startswith('.')),
cookie.path,
self._true_or_false(cookie.secure),
str_or_none(cookie.expires, default=''),
name, value,
))))
def save(self, filename=None, ignore_discard=True, ignore_expires=True):
"""
Save cookies to a file.
Code is taken from CPython 3.6
https://github.com/python/cpython/blob/8d999cbf4adea053be6dbb612b9844635c4dfb8e/Lib/http/cookiejar.py#L2091-L2117 """
if filename is None:
if self.filename is not None:
filename = self.filename
else:
raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT)
# Store session cookies with `expires` set to 0 instead of an empty string
for cookie in self:
if cookie.expires is None:
cookie.expires = 0
with self.open(filename, write=True) as f:
f.write(self._HEADER)
self._really_save(f, ignore_discard, ignore_expires)
def load(self, filename=None, ignore_discard=True, ignore_expires=True):
"""Load cookies from a file."""
if filename is None:
if self.filename is not None:
filename = self.filename
else:
raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT)
def prepare_line(line):
if line.startswith(self._HTTPONLY_PREFIX):
line = line[len(self._HTTPONLY_PREFIX):]
# comments and empty lines are fine
if line.startswith('#') or not line.strip():
return line
cookie_list = line.split('\t')
if len(cookie_list) != self._ENTRY_LEN:
raise http.cookiejar.LoadError(f'invalid length {len(cookie_list)}')
cookie = self._CookieFileEntry(*cookie_list)
if cookie.expires_at and not cookie.expires_at.isdigit():
raise http.cookiejar.LoadError(f'invalid expires at {cookie.expires_at}')
return line
cf = io.StringIO()
with self.open(filename) as f:
for line in f:
try:
cf.write(prepare_line(line))
except http.cookiejar.LoadError as e:
if f'{line.strip()} '[0] in '[{"':
raise http.cookiejar.LoadError(
'Cookies file must be Netscape formatted, not JSON. See '
'https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp')
write_string(f'WARNING: skipping cookie file entry due to {e}: {line!r}\n')
continue
cf.seek(0)
self._really_load(cf, filename, ignore_discard, ignore_expires)
# Session cookies are denoted by either `expires` field set to
# an empty string or 0. MozillaCookieJar only recognizes the former
# (see [1]). So we need force the latter to be recognized as session
# cookies on our own.
# Session cookies may be important for cookies-based authentication,
# e.g. usually, when user does not check 'Remember me' check box while
# logging in on a site, some important cookies are stored as session
# cookies so that not recognizing them will result in failed login.
# 1. https://bugs.python.org/issue17164
for cookie in self:
# Treat `expires=0` cookies as session cookies
if cookie.expires == 0:
cookie.expires = None
cookie.discard = True
def get_cookie_header(self, url):
"""Generate a Cookie HTTP header for a given url"""
cookie_req = urllib.request.Request(normalize_url(sanitize_url(url)))
self.add_cookie_header(cookie_req)
return cookie_req.get_header('Cookie')
def get_cookies_for_url(self, url):
"""Generate a list of Cookie objects for a given url"""
# Policy `_now` attribute must be set before calling `_cookies_for_request`
# Ref: https://github.com/python/cpython/blob/3.7/Lib/http/cookiejar.py#L1360
self._policy._now = self._now = int(time.time())
return self._cookies_for_request(urllib.request.Request(normalize_url(sanitize_url(url))))
def clear(self, *args, **kwargs):
with contextlib.suppress(KeyError):
return super().clear(*args, **kwargs)
yt_dlp/dependencies/Cryptodome.py
+25 -17
View File
@@ -1,6 +1,3 @@
import types
from ..compat import functools
from ..compat.compat_utils import passthrough_module
try:
@@ -9,22 +6,33 @@ except ImportError:
try:
import Crypto as _parent
except (ImportError, SyntaxError): # Old Crypto gives SyntaxError in newer Python
_parent = types.ModuleType('no_Cryptodome')
_parent = passthrough_module(__name__, 'no_Cryptodome')
__bool__ = lambda: False
passthrough_module(__name__, _parent, (..., '__version__'))
del passthrough_module
__version__ = ''
AES = PKCS1_v1_5 = Blowfish = PKCS1_OAEP = SHA1 = CMAC = RSA = None
try:
if _parent.__name__ == 'Cryptodome':
from Cryptodome import __version__
from Cryptodome.Cipher import AES, PKCS1_OAEP, Blowfish, PKCS1_v1_5
from Cryptodome.Hash import CMAC, SHA1
from Cryptodome.PublicKey import RSA
elif _parent.__name__ == 'Crypto':
from Crypto import __version__
from Crypto.Cipher import AES, PKCS1_OAEP, Blowfish, PKCS1_v1_5 # noqa: F401
from Crypto.Hash import CMAC, SHA1 # noqa: F401
from Crypto.PublicKey import RSA # noqa: F401
except ImportError:
__version__ = f'broken {__version__}'.strip()
@property
@functools.cache
def _yt_dlp__identifier():
if _parent.__name__ == 'Crypto':
from Crypto.Cipher import AES
try:
# In pycrypto, mode defaults to ECB. See:
# https://www.pycryptodome.org/en/latest/src/vs_pycrypto.html#:~:text=not%20have%20ECB%20as%20default%20mode
AES.new(b'abcdefghijklmnop')
except TypeError:
return 'pycrypto'
return _parent.__name__
_yt_dlp__identifier = _parent.__name__
if AES and _yt_dlp__identifier == 'Crypto':
try:
# In pycrypto, mode defaults to ECB. See:
# https://www.pycryptodome.org/en/latest/src/vs_pycrypto.html#:~:text=not%20have%20ECB%20as%20default%20mode
AES.new(b'abcdefghijklmnop')
except TypeError:
_yt_dlp__identifier = 'pycrypto'
yt_dlp/dependencies/__init__.py
+18 -5
View File
@@ -43,19 +43,28 @@ except Exception as _err:
try:
import sqlite3
# We need to get the underlying `sqlite` version, see https://github.com/yt-dlp/yt-dlp/issues/8152
sqlite3._yt_dlp__version = sqlite3.sqlite_version
except ImportError:
# although sqlite3 is part of the standard library, it is possible to compile python without
# although sqlite3 is part of the standard library, it is possible to compile Python without
# sqlite support. See: https://github.com/yt-dlp/yt-dlp/issues/544
sqlite3 = None
try:
import websockets
except (ImportError, SyntaxError):
# websockets 3.10 on python 3.6 causes SyntaxError
# See https://github.com/yt-dlp/yt-dlp/issues/2633
except ImportError:
websockets = None
try:
import urllib3
except ImportError:
urllib3 = None
try:
import requests
except ImportError:
requests = None
try:
import xattr # xattr or pyxattr
@@ -65,6 +74,10 @@ else:
if hasattr(xattr, 'set'): # pyxattr
xattr._yt_dlp__identifier = 'pyxattr'
try:
import curl_cffi
except ImportError:
curl_cffi = None
from . import Cryptodome
@@ -73,7 +86,7 @@ available_dependencies = {k: v for k, v in all_dependencies.items() if v}
# Deprecated
Cryptodome_AES = Cryptodome.Cipher.AES if Cryptodome else None
Cryptodome_AES = Cryptodome.AES
__all__ = [
yt_dlp/downloader/__init__.py
+2 -1
View File
@@ -30,7 +30,7 @@ from .hls import HlsFD
from .http import HttpFD
from .ism import IsmFD
from .mhtml import MhtmlFD
from .niconico import NiconicoDmcFD
from .niconico import NiconicoDmcFD, NiconicoLiveFD
from .rtmp import RtmpFD
from .rtsp import RtspFD
from .websocket import WebSocketFragmentFD
@@ -50,6 +50,7 @@ PROTOCOL_MAP = {
'ism': IsmFD,
'mhtml': MhtmlFD,
'niconico_dmc': NiconicoDmcFD,
'niconico_live': NiconicoLiveFD,
'fc2_live': FC2LiveFD,
'websocket_frag': WebSocketFragmentFD,
'youtube_live_chat': YoutubeLiveChatFD,
yt_dlp/downloader/common.py
+37 -15
View File
@@ -4,6 +4,7 @@ import functools
import os
import random
import re
import threading
import time
from ..minicurses import (
@@ -49,10 +50,10 @@ class FileDownloader:
verbose: Print additional info to stdout.
quiet: Do not print messages to stdout.
ratelimit: Download speed limit, in bytes/sec.
continuedl: Attempt to continue downloads if possible
throttledratelimit: Assume the download is being throttled below this speed (bytes/sec)
retries: Number of times to retry for HTTP error 5xx
file_access_retries: Number of times to retry on file access error
retries: Number of times to retry for expected network errors.
Default is 0 for API, but 10 for CLI
file_access_retries: Number of times to retry on file access error (default: 3)
buffersize: Size of download buffer in bytes.
noresizebuffer: Do not automatically resize the download buffer.
continuedl: Try to continue downloads if possible.
@@ -63,6 +64,7 @@ class FileDownloader:
min_filesize: Skip files smaller than this size
max_filesize: Skip files larger than this size
xattr_set_filesize: Set ytdl.filesize user xattribute with expected size.
progress_delta: The minimum time between progress output, in seconds
external_downloader_args: A dictionary of downloader keys (in lower case)
and a list of additional command-line arguments for the
executable. Use 'default' as the name for arguments to be
@@ -88,6 +90,9 @@ class FileDownloader:
self.params = params
self._prepare_multiline_status()
self.add_progress_hook(self.report_progress)
if self.params.get('progress_delta'):
self._progress_delta_lock = threading.Lock()
self._progress_delta_time = time.monotonic()
def _set_ydl(self, ydl):
self.ydl = ydl
@@ -138,17 +143,21 @@ class FileDownloader:
def format_percent(percent):
return ' N/A%' if percent is None else f'{percent:>5.1f}%'
@staticmethod
def calc_eta(start, now, total, current):
@classmethod
def calc_eta(cls, start_or_rate, now_or_remaining, total=NO_DEFAULT, current=NO_DEFAULT):
if total is NO_DEFAULT:
rate, remaining = start_or_rate, now_or_remaining
if None in (rate, remaining):
return None
return int(float(remaining) / rate)
start, now = start_or_rate, now_or_remaining
if total is None:
return None
if now is None:
now = time.time()
dif = now - start
if current == 0 or dif < 0.001: # One millisecond
return None
rate = float(current) / dif
return int((float(total) - float(current)) / rate)
rate = cls.calc_speed(start, now, current)
return rate and int((float(total) - float(current)) / rate)
@staticmethod
def calc_speed(start, now, bytes):
@@ -165,6 +174,12 @@ class FileDownloader:
def format_retries(retries):
return 'inf' if retries == float('inf') else int(retries)
@staticmethod
def filesize_or_none(unencoded_filename):
if os.path.isfile(unencoded_filename):
return os.path.getsize(unencoded_filename)
return 0
@staticmethod
def best_block_size(elapsed_time, bytes):
new_min = max(bytes / 2.0, 1.0)
@@ -225,7 +240,7 @@ class FileDownloader:
sleep_func=fd.params.get('retry_sleep_functions', {}).get('file_access'))
def wrapper(self, func, *args, **kwargs):
for retry in RetryManager(self.params.get('file_access_retries'), error_callback, fd=self):
for retry in RetryManager(self.params.get('file_access_retries', 3), error_callback, fd=self):
try:
return func(self, *args, **kwargs)
except OSError as err:
@@ -245,7 +260,8 @@ class FileDownloader:
@wrap_file_access('remove')
def try_remove(self, filename):
os.remove(filename)
if os.path.isfile(filename):
os.remove(filename)
@wrap_file_access('rename')
def try_rename(self, old_filename, new_filename):
@@ -285,7 +301,8 @@ class FileDownloader:
self._multiline = BreaklineStatusPrinter(self.ydl._out_files.out, lines)
else:
self._multiline = MultilinePrinter(self.ydl._out_files.out, lines, not self.params.get('quiet'))
self._multiline.allow_colors = self._multiline._HAVE_FULLCAP and not self.params.get('no_color')
self._multiline.allow_colors = self.ydl._allow_colors.out and self.ydl._allow_colors.out != 'no_color'
self._multiline._HAVE_FULLCAP = self.ydl._allow_colors.out
def _finish_multiline_status(self):
self._multiline.end()
@@ -354,6 +371,12 @@ class FileDownloader:
if s['status'] != 'downloading':
return
if update_delta := self.params.get('progress_delta'):
with self._progress_delta_lock:
if time.monotonic() < self._progress_delta_time:
return
self._progress_delta_time += update_delta
s.update({
'_eta_str': self.format_eta(s.get('eta')).strip(),
'_speed_str': self.format_speed(s.get('speed')),
@@ -381,7 +404,7 @@ class FileDownloader:
def report_resuming_byte(self, resume_len):
"""Report attempt to resume at given byte."""
self.to_screen('[download] Resuming download at byte %s' % resume_len)
self.to_screen(f'[download] Resuming download at byte {resume_len}')
def report_retry(self, err, count, retries, frag_index=NO_DEFAULT, fatal=True):
"""Report retry"""
@@ -407,7 +430,6 @@ class FileDownloader:
"""Download to a filename using the info from info_dict
Return True on success and False otherwise
"""
nooverwrites_and_exists = (
not self.params.get('overwrites', True)
and os.path.exists(encodeFilename(filename))
yt_dlp/downloader/dash.py
+7 -4
View File
@@ -15,12 +15,15 @@ class DashSegmentsFD(FragmentFD):
FD_NAME = 'dashsegments'
def real_download(self, filename, info_dict):
if info_dict.get('is_live') and set(info_dict['protocol'].split('+')) != {'http_dash_segments_generator'}:
self.report_error('Live DASH videos are not supported')
if 'http_dash_segments_generator' in info_dict['protocol'].split('+'):
real_downloader = None # No external FD can support --live-from-start
else:
if info_dict.get('is_live'):
self.report_error('Live DASH videos are not supported')
real_downloader = get_suitable_downloader(
info_dict, self.params, None, protocol='dash_frag_urls', to_stdout=(filename == '-'))
real_start = time.time()
real_downloader = get_suitable_downloader(
info_dict, self.params, None, protocol='dash_frag_urls', to_stdout=(filename == '-'))
requested_formats = [{**info_dict, **fmt} for fmt in info_dict.get('requested_formats', [])]
args = []
yt_dlp/downloader/external.py
+66 -28
View File
@@ -1,14 +1,16 @@
import enum
import functools
import json
import os.path
import os
import re
import subprocess
import sys
import tempfile
import time
import uuid
from .fragment import FragmentFD
from ..compat import functools
from ..networking import Request
from ..postprocessor.ffmpeg import EXT_TO_OUT_FORMATS, FFmpegPostProcessor
from ..utils import (
Popen,
@@ -23,9 +25,7 @@ from ..utils import (
encodeArgument,
encodeFilename,
find_available_port,
handle_youtubedl_headers,
remove_end,
sanitized_Request,
traverse_obj,
)
@@ -43,6 +43,7 @@ class ExternalFD(FragmentFD):
def real_download(self, filename, info_dict):
self.report_destination(filename)
tmpfilename = self.temp_name(filename)
self._cookies_tempfile = None
try:
started = time.time()
@@ -54,7 +55,10 @@ class ExternalFD(FragmentFD):
# correct and expected termination thus all postprocessing
# should take place
retval = 0
self.to_screen('[%s] Interrupted by user' % self.get_basename())
self.to_screen(f'[{self.get_basename()}] Interrupted by user')
finally:
if self._cookies_tempfile:
self.try_remove(self._cookies_tempfile)
if retval == 0:
status = {
@@ -104,7 +108,7 @@ class ExternalFD(FragmentFD):
return all((
not info_dict.get('to_stdout') or Features.TO_STDOUT in cls.SUPPORTED_FEATURES,
'+' not in info_dict['protocol'] or Features.MULTIPLE_FORMATS in cls.SUPPORTED_FEATURES,
not traverse_obj(info_dict, ('hls_aes', ...), 'extra_param_to_segment_url'),
not traverse_obj(info_dict, ('hls_aes', ...), 'extra_param_to_segment_url', 'extra_param_to_key_url'),
all(proto in cls.SUPPORTED_PROTOCOLS for proto in info_dict['protocol'].split('+')),
))
@@ -126,6 +130,16 @@ class ExternalFD(FragmentFD):
self.get_basename(), self.params.get('external_downloader_args'), self.EXE_NAME,
keys, *args, **kwargs)
def _write_cookies(self):
if not self.ydl.cookiejar.filename:
tmp_cookies = tempfile.NamedTemporaryFile(suffix='.cookies', delete=False)
tmp_cookies.close()
self._cookies_tempfile = tmp_cookies.name
self.to_screen(f'[download] Writing temporary cookies file to "{self._cookies_tempfile}"')
# real_download resets _cookies_tempfile; if it's None then save() will write to cookiejar.filename
self.ydl.cookiejar.save(self._cookies_tempfile)
return self.ydl.cookiejar.filename or self._cookies_tempfile
def _call_downloader(self, tmpfilename, info_dict):
""" Either overwrite this or implement _make_cmd """
cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
@@ -158,7 +172,7 @@ class ExternalFD(FragmentFD):
decrypt_fragment = self.decrypter(info_dict)
dest, _ = self.sanitize_open(tmpfilename, 'wb')
for frag_index, fragment in enumerate(info_dict['fragments']):
fragment_filename = '%s-Frag%d' % (tmpfilename, frag_index)
fragment_filename = f'{tmpfilename}-Frag{frag_index}'
try:
src, _ = self.sanitize_open(fragment_filename, 'rb')
except OSError as err:
@@ -172,11 +186,11 @@ class ExternalFD(FragmentFD):
if not self.params.get('keep_fragments', False):
self.try_remove(encodeFilename(fragment_filename))
dest.close()
self.try_remove(encodeFilename('%s.frag.urls' % tmpfilename))
self.try_remove(encodeFilename(f'{tmpfilename}.frag.urls'))
return 0
def _call_process(self, cmd, info_dict):
return Popen.run(cmd, text=True, stderr=subprocess.PIPE)
return Popen.run(cmd, text=True, stderr=subprocess.PIPE if self._CAPTURE_STDERR else None)
class CurlFD(ExternalFD):
@@ -185,6 +199,9 @@ class CurlFD(ExternalFD):
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '--location', '-o', tmpfilename, '--compressed']
cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url'])
if cookie_header:
cmd += ['--cookie', cookie_header]
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
cmd += ['--header', f'{key}: {val}']
@@ -215,6 +232,9 @@ class AxelFD(ExternalFD):
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
cmd += ['-H', f'{key}: {val}']
cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url'])
if cookie_header:
cmd += ['-H', f'Cookie: {cookie_header}', '--max-redirect=0']
cmd += self._configuration_args()
cmd += ['--', info_dict['url']]
return cmd
@@ -224,7 +244,9 @@ class WgetFD(ExternalFD):
AVAILABLE_OPT = '--version'
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies', '--compression=auto']
cmd = [self.exe, '-O', tmpfilename, '-nv', '--compression=auto']
if self.ydl.cookiejar.get_cookie_header(info_dict['url']):
cmd += ['--load-cookies', self._write_cookies()]
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
cmd += ['--header', f'{key}: {val}']
@@ -272,7 +294,7 @@ class Aria2cFD(ExternalFD):
return super()._call_downloader(tmpfilename, info_dict)
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '-c',
cmd = [self.exe, '-c', '--no-conf',
'--console-log-level=warn', '--summary-interval=0', '--download-result=hide',
'--http-accept-gzip=true', '--file-allocation=none', '-x16', '-j16', '-s16']
if 'fragments' in info_dict:
@@ -280,6 +302,8 @@ class Aria2cFD(ExternalFD):
else:
cmd += ['--min-split-size', '1M']
if self.ydl.cookiejar.get_cookie_header(info_dict['url']):
cmd += [f'--load-cookies={self._write_cookies()}']
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
cmd += ['--header', f'{key}: {val}']
@@ -311,12 +335,12 @@ class Aria2cFD(ExternalFD):
cmd += ['--auto-file-renaming=false']
if 'fragments' in info_dict:
cmd += ['--file-allocation=none', '--uri-selector=inorder']
url_list_file = '%s.frag.urls' % tmpfilename
cmd += ['--uri-selector=inorder']
url_list_file = f'{tmpfilename}.frag.urls'
url_list = []
for frag_index, fragment in enumerate(info_dict['fragments']):
fragment_filename = '%s-Frag%d' % (os.path.basename(tmpfilename), frag_index)
url_list.append('%s\n\tout=%s' % (fragment['url'], self._aria2c_filename(fragment_filename)))
fragment_filename = f'{os.path.basename(tmpfilename)}-Frag{frag_index}'
url_list.append('{}\n\tout={}'.format(fragment['url'], self._aria2c_filename(fragment_filename)))
stream, _ = self.sanitize_open(url_list_file, 'wb')
stream.write('\n'.join(url_list).encode())
stream.close()
@@ -333,14 +357,13 @@ class Aria2cFD(ExternalFD):
'id': sanitycheck,
'method': method,
'params': [f'token:{rpc_secret}', *params],
}).encode('utf-8')
request = sanitized_Request(
}).encode()
request = Request(
f'http://localhost:{rpc_port}/jsonrpc',
data=d, headers={
'Content-Type': 'application/json',
'Content-Length': f'{len(d)}',
'Ytdl-request-proxy': '__noproxy__',
})
}, proxies={'all': None})
with self.ydl.urlopen(request) as r:
resp = json.load(r)
assert resp.get('id') == sanitycheck, 'Something went wrong with RPC server'
@@ -393,7 +416,7 @@ class Aria2cFD(ExternalFD):
'total_bytes_estimate': total,
'eta': (total - downloaded) / (speed or 1),
'fragment_index': min(frag_count, len(completed) + 1) if fragmented else None,
'elapsed': time.time() - started
'elapsed': time.time() - started,
})
self._hook_progress(status, info_dict)
@@ -418,6 +441,14 @@ class HttpieFD(ExternalFD):
if info_dict.get('http_headers') is not None:
for key, val in info_dict['http_headers'].items():
cmd += [f'{key}:{val}']
# httpie 3.1.0+ removes the Cookie header on redirect, so this should be safe for now. [1]
# If we ever need cookie handling for redirects, we can export the cookiejar into a session. [2]
# 1: https://github.com/httpie/httpie/security/advisories/GHSA-9w4w-cpc8-h2fq
# 2: https://httpie.io/docs/cli/sessions
cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url'])
if cookie_header:
cmd += [f'Cookie:{cookie_header}']
return cmd
@@ -460,7 +491,7 @@ class FFmpegFD(ExternalFD):
if not self.params.get('verbose'):
args += ['-hide_banner']
args += traverse_obj(info_dict, ('downloader_options', 'ffmpeg_args'), default=[])
args += traverse_obj(info_dict, ('downloader_options', 'ffmpeg_args', ...))
# These exists only for compatibility. Extractors should use
# info_dict['downloader_options']['ffmpeg_args'] instead
@@ -478,12 +509,12 @@ class FFmpegFD(ExternalFD):
proxy = self.params.get('proxy')
if proxy:
if not re.match(r'^[\da-zA-Z]+://', proxy):
proxy = 'http://%s' % proxy
proxy = f'http://{proxy}'
if proxy.startswith('socks'):
self.report_warning(
'%s does not support SOCKS proxies. Downloading is likely to fail. '
'Consider adding --hls-prefer-native to your command.' % self.get_basename())
f'{self.get_basename()} does not support SOCKS proxies. Downloading is likely to fail. '
'Consider adding --hls-prefer-native to your command.')
# Since December 2015 ffmpeg supports -http_proxy option (see
# http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd)
@@ -528,18 +559,23 @@ class FFmpegFD(ExternalFD):
selected_formats = info_dict.get('requested_formats') or [info_dict]
for i, fmt in enumerate(selected_formats):
if fmt.get('http_headers') and re.match(r'^https?://', fmt['url']):
headers_dict = handle_youtubedl_headers(fmt['http_headers'])
is_http = re.match(r'^https?://', fmt['url'])
cookies = self.ydl.cookiejar.get_cookies_for_url(fmt['url']) if is_http else []
if cookies:
args.extend(['-cookies', ''.join(
f'{cookie.name}={cookie.value}; path={cookie.path}; domain={cookie.domain};\r\n'
for cookie in cookies)])
if fmt.get('http_headers') and is_http:
# Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv:
# [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header.
args.extend(['-headers', ''.join(f'{key}: {val}\r\n' for key, val in headers_dict.items())])
args.extend(['-headers', ''.join(f'{key}: {val}\r\n' for key, val in fmt['http_headers'].items())])
if start_time:
args += ['-ss', str(start_time)]
if end_time:
args += ['-t', str(end_time - start_time)]
args += self._configuration_args((f'_i{i + 1}', '_i')) + ['-i', fmt['url']]
args += [*self._configuration_args((f'_i{i + 1}', '_i')), '-i', fmt['url']]
if not (start_time or end_time) or not self.params.get('force_keyframes_at_cuts'):
args += ['-c', 'copy']
@@ -579,6 +615,8 @@ class FFmpegFD(ExternalFD):
else:
args += ['-f', EXT_TO_OUT_FORMATS.get(ext, ext)]
args += traverse_obj(info_dict, ('downloader_options', 'ffmpeg_args_out', ...))
args += self._configuration_args(('_o1', '_o', ''))
args = [encodeArgument(opt) for opt in args]
yt_dlp/downloader/f4m.py
+15 -15
View File
@@ -3,11 +3,11 @@ import io
import itertools
import struct
import time
import urllib.error
import urllib.parse
from .fragment import FragmentFD
from ..compat import compat_etree_fromstring
from ..networking.exceptions import HTTPError
from ..utils import fix_xml_ampersands, xpath_text
@@ -67,12 +67,12 @@ class FlvReader(io.BytesIO):
self.read_bytes(3)
quality_entry_count = self.read_unsigned_char()
# QualityEntryCount
for i in range(quality_entry_count):
for _ in range(quality_entry_count):
self.read_string()
segment_run_count = self.read_unsigned_int()
segments = []
for i in range(segment_run_count):
for _ in range(segment_run_count):
first_segment = self.read_unsigned_int()
fragments_per_segment = self.read_unsigned_int()
segments.append((first_segment, fragments_per_segment))
@@ -91,12 +91,12 @@ class FlvReader(io.BytesIO):
quality_entry_count = self.read_unsigned_char()
# QualitySegmentUrlModifiers
for i in range(quality_entry_count):
for _ in range(quality_entry_count):
self.read_string()
fragments_count = self.read_unsigned_int()
fragments = []
for i in range(fragments_count):
for _ in range(fragments_count):
first = self.read_unsigned_int()
first_ts = self.read_unsigned_long_long()
duration = self.read_unsigned_int()
@@ -135,11 +135,11 @@ class FlvReader(io.BytesIO):
self.read_string() # MovieIdentifier
server_count = self.read_unsigned_char()
# ServerEntryTable
for i in range(server_count):
for _ in range(server_count):
self.read_string()
quality_count = self.read_unsigned_char()
# QualityEntryTable
for i in range(quality_count):
for _ in range(quality_count):
self.read_string()
# DrmData
self.read_string()
@@ -148,14 +148,14 @@ class FlvReader(io.BytesIO):
segments_count = self.read_unsigned_char()
segments = []
for i in range(segments_count):
for _ in range(segments_count):
box_size, box_type, box_data = self.read_box_info()
assert box_type == b'asrt'
segment = FlvReader(box_data).read_asrt()
segments.append(segment)
fragments_run_count = self.read_unsigned_char()
fragments = []
for i in range(fragments_run_count):
for _ in range(fragments_run_count):
box_size, box_type, box_data = self.read_box_info()
assert box_type == b'afrt'
fragments.append(FlvReader(box_data).read_afrt())
@@ -309,10 +309,10 @@ class F4mFD(FragmentFD):
def real_download(self, filename, info_dict):
man_url = info_dict['url']
requested_bitrate = info_dict.get('tbr')
self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME)
self.to_screen(f'[{self.FD_NAME}] Downloading f4m manifest')
urlh = self.ydl.urlopen(self._prepare_url(info_dict, man_url))
man_url = urlh.geturl()
man_url = urlh.url
# Some manifests may be malformed, e.g. prosiebensat1 generated manifests
# (see https://github.com/ytdl-org/youtube-dl/issues/6215#issuecomment-121704244
# and https://github.com/ytdl-org/youtube-dl/issues/7823)
@@ -326,8 +326,8 @@ class F4mFD(FragmentFD):
formats = sorted(formats, key=lambda f: f[0])
rate, media = formats[-1]
else:
rate, media = list(filter(
lambda f: int(f[0]) == requested_bitrate, formats))[0]
rate, media = next(filter(
lambda f: int(f[0]) == requested_bitrate, formats))
# Prefer baseURL for relative URLs as per 11.2 of F4M 3.0 spec.
man_base_url = get_base_url(doc) or man_url
@@ -407,8 +407,8 @@ class F4mFD(FragmentFD):
if box_type == b'mdat':
self._append_fragment(ctx, box_data)
break
except urllib.error.HTTPError as err:
if live and (err.code == 404 or err.code == 410):
except HTTPError as err:
if live and (err.status == 404 or err.status == 410):
# We didn't keep up with the live window. Continue
# with the next available fragment.
msg = 'Fragment %d unavailable' % frag_i
yt_dlp/downloader/fragment.py
+55 -61
View File
@@ -1,24 +1,20 @@
import concurrent.futures
import contextlib
import http.client
import json
import math
import os
import struct
import time
import urllib.error
from .common import FileDownloader
from .http import HttpFD
from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
from ..compat import compat_os_name
from ..utils import (
DownloadError,
RetryManager,
encodeFilename,
sanitized_Request,
traverse_obj,
)
from ..networking import Request
from ..networking.exceptions import HTTPError, IncompleteRead
from ..utils import DownloadError, RetryManager, encodeFilename, traverse_obj
from ..utils.networking import HTTPHeaderDict
from ..utils.progress import ProgressCalculator
class HttpQuietDownloader(HttpFD):
@@ -34,8 +30,8 @@ class FragmentFD(FileDownloader):
Available options:
fragment_retries: Number of times to retry a fragment for HTTP error (DASH
and hlsnative only)
fragment_retries: Number of times to retry a fragment for HTTP error
(DASH and hlsnative only). Default is 0 for API, but 10 for CLI
skip_unavailable_fragments:
Skip unavailable fragments (DASH and hlsnative only)
keep_fragments: Keep downloaded fragments on disk after downloading is
@@ -75,7 +71,7 @@ class FragmentFD(FileDownloader):
def _prepare_url(self, info_dict, url):
headers = info_dict.get('http_headers')
return sanitized_Request(url, None, headers) if headers else url
return Request(url, None, headers) if headers else url
def _prepare_and_start_frag_download(self, ctx, info_dict):
self._prepare_frag_download(ctx)
@@ -121,6 +117,11 @@ class FragmentFD(FileDownloader):
'request_data': request_data,
'ctx_id': ctx.get('ctx_id'),
}
frag_resume_len = 0
if ctx['dl'].params.get('continuedl', True):
frag_resume_len = self.filesize_or_none(self.temp_name(fragment_filename))
fragment_info_dict['frag_resume_len'] = ctx['frag_resume_len'] = frag_resume_len
success, _ = ctx['dl'].download(fragment_filename, fragment_info_dict)
if not success:
return False
@@ -155,9 +156,7 @@ class FragmentFD(FileDownloader):
del ctx['fragment_filename_sanitized']
def _prepare_frag_download(self, ctx):
if 'live' not in ctx:
ctx['live'] = False
if not ctx['live']:
if not ctx.setdefault('live', False):
total_frags_str = '%d' % ctx['total_frags']
ad_frags = ctx.get('ad_frags', 0)
if ad_frags:
@@ -170,15 +169,17 @@ class FragmentFD(FileDownloader):
**self.params,
'noprogress': True,
'test': False,
'sleep_interval': 0,
'max_sleep_interval': 0,
'sleep_interval_subtitles': 0,
})
tmpfilename = self.temp_name(ctx['filename'])
open_mode = 'wb'
resume_len = 0
# Establish possible resume length
if os.path.isfile(encodeFilename(tmpfilename)):
resume_len = self.filesize_or_none(tmpfilename)
if resume_len > 0:
open_mode = 'ab'
resume_len = os.path.getsize(encodeFilename(tmpfilename))
# Should be initialized before ytdl file check
ctx.update({
@@ -187,7 +188,9 @@ class FragmentFD(FileDownloader):
})
if self.__do_ytdl_file(ctx):
if os.path.isfile(encodeFilename(self.ytdl_filename(ctx['filename']))):
ytdl_file_exists = os.path.isfile(encodeFilename(self.ytdl_filename(ctx['filename'])))
continuedl = self.params.get('continuedl', True)
if continuedl and ytdl_file_exists:
self._read_ytdl_file(ctx)
is_corrupt = ctx.get('ytdl_corrupt') is True
is_inconsistent = ctx['fragment_index'] > 0 and resume_len == 0
@@ -196,12 +199,17 @@ class FragmentFD(FileDownloader):
'.ytdl file is corrupt' if is_corrupt else
'Inconsistent state of incomplete fragment download')
self.report_warning(
'%s. Restarting from the beginning ...' % message)
f'{message}. Restarting from the beginning ...')
ctx['fragment_index'] = resume_len = 0
if 'ytdl_corrupt' in ctx:
del ctx['ytdl_corrupt']
self._write_ytdl_file(ctx)
else:
if not continuedl:
if ytdl_file_exists:
self._read_ytdl_file(ctx)
ctx['fragment_index'] = resume_len = 0
self._write_ytdl_file(ctx)
assert ctx['fragment_index'] == 0
@@ -219,8 +227,7 @@ class FragmentFD(FileDownloader):
resume_len = ctx['complete_frags_downloaded_bytes']
total_frags = ctx['total_frags']
ctx_id = ctx.get('ctx_id')
# This dict stores the download progress, it's updated by the progress
# hook
# Stores the download progress, updated by the progress hook
state = {
'status': 'downloading',
'downloaded_bytes': resume_len,
@@ -230,14 +237,8 @@ class FragmentFD(FileDownloader):
'tmpfilename': ctx['tmpfilename'],
}
start = time.time()
ctx.update({
'started': start,
'fragment_started': start,
# Amount of fragment's bytes downloaded by the time of the previous
# frag progress hook invocation
'prev_frag_downloaded_bytes': 0,
})
ctx['started'] = time.time()
progress = ProgressCalculator(resume_len)
def frag_progress_hook(s):
if s['status'] not in ('downloading', 'finished'):
@@ -252,52 +253,45 @@ class FragmentFD(FileDownloader):
state['max_progress'] = ctx.get('max_progress')
state['progress_idx'] = ctx.get('progress_idx')
time_now = time.time()
state['elapsed'] = time_now - start
state['elapsed'] = progress.elapsed
frag_total_bytes = s.get('total_bytes') or 0
s['fragment_info_dict'] = s.pop('info_dict', {})
# XXX: Fragment resume is not accounted for here
if not ctx['live']:
estimated_size = (
(ctx['complete_frags_downloaded_bytes'] + frag_total_bytes)
/ (state['fragment_index'] + 1) * total_frags)
state['total_bytes_estimate'] = estimated_size
progress.total = estimated_size
progress.update(s.get('downloaded_bytes'))
state['total_bytes_estimate'] = progress.total
else:
progress.update(s.get('downloaded_bytes'))
if s['status'] == 'finished':
state['fragment_index'] += 1
ctx['fragment_index'] = state['fragment_index']
state['downloaded_bytes'] += frag_total_bytes - ctx['prev_frag_downloaded_bytes']
ctx['complete_frags_downloaded_bytes'] = state['downloaded_bytes']
ctx['speed'] = state['speed'] = self.calc_speed(
ctx['fragment_started'], time_now, frag_total_bytes)
ctx['fragment_started'] = time.time()
ctx['prev_frag_downloaded_bytes'] = 0
else:
frag_downloaded_bytes = s['downloaded_bytes']
state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes']
if not ctx['live']:
state['eta'] = self.calc_eta(
start, time_now, estimated_size - resume_len,
state['downloaded_bytes'] - resume_len)
ctx['speed'] = state['speed'] = self.calc_speed(
ctx['fragment_started'], time_now, frag_downloaded_bytes)
ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes
progress.thread_reset()
state['downloaded_bytes'] = ctx['complete_frags_downloaded_bytes'] = progress.downloaded
state['speed'] = ctx['speed'] = progress.speed.smooth
state['eta'] = progress.eta.smooth
self._hook_progress(state, info_dict)
ctx['dl'].add_progress_hook(frag_progress_hook)
return start
return ctx['started']
def _finish_frag_download(self, ctx, info_dict):
ctx['dest_stream'].close()
if self.__do_ytdl_file(ctx):
ytdl_filename = encodeFilename(self.ytdl_filename(ctx['filename']))
if os.path.isfile(ytdl_filename):
self.try_remove(ytdl_filename)
self.try_remove(self.ytdl_filename(ctx['filename']))
elapsed = time.time() - ctx['started']
to_file = ctx['tmpfilename'] != '-'
if to_file:
downloaded_bytes = os.path.getsize(encodeFilename(ctx['tmpfilename']))
downloaded_bytes = self.filesize_or_none(ctx['tmpfilename'])
else:
downloaded_bytes = ctx['complete_frags_downloaded_bytes']
@@ -372,10 +366,10 @@ class FragmentFD(FileDownloader):
return decrypt_fragment
def download_and_append_fragments_multiple(self, *args, **kwargs):
'''
"""
@params (ctx1, fragments1, info_dict1), (ctx2, fragments2, info_dict2), ...
all args must be either tuple or list
'''
"""
interrupt_trigger = [True]
max_progress = len(args)
if max_progress == 1:
@@ -430,7 +424,7 @@ class FragmentFD(FileDownloader):
finally:
tpe.shutdown(wait=True)
if not interrupt_trigger[0] and not is_live:
raise KeyboardInterrupt()
raise KeyboardInterrupt
# we expect the user wants to stop and DO WANT the preceding postprocessors to run;
# so returning a intermediate result here instead of KeyboardInterrupt on live
return result
@@ -449,7 +443,7 @@ class FragmentFD(FileDownloader):
frag_index = ctx['fragment_index'] = fragment['frag_index']
ctx['last_error'] = None
headers = info_dict.get('http_headers', {}).copy()
headers = HTTPHeaderDict(info_dict.get('http_headers'))
byte_range = fragment.get('byte_range')
if byte_range:
headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
@@ -466,9 +460,10 @@ class FragmentFD(FileDownloader):
for retry in RetryManager(self.params.get('fragment_retries'), error_callback):
try:
ctx['fragment_count'] = fragment.get('fragment_count')
if not self._download_fragment(ctx, fragment['url'], info_dict, headers):
if not self._download_fragment(
ctx, fragment['url'], info_dict, headers, info_dict.get('request_data')):
return
except (urllib.error.HTTPError, http.client.IncompleteRead) as err:
except (HTTPError, IncompleteRead) as err:
retry.error = err
continue
except DownloadError: # has own retry settings
@@ -496,7 +491,6 @@ class FragmentFD(FileDownloader):
download_fragment(fragment, ctx_copy)
return fragment, fragment['frag_index'], ctx_copy.get('fragment_filename_sanitized')
self.report_warning('The download speed shown is only of one thread. This is a known issue and patches are welcome')
with tpe or concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
try:
for fragment, frag_index, frag_filename in pool.map(_download_fragment, fragments):
yt_dlp/downloader/hls.py
+45 -29
View File
@@ -28,7 +28,16 @@ class HlsFD(FragmentFD):
FD_NAME = 'hlsnative'
@staticmethod
def can_download(manifest, info_dict, allow_unplayable_formats=False):
def _has_drm(manifest): # TODO: https://github.com/yt-dlp/yt-dlp/pull/5039
return bool(re.search('|'.join((
r'#EXT-X-(?:SESSION-)?KEY:.*?URI="skd://', # Apple FairPlay
r'#EXT-X-(?:SESSION-)?KEY:.*?KEYFORMAT="com\.apple\.streamingkeydelivery"', # Apple FairPlay
r'#EXT-X-(?:SESSION-)?KEY:.*?KEYFORMAT="com\.microsoft\.playready"', # Microsoft PlayReady
r'#EXT-X-FAXS-CM:', # Adobe Flash Access
)), manifest))
@classmethod
def can_download(cls, manifest, info_dict, allow_unplayable_formats=False):
UNSUPPORTED_FEATURES = [
# r'#EXT-X-BYTERANGE', # playlists composed of byte ranges of media files [2]
@@ -50,27 +59,29 @@ class HlsFD(FragmentFD):
]
if not allow_unplayable_formats:
UNSUPPORTED_FEATURES += [
r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)', # encrypted streams [1]
r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)', # encrypted streams [1], but not necessarily DRM
]
def check_results():
yield not info_dict.get('is_live')
for feature in UNSUPPORTED_FEATURES:
yield not re.search(feature, manifest)
if not allow_unplayable_formats:
yield not cls._has_drm(manifest)
return all(check_results())
def real_download(self, filename, info_dict):
man_url = info_dict['url']
self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME)
self.to_screen(f'[{self.FD_NAME}] Downloading m3u8 manifest')
urlh = self.ydl.urlopen(self._prepare_url(info_dict, man_url))
man_url = urlh.geturl()
man_url = urlh.url
s = urlh.read().decode('utf-8', 'ignore')
can_download, message = self.can_download(s, info_dict, self.params.get('allow_unplayable_formats')), None
if can_download:
has_ffmpeg = FFmpegFD.available()
no_crypto = not Cryptodome and '#EXT-X-KEY:METHOD=AES-128' in s
no_crypto = not Cryptodome.AES and '#EXT-X-KEY:METHOD=AES-128' in s
if no_crypto and has_ffmpeg:
can_download, message = False, 'The stream has AES-128 encryption and pycryptodomex is not available'
elif no_crypto:
@@ -81,14 +92,13 @@ class HlsFD(FragmentFD):
message = ('Live HLS streams are not supported by the native downloader. If this is a livestream, '
f'please {install_ffmpeg}add "--downloader ffmpeg --hls-use-mpegts" to your command')
if not can_download:
has_drm = re.search('|'.join([
r'#EXT-X-FAXS-CM:', # Adobe Flash Access
r'#EXT-X-(?:SESSION-)?KEY:.*?URI="skd://', # Apple FairPlay
]), s)
if has_drm and not self.params.get('allow_unplayable_formats'):
self.report_error(
'This video is DRM protected; Try selecting another format with --format or '
'add --check-formats to automatically fallback to the next best format')
if self._has_drm(s) and not self.params.get('allow_unplayable_formats'):
if info_dict.get('has_drm') and self.params.get('test'):
self.to_screen(f'[{self.FD_NAME}] This format is DRM protected', skip_eol=True)
else:
self.report_error(
'This format is DRM protected; Try selecting another format with --format or '
'add --check-formats to automatically fallback to the next best format', tb=False)
return False
message = message or 'Unsupported features have been detected'
fd = FFmpegFD(self.ydl, self.params)
@@ -150,10 +160,12 @@ class HlsFD(FragmentFD):
extra_state = ctx.setdefault('extra_state', {})
format_index = info_dict.get('format_index')
extra_query = None
extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
if extra_param_to_segment_url:
extra_query = urllib.parse.parse_qs(extra_param_to_segment_url)
extra_segment_query = None
if extra_param_to_segment_url := info_dict.get('extra_param_to_segment_url'):
extra_segment_query = urllib.parse.parse_qs(extra_param_to_segment_url)
extra_key_query = None
if extra_param_to_key_url := info_dict.get('extra_param_to_key_url'):
extra_key_query = urllib.parse.parse_qs(extra_param_to_key_url)
i = 0
media_sequence = 0
decrypt_info = {'METHOD': 'NONE'}
@@ -180,8 +192,8 @@ class HlsFD(FragmentFD):
if frag_index <= ctx['fragment_index']:
continue
frag_url = urljoin(man_url, line)
if extra_query:
frag_url = update_url_query(frag_url, extra_query)
if extra_segment_query:
frag_url = update_url_query(frag_url, extra_segment_query)
fragments.append({
'frag_index': frag_index,
@@ -202,8 +214,8 @@ class HlsFD(FragmentFD):
frag_index += 1
map_info = parse_m3u8_attributes(line[11:])
frag_url = urljoin(man_url, map_info.get('URI'))
if extra_query:
frag_url = update_url_query(frag_url, extra_query)
if extra_segment_query:
frag_url = update_url_query(frag_url, extra_segment_query)
if map_info.get('BYTERANGE'):
splitted_byte_range = map_info.get('BYTERANGE').split('@')
@@ -218,7 +230,7 @@ class HlsFD(FragmentFD):
'url': frag_url,
'decrypt_info': decrypt_info,
'byte_range': byte_range,
'media_sequence': media_sequence
'media_sequence': media_sequence,
})
media_sequence += 1
@@ -234,8 +246,10 @@ class HlsFD(FragmentFD):
decrypt_info['KEY'] = external_aes_key
else:
decrypt_info['URI'] = urljoin(man_url, decrypt_info['URI'])
if extra_query:
decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
if extra_key_query or extra_segment_query:
# Fall back to extra_segment_query to key for backwards compat
decrypt_info['URI'] = update_url_query(
decrypt_info['URI'], extra_key_query or extra_segment_query)
if decrypt_url != decrypt_info['URI']:
decrypt_info['KEY'] = None
@@ -340,9 +354,8 @@ class HlsFD(FragmentFD):
# XXX: this should probably be silent as well
# or verify that all segments contain the same data
self.report_warning(bug_reports_message(
'Discarding a %s block found in the middle of the stream; '
'if the subtitles display incorrectly,'
% (type(block).__name__)))
f'Discarding a {type(block).__name__} block found in the middle of the stream; '
'if the subtitles display incorrectly,'))
continue
block.write_into(output)
@@ -359,7 +372,10 @@ class HlsFD(FragmentFD):
return output.getvalue().encode()
self.download_and_append_fragments(
ctx, fragments, info_dict, pack_func=pack_fragment, finish_func=fin_fragments)
if len(fragments) == 1:
self.download_and_append_fragments(ctx, fragments, info_dict)
else:
self.download_and_append_fragments(
ctx, fragments, info_dict, pack_func=pack_fragment, finish_func=fin_fragments)
else:
return self.download_and_append_fragments(ctx, fragments, info_dict)
yt_dlp/downloader/http.py
+46 -53
View File
@@ -1,12 +1,14 @@
import http.client
import os
import random
import socket
import ssl
import time
import urllib.error
from .common import FileDownloader
from ..networking import Request
from ..networking.exceptions import (
CertificateVerifyError,
HTTPError,
TransportError,
)
from ..utils import (
ContentTooShortError,
RetryManager,
@@ -16,18 +18,10 @@ from ..utils import (
encodeFilename,
int_or_none,
parse_http_range,
sanitized_Request,
try_call,
write_xattr,
)
RESPONSE_READ_EXCEPTIONS = (
TimeoutError,
socket.timeout, # compat: py < 3.10
ConnectionError,
ssl.SSLError,
http.client.HTTPException
)
from ..utils.networking import HTTPHeaderDict
class HttpFD(FileDownloader):
@@ -45,11 +39,8 @@ class HttpFD(FileDownloader):
ctx.tmpfilename = self.temp_name(filename)
ctx.stream = None
# Do not include the Accept-Encoding header
headers = {'Youtubedl-no-compression': 'True'}
add_headers = info_dict.get('http_headers')
if add_headers:
headers.update(add_headers)
# Disable compression
headers = HTTPHeaderDict({'Accept-Encoding': 'identity'}, info_dict.get('http_headers'))
is_test = self.params.get('test', False)
chunk_size = self._TEST_FILE_SIZE if is_test else (
@@ -120,10 +111,10 @@ class HttpFD(FileDownloader):
if try_call(lambda: range_end >= ctx.content_len):
range_end = ctx.content_len - 1
request = sanitized_Request(url, request_data, headers)
request = Request(url, request_data, headers)
has_range = range_start is not None
if has_range:
request.add_header('Range', f'bytes={int(range_start)}-{int_or_none(range_end) or ""}')
request.headers['Range'] = f'bytes={int(range_start)}-{int_or_none(range_end) or ""}'
# Establish connection
try:
ctx.data = self.ydl.urlopen(request)
@@ -150,20 +141,21 @@ class HttpFD(FileDownloader):
# Content-Range is either not present or invalid. Assuming remote webserver is
# trying to send the whole file, resume is not possible, so wiping the local file
# and performing entire redownload
self.report_unable_to_resume()
elif range_start > 0:
self.report_unable_to_resume()
ctx.resume_len = 0
ctx.open_mode = 'wb'
ctx.data_len = ctx.content_len = int_or_none(ctx.data.info().get('Content-length', None))
except urllib.error.HTTPError as err:
if err.code == 416:
ctx.data_len = ctx.content_len = int_or_none(ctx.data.headers.get('Content-length', None))
except HTTPError as err:
if err.status == 416:
# Unable to resume (requested range not satisfiable)
try:
# Open the connection again without the range header
ctx.data = self.ydl.urlopen(
sanitized_Request(url, request_data, headers))
content_length = ctx.data.info()['Content-Length']
except urllib.error.HTTPError as err:
if err.code < 500 or err.code >= 600:
Request(url, request_data, headers))
content_length = ctx.data.headers['Content-Length']
except HTTPError as err:
if err.status < 500 or err.status >= 600:
raise
else:
# Examine the reported length
@@ -184,36 +176,32 @@ class HttpFD(FileDownloader):
'downloaded_bytes': ctx.resume_len,
'total_bytes': ctx.resume_len,
}, info_dict)
raise SucceedDownload()
raise SucceedDownload
else:
# The length does not match, we start the download over
self.report_unable_to_resume()
ctx.resume_len = 0
ctx.open_mode = 'wb'
return
elif err.code < 500 or err.code >= 600:
elif err.status < 500 or err.status >= 600:
# Unexpected HTTP error
raise
raise RetryDownload(err)
except urllib.error.URLError as err:
if isinstance(err.reason, ssl.CertificateError):
raise
raise RetryDownload(err)
# In urllib.request.AbstractHTTPHandler, the response is partially read on request.
# Any errors that occur during this will not be wrapped by URLError
except RESPONSE_READ_EXCEPTIONS as err:
except CertificateVerifyError:
raise
except TransportError as err:
raise RetryDownload(err)
def close_stream():
if ctx.stream is not None:
if not ctx.tmpfilename == '-':
if ctx.tmpfilename != '-':
ctx.stream.close()
ctx.stream = None
def download():
data_len = ctx.data.info().get('Content-length')
data_len = ctx.data.headers.get('Content-length')
if ctx.data.info().get('Content-encoding'):
if ctx.data.headers.get('Content-encoding'):
# Content-encoding is present, Content-length is not reliable anymore as we are
# doing auto decompression. (See: https://github.com/yt-dlp/yt-dlp/pull/6176)
data_len = None
@@ -249,15 +237,20 @@ class HttpFD(FileDownloader):
def retry(e):
close_stream()
ctx.resume_len = (byte_counter if ctx.tmpfilename == '-'
else os.path.getsize(encodeFilename(ctx.tmpfilename)))
if ctx.tmpfilename == '-':
ctx.resume_len = byte_counter
else:
try:
ctx.resume_len = os.path.getsize(encodeFilename(ctx.tmpfilename))
except FileNotFoundError:
ctx.resume_len = 0
raise RetryDownload(e)
while True:
try:
# Download and write
data_block = ctx.data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
except RESPONSE_READ_EXCEPTIONS as err:
except TransportError as err:
retry(err)
byte_counter += len(data_block)
@@ -275,20 +268,20 @@ class HttpFD(FileDownloader):
ctx.filename = self.undo_temp_name(ctx.tmpfilename)
self.report_destination(ctx.filename)
except OSError as err:
self.report_error('unable to open for writing: %s' % str(err))
self.report_error(f'unable to open for writing: {err}')
return False
if self.params.get('xattr_set_filesize', False) and data_len is not None:
try:
write_xattr(ctx.tmpfilename, 'user.ytdl.filesize', str(data_len).encode())
except (XAttrUnavailableError, XAttrMetadataError) as err:
self.report_error('unable to set filesize xattr: %s' % str(err))
self.report_error(f'unable to set filesize xattr: {err}')
try:
ctx.stream.write(data_block)
except OSError as err:
self.to_stderr('\n')
self.report_error('unable to write data: %s' % str(err))
self.report_error(f'unable to write data: {err}')
return False
# Apply rate limit
@@ -334,19 +327,19 @@ class HttpFD(FileDownloader):
elif now - ctx.throttle_start > 3:
if ctx.stream is not None and ctx.tmpfilename != '-':
ctx.stream.close()
raise ThrottledDownload()
raise ThrottledDownload
elif speed:
ctx.throttle_start = None
if not is_test and ctx.chunk_size and ctx.content_len is not None and byte_counter < ctx.content_len:
ctx.resume_len = byte_counter
# ctx.block_size = block_size
raise NextFragment()
if ctx.stream is None:
self.to_stderr('\n')
self.report_error('Did not get any data blocks')
return False
if not is_test and ctx.chunk_size and ctx.content_len is not None and byte_counter < ctx.content_len:
ctx.resume_len = byte_counter
raise NextFragment
if ctx.tmpfilename != '-':
ctx.stream.close()
@@ -358,7 +351,7 @@ class HttpFD(FileDownloader):
# Update file modification time
if self.params.get('updatetime', True):
info_dict['filetime'] = self.try_utime(ctx.filename, ctx.data.info().get('last-modified', None))
info_dict['filetime'] = self.try_utime(ctx.filename, ctx.data.headers.get('last-modified', None))
self._hook_progress({
'downloaded_bytes': byte_counter,
yt_dlp/downloader/ism.py
+3 -3
View File
@@ -2,9 +2,9 @@ import binascii
import io
import struct
import time
import urllib.error
from .fragment import FragmentFD
from ..networking.exceptions import HTTPError
from ..utils import RetryManager
u8 = struct.Struct('>B')
@@ -251,7 +251,7 @@ class IsmFD(FragmentFD):
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
frag_index = 0
for i, segment in enumerate(segments):
for segment in segments:
frag_index += 1
if frag_index <= ctx['fragment_index']:
continue
@@ -271,7 +271,7 @@ class IsmFD(FragmentFD):
write_piff_header(ctx['dest_stream'], info_dict['_download_params'])
extra_state['ism_track_written'] = True
self._append_fragment(ctx, frag_content)
except urllib.error.HTTPError as err:
except HTTPError as err:
retry.error = err
continue
yt_dlp/downloader/mhtml.py
+19 -34
View File
@@ -10,7 +10,7 @@ from ..version import __version__ as YT_DLP_VERSION
class MhtmlFD(FragmentFD):
_STYLESHEET = """\
_STYLESHEET = '''\
html, body {
margin: 0;
padding: 0;
@@ -45,7 +45,7 @@ body > figure > img {
max-width: 100%;
max-height: calc(100vh - 5em);
}
"""
'''
_STYLESHEET = re.sub(r'\s+', ' ', _STYLESHEET)
_STYLESHEET = re.sub(r'\B \B|(?<=[\w\-]) (?=[^\w\-])|(?<=[^\w\-]) (?=[\w\-])', '', _STYLESHEET)
@@ -57,24 +57,19 @@ body > figure > img {
)).decode('us-ascii') + '?='
def _gen_cid(self, i, fragment, frag_boundary):
return '%u.%s@yt-dlp.github.io.invalid' % (i, frag_boundary)
return f'{i}.{frag_boundary}@yt-dlp.github.io.invalid'
def _gen_stub(self, *, fragments, frag_boundary, title):
output = io.StringIO()
output.write((
output.write(
'<!DOCTYPE html>'
'<html>'
'<head>'
'' '<meta name="generator" content="yt-dlp {version}">'
'' '<title>{title}</title>'
'' '<style>{styles}</style>'
'<body>'
).format(
version=escapeHTML(YT_DLP_VERSION),
styles=self._STYLESHEET,
title=escapeHTML(title)
))
f'<meta name="generator" content="yt-dlp {escapeHTML(YT_DLP_VERSION)}">'
f'<title>{escapeHTML(title)}</title>'
f'<style>{self._STYLESHEET}</style>'
'<body>')
t0 = 0
for i, frag in enumerate(fragments):
@@ -87,15 +82,12 @@ body > figure > img {
num=i + 1,
t0=srt_subtitles_timecode(t0),
t1=srt_subtitles_timecode(t1),
duration=formatSeconds(frag['duration'], msec=True)
duration=formatSeconds(frag['duration'], msec=True),
))
except (KeyError, ValueError, TypeError):
t1 = None
output.write((
'<figcaption>Slide #{num}</figcaption>'
).format(num=i + 1))
output.write('<img src="cid:{cid}">'.format(
cid=self._gen_cid(i, frag, frag_boundary)))
output.write(f'<figcaption>Slide #{i + 1}</figcaption>')
output.write(f'<img src="cid:{self._gen_cid(i, frag, frag_boundary)}">')
output.write('</figure>')
t0 = t1
@@ -126,31 +118,24 @@ body > figure > img {
stub = self._gen_stub(
fragments=fragments,
frag_boundary=frag_boundary,
title=title
title=title,
)
ctx['dest_stream'].write((
'MIME-Version: 1.0\r\n'
'From: <nowhere@yt-dlp.github.io.invalid>\r\n'
'To: <nowhere@yt-dlp.github.io.invalid>\r\n'
'Subject: {title}\r\n'
f'Subject: {self._escape_mime(title)}\r\n'
'Content-type: multipart/related; '
'' 'boundary="{boundary}"; '
'' 'type="text/html"\r\n'
'X.yt-dlp.Origin: {origin}\r\n'
f'boundary="{frag_boundary}"; '
'type="text/html"\r\n'
f'X.yt-dlp.Origin: {origin}\r\n'
'\r\n'
'--{boundary}\r\n'
f'--{frag_boundary}\r\n'
'Content-Type: text/html; charset=utf-8\r\n'
'Content-Length: {length}\r\n'
f'Content-Length: {len(stub)}\r\n'
'\r\n'
'{stub}\r\n'
).format(
origin=origin,
boundary=frag_boundary,
length=len(stub),
title=self._escape_mime(title),
stub=stub
).encode())
f'{stub}\r\n').encode())
extra_state['header_written'] = True
for i, fragment in enumerate(fragments):
yt_dlp/downloader/niconico.py
+92 -4
View File
@@ -1,8 +1,12 @@
import json
import threading
import time
from . import get_suitable_downloader
from .common import FileDownloader
from ..utils import sanitized_Request
from .external import FFmpegFD
from ..networking import Request
from ..utils import DownloadError, str_or_none, try_get
class NiconicoDmcFD(FileDownloader):
@@ -11,7 +15,7 @@ class NiconicoDmcFD(FileDownloader):
def real_download(self, filename, info_dict):
from ..extractor.niconico import NiconicoIE
self.to_screen('[%s] Downloading from DMC' % self.FD_NAME)
self.to_screen(f'[{self.FD_NAME}] Downloading from DMC')
ie = NiconicoIE(self.ydl)
info_dict, heartbeat_info_dict = ie._get_heartbeat_info(info_dict)
@@ -24,13 +28,13 @@ class NiconicoDmcFD(FileDownloader):
heartbeat_data = heartbeat_info_dict['data'].encode()
heartbeat_interval = heartbeat_info_dict.get('interval', 30)
request = sanitized_Request(heartbeat_url, heartbeat_data)
request = Request(heartbeat_url, heartbeat_data)
def heartbeat():
try:
self.ydl.urlopen(request).read()
except Exception:
self.to_screen('[%s] Heartbeat failed' % self.FD_NAME)
self.to_screen(f'[{self.FD_NAME}] Heartbeat failed')
with heartbeat_lock:
if not download_complete:
@@ -50,3 +54,87 @@ class NiconicoDmcFD(FileDownloader):
timer[0].cancel()
download_complete = True
return success
class NiconicoLiveFD(FileDownloader):
""" Downloads niconico live without being stopped """
def real_download(self, filename, info_dict):
video_id = info_dict['video_id']
ws_url = info_dict['url']
ws_extractor = info_dict['ws']
ws_origin_host = info_dict['origin']
live_quality = info_dict.get('live_quality', 'high')
live_latency = info_dict.get('live_latency', 'high')
dl = FFmpegFD(self.ydl, self.params or {})
new_info_dict = info_dict.copy()
new_info_dict.update({
'protocol': 'm3u8',
})
def communicate_ws(reconnect):
if reconnect:
ws = self.ydl.urlopen(Request(ws_url, headers={'Origin': f'https://{ws_origin_host}'}))
if self.ydl.params.get('verbose', False):
self.to_screen('[debug] Sending startWatching request')
ws.send(json.dumps({
'type': 'startWatching',
'data': {
'stream': {
'quality': live_quality,
'protocol': 'hls+fmp4',
'latency': live_latency,
'chasePlay': False,
},
'room': {
'protocol': 'webSocket',
'commentable': True,
},
'reconnect': True,
},
}))
else:
ws = ws_extractor
with ws:
while True:
recv = ws.recv()
if not recv:
continue
data = json.loads(recv)
if not data or not isinstance(data, dict):
continue
if data.get('type') == 'ping':
# pong back
ws.send(r'{"type":"pong"}')
ws.send(r'{"type":"keepSeat"}')
elif data.get('type') == 'disconnect':
self.write_debug(data)
return True
elif data.get('type') == 'error':
self.write_debug(data)
message = try_get(data, lambda x: x['body']['code'], str) or recv
return DownloadError(message)
elif self.ydl.params.get('verbose', False):
if len(recv) > 100:
recv = recv[:100] + '...'
self.to_screen(f'[debug] Server said: {recv}')
def ws_main():
reconnect = False
while True:
try:
ret = communicate_ws(reconnect)
if ret is True:
return
except BaseException as e:
self.to_screen('[{}] {}: Connection error occured, reconnecting after 10 seconds: {}'.format('niconico:live', video_id, str_or_none(e)))
time.sleep(10)
continue
finally:
reconnect = True
thread = threading.Thread(target=ws_main, daemon=True)
thread.start()
return dl.download(filename, new_info_dict)
yt_dlp/downloader/rtmp.py
+3 -3
View File
@@ -180,9 +180,9 @@ class RtmpFD(FileDownloader):
while retval in (RD_INCOMPLETE, RD_FAILED) and not test and not live:
prevsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('[rtmpdump] Downloaded %s bytes' % prevsize)
self.to_screen(f'[rtmpdump] Downloaded {prevsize} bytes')
time.sleep(5.0) # This seems to be needed
args = basic_args + ['--resume']
args = [*basic_args, '--resume']
if retval == RD_FAILED:
args += ['--skip', '1']
args = [encodeArgument(a) for a in args]
@@ -197,7 +197,7 @@ class RtmpFD(FileDownloader):
break
if retval == RD_SUCCESS or (test and retval == RD_INCOMPLETE):
fsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('[rtmpdump] Downloaded %s bytes' % fsize)
self.to_screen(f'[rtmpdump] Downloaded {fsize} bytes')
self.try_rename(tmpfilename, filename)
self._hook_progress({
'downloaded_bytes': fsize,
yt_dlp/downloader/youtube_live_chat.py
+5 -7
View File
@@ -1,8 +1,8 @@
import json
import time
import urllib.error
from .fragment import FragmentFD
from ..networking.exceptions import HTTPError
from ..utils import (
RegexNotFoundError,
RetryManager,
@@ -10,6 +10,7 @@ from ..utils import (
int_or_none,
try_get,
)
from ..utils.networking import HTTPHeaderDict
class YoutubeLiveChatFD(FragmentFD):
@@ -17,7 +18,7 @@ class YoutubeLiveChatFD(FragmentFD):
def real_download(self, filename, info_dict):
video_id = info_dict['video_id']
self.to_screen('[%s] Downloading live chat' % self.FD_NAME)
self.to_screen(f'[{self.FD_NAME}] Downloading live chat')
if not self.params.get('skip_download') and info_dict['protocol'] == 'youtube_live_chat':
self.report_warning('Live chat download runs until the livestream ends. '
'If you wish to download the video simultaneously, run a separate yt-dlp instance')
@@ -37,10 +38,7 @@ class YoutubeLiveChatFD(FragmentFD):
start_time = int(time.time() * 1000)
def dl_fragment(url, data=None, headers=None):
http_headers = info_dict.get('http_headers', {})
if headers:
http_headers = http_headers.copy()
http_headers.update(headers)
http_headers = HTTPHeaderDict(info_dict.get('http_headers'), headers)
return self._download_fragment(ctx, url, info_dict, http_headers, data)
def parse_actions_replay(live_chat_continuation):
@@ -129,7 +127,7 @@ class YoutubeLiveChatFD(FragmentFD):
or frag_index == 1 and try_refresh_replay_beginning
or parse_actions_replay)
return (True, *func(live_chat_continuation))
except urllib.error.HTTPError as err:
except HTTPError as err:
retry.error = err
continue
return False, None, None, None
yt_dlp/extractor/_extractors.py
+864 -679
View File
File diff suppressed because it is too large Load Diff
yt_dlp/extractor/abc.py
+129 -23
View File
@@ -4,17 +4,18 @@ import re
import time
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
dict_get,
ExtractorError,
js_to_json,
dict_get,
int_or_none,
js_to_json,
parse_iso8601,
str_or_none,
traverse_obj,
try_get,
unescapeHTML,
update_url_query,
url_or_none,
)
@@ -65,7 +66,7 @@ class ABCIE(InfoExtractor):
'ext': 'mp4',
'title': 'WWI Centenary',
'description': 'md5:c2379ec0ca84072e86b446e536954546',
}
},
}, {
'url': 'https://www.abc.net.au/news/programs/the-world/2020-06-10/black-lives-matter-protests-spawn-support-for/12342074',
'info_dict': {
@@ -73,7 +74,7 @@ class ABCIE(InfoExtractor):
'ext': 'mp4',
'title': 'Black Lives Matter protests spawn support for Papuans in Indonesia',
'description': 'md5:2961a17dc53abc558589ccd0fb8edd6f',
}
},
}, {
'url': 'https://www.abc.net.au/btn/newsbreak/btn-newsbreak-20200814/12560476',
'info_dict': {
@@ -84,7 +85,16 @@ class ABCIE(InfoExtractor):
'upload_date': '20200813',
'uploader': 'Behind the News',
'uploader_id': 'behindthenews',
}
},
}, {
'url': 'https://www.abc.net.au/news/2023-06-25/wagner-boss-orders-troops-back-to-bases-to-avoid-bloodshed/102520540',
'info_dict': {
'id': '102520540',
'title': 'Wagner Group retreating from Russia, leader Prigozhin to move to Belarus',
'ext': 'mp4',
'description': 'Wagner troops leave Rostov-on-Don and\xa0Yevgeny Prigozhin will move to Belarus under a deal brokered by Belarusian President Alexander Lukashenko to end the mutiny.',
'thumbnail': 'https://live-production.wcms.abc-cdn.net.au/0c170f5b57f0105c432f366c0e8e267b?impolicy=wcms_crop_resize&cropH=2813&cropW=5000&xPos=0&yPos=249&width=862&height=485',
},
}]
def _real_extract(self, url):
@@ -107,7 +117,7 @@ class ABCIE(InfoExtractor):
video = True
if mobj is None:
mobj = re.search(r'(?P<type>)"sources": (?P<json_data>\[[^\]]+\]),', webpage)
mobj = re.search(r'(?P<type>)"(?:sources|files|renditions)":\s*(?P<json_data>\[[^\]]+\])', webpage)
if mobj is None:
mobj = re.search(
r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
@@ -115,13 +125,14 @@ class ABCIE(InfoExtractor):
if mobj is None:
expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
if expired:
raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
raise ExtractorError(f'{self.IE_NAME} said: {expired}', expected=True)
raise ExtractorError('Unable to extract video urls')
urls_info = self._parse_json(
mobj.group('json_data'), video_id, transform_source=js_to_json)
youtube = mobj.group('type') == 'YouTube'
video = mobj.group('type') == 'Video' or urls_info[0]['contentType'] == 'video/mp4'
video = mobj.group('type') == 'Video' or traverse_obj(
urls_info, (0, ('contentType', 'MIMEType')), get_all=False) == 'video/mp4'
if not isinstance(urls_info, list):
urls_info = [urls_info]
@@ -152,7 +163,7 @@ class ABCIE(InfoExtractor):
'height': height,
'tbr': bitrate,
'filesize': int_or_none(url_info.get('filesize')),
'format_id': format_id
'format_id': format_id,
})
return {
@@ -169,20 +180,100 @@ class ABCIViewIE(InfoExtractor):
_VALID_URL = r'https?://iview\.abc\.net\.au/(?:[^/]+/)*video/(?P<id>[^/?#]+)'
_GEO_COUNTRIES = ['AU']
# ABC iview programs are normally available for 14 days only.
_TESTS = [{
'url': 'https://iview.abc.net.au/show/utopia/series/1/video/CO1211V001S00',
'md5': '52a942bfd7a0b79a6bfe9b4ce6c9d0ed',
'info_dict': {
'id': 'CO1211V001S00',
'ext': 'mp4',
'title': 'Series 1 Ep 1 Wood For The Trees',
'series': 'Utopia',
'description': 'md5:0cfb2c183c1b952d1548fd65c8a95c00',
'upload_date': '20230726',
'uploader_id': 'abc1',
'series_id': 'CO1211V',
'episode_id': 'CO1211V001S00',
'season_number': 1,
'season': 'Season 1',
'episode_number': 1,
'episode': 'Wood For The Trees',
'thumbnail': 'https://cdn.iview.abc.net.au/thumbs/i/co/CO1211V001S00_5ad8353f4df09_1280.jpg',
'timestamp': 1690403700,
},
'params': {
'skip_download': True,
},
}, {
'note': 'No episode name',
'url': 'https://iview.abc.net.au/show/gruen/series/11/video/LE1927H001S00',
'md5': '67715ce3c78426b11ba167d875ac6abf',
'info_dict': {
'id': 'LE1927H001S00',
'ext': 'mp4',
'title': "Series 11 Ep 1",
'series': "Gruen",
'title': 'Series 11 Ep 1',
'series': 'Gruen',
'description': 'md5:52cc744ad35045baf6aded2ce7287f67',
'upload_date': '20190925',
'uploader_id': 'abc1',
'series_id': 'LE1927H',
'episode_id': 'LE1927H001S00',
'season_number': 11,
'season': 'Season 11',
'episode_number': 1,
'episode': 'Episode 1',
'thumbnail': 'https://cdn.iview.abc.net.au/thumbs/i/le/LE1927H001S00_5d954fbd79e25_1280.jpg',
'timestamp': 1569445289,
},
'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
'params': {
'skip_download': True,
},
}, {
'note': 'No episode number',
'url': 'https://iview.abc.net.au/show/four-corners/series/2022/video/NC2203H039S00',
'md5': '77cb7d8434440e3b28fbebe331c2456a',
'info_dict': {
'id': 'NC2203H039S00',
'ext': 'mp4',
'title': 'Series 2022 Locking Up Kids',
'series': 'Four Corners',
'description': 'md5:54829ca108846d1a70e1fcce2853e720',
'upload_date': '20221114',
'uploader_id': 'abc1',
'series_id': 'NC2203H',
'episode_id': 'NC2203H039S00',
'season_number': 2022,
'season': 'Season 2022',
'episode': 'Locking Up Kids',
'thumbnail': 'https://cdn.iview.abc.net.au/thumbs/i/nc/NC2203H039S00_636d8a0944a22_1920.jpg',
'timestamp': 1668460497,
},
'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
'params': {
'skip_download': True,
},
}, {
'note': 'No episode name or number',
'url': 'https://iview.abc.net.au/show/landline/series/2021/video/RF2004Q043S00',
'md5': '2e17dec06b13cc81dc119d2565289396',
'info_dict': {
'id': 'RF2004Q043S00',
'ext': 'mp4',
'title': 'Series 2021',
'series': 'Landline',
'description': 'md5:c9f30d9c0c914a7fd23842f6240be014',
'upload_date': '20211205',
'uploader_id': 'abc1',
'series_id': 'RF2004Q',
'episode_id': 'RF2004Q043S00',
'season_number': 2021,
'season': 'Season 2021',
'thumbnail': 'https://cdn.iview.abc.net.au/thumbs/i/rf/RF2004Q043S00_61a950639dbc0_1920.jpg',
'timestamp': 1638710705,
},
'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
'params': {
'skip_download': True,
},
@@ -196,13 +287,12 @@ class ABCIViewIE(InfoExtractor):
stream = next(s for s in video_params['playlist'] if s.get('type') in ('program', 'livestream'))
house_number = video_params.get('episodeHouseNumber') or video_id
path = '/auth/hls/sign?ts={0}&hn={1}&d=android-tablet'.format(
int(time.time()), house_number)
path = f'/auth/hls/sign?ts={int(time.time())}&hn={house_number}&d=android-tablet'
sig = hmac.new(
b'android.content.res.Resources',
path.encode('utf-8'), hashlib.sha256).hexdigest()
path.encode(), hashlib.sha256).hexdigest()
token = self._download_webpage(
'http://iview.abc.net.au{0}&sig={1}'.format(path, sig), video_id)
f'http://iview.abc.net.au{path}&sig={sig}', video_id)
def tokenize_url(url, token):
return update_url_query(url, {
@@ -211,7 +301,7 @@ class ABCIViewIE(InfoExtractor):
for sd in ('1080', '720', 'sd', 'sd-low'):
sd_url = try_get(
stream, lambda x: x['streams']['hls'][sd], compat_str)
stream, lambda x: x['streams']['hls'][sd], str)
if not sd_url:
continue
formats = self._extract_m3u8_formats(
@@ -244,6 +334,8 @@ class ABCIViewIE(InfoExtractor):
'episode_number': int_or_none(self._search_regex(
r'\bEp\s+(\d+)\b', title, 'episode number', default=None)),
'episode_id': house_number,
'episode': self._search_regex(
r'^(?:Series\s+\d+)?\s*(?:Ep\s+\d+)?\s*(.*)$', title, 'episode', default='') or None,
'uploader_id': video_params.get('channel'),
'formats': formats,
'subtitles': subtitles,
@@ -264,7 +356,7 @@ class ABCIViewShowSeriesIE(InfoExtractor):
'description': 'md5:93119346c24a7c322d446d8eece430ff',
'series': 'Upper Middle Bogan',
'season': 'Series 1',
'thumbnail': r're:^https?://cdn\.iview\.abc\.net\.au/thumbs/.*\.jpg$'
'thumbnail': r're:^https?://cdn\.iview\.abc\.net\.au/thumbs/.*\.jpg$',
},
'playlist_count': 8,
}, {
@@ -283,6 +375,18 @@ class ABCIViewShowSeriesIE(InfoExtractor):
'noplaylist': True,
'skip_download': 'm3u8',
},
}, {
# 'videoEpisodes' is a dict with `items` key
'url': 'https://iview.abc.net.au/show/7-30-mark-humphries-satire',
'info_dict': {
'id': '178458-0',
'title': 'Episodes',
'description': 'Satirist Mark Humphries brings his unique perspective on current political events for 7.30.',
'series': '7.30 Mark Humphries Satire',
'season': 'Episodes',
'thumbnail': r're:^https?://cdn\.iview\.abc\.net\.au/thumbs/.*\.jpg$',
},
'playlist_count': 15,
}]
def _real_extract(self, url):
@@ -292,7 +396,7 @@ class ABCIViewShowSeriesIE(InfoExtractor):
r'window\.__INITIAL_STATE__\s*=\s*[\'"](.+?)[\'"]\s*;',
webpage, 'initial state')
video_data = self._parse_json(
unescapeHTML(webpage_data).encode('utf-8').decode('unicode_escape'), show_id)
unescapeHTML(webpage_data).encode().decode('unicode_escape'), show_id)
video_data = video_data['route']['pageData']['_embedded']
highlight = try_get(video_data, lambda x: x['highlightVideo']['shareUrl'])
@@ -302,12 +406,14 @@ class ABCIViewShowSeriesIE(InfoExtractor):
series = video_data['selectedSeries']
return {
'_type': 'playlist',
'entries': [self.url_result(episode['shareUrl'])
for episode in series['_embedded']['videoEpisodes']],
'entries': [self.url_result(episode_url, ABCIViewIE)
for episode_url in traverse_obj(series, (
'_embedded', 'videoEpisodes', (None, 'items'), ..., 'shareUrl', {url_or_none}))],
'id': series.get('id'),
'title': dict_get(series, ('title', 'displaySubtitle')),
'description': series.get('description'),
'series': dict_get(series, ('showTitle', 'displayTitle')),
'season': dict_get(series, ('title', 'displaySubtitle')),
'thumbnail': series.get('thumbnail'),
'thumbnail': traverse_obj(
series, 'thumbnail', ('images', lambda _, v: v['name'] == 'seriesThumbnail', 'url'), get_all=False),
}
yt_dlp/extractor/abcnews.py
+1 -1
View File
@@ -58,7 +58,7 @@ class AbcNewsVideoIE(AMPIE):
display_id = mobj.group('display_id')
video_id = mobj.group('id')
info_dict = self._extract_feed_info(
'http://abcnews.go.com/video/itemfeed?id=%s' % video_id)
f'http://abcnews.go.com/video/itemfeed?id={video_id}')
info_dict.update({
'id': video_id,
'display_id': display_id,
yt_dlp/extractor/abcotvs.py
+2 -3
View File
@@ -1,5 +1,4 @@
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
dict_get,
int_or_none,
@@ -57,11 +56,11 @@ class ABCOTVSIE(InfoExtractor):
data = self._download_json(
'https://api.abcotvs.com/v2/content', display_id, query={
'id': video_id,
'key': 'otv.web.%s.story' % station,
'key': f'otv.web.{station}.story',
'station': station,
})['data']
video = try_get(data, lambda x: x['featuredMedia']['video'], dict) or data
video_id = compat_str(dict_get(video, ('id', 'publishedKey'), video_id))
video_id = str(dict_get(video, ('id', 'publishedKey'), video_id))
title = video.get('title') or video['linkText']
formats = []
yt_dlp/extractor/abematv.py
+96 -139
View File
@@ -9,107 +9,53 @@ import re
import struct
import time
import urllib.parse
import urllib.request
import urllib.response
import uuid
from .common import InfoExtractor
from ..aes import aes_ecb_decrypt
from ..networking import RequestHandler, Response
from ..networking.exceptions import TransportError
from ..utils import (
ExtractorError,
OnDemandPagedList,
bytes_to_intlist,
decode_base_n,
int_or_none,
intlist_to_bytes,
OnDemandPagedList,
request_to_url,
time_seconds,
traverse_obj,
update_url_query,
)
# NOTE: network handler related code is temporary thing until network stack overhaul PRs are merged (#2861/#2862)
class AbemaLicenseRH(RequestHandler):
_SUPPORTED_URL_SCHEMES = ('abematv-license',)
_SUPPORTED_PROXY_SCHEMES = None
_SUPPORTED_FEATURES = None
RH_NAME = 'abematv_license'
def add_opener(ydl, handler):
''' Add a handler for opening URLs, like _download_webpage '''
# https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426
# https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605
assert isinstance(ydl._opener, urllib.request.OpenerDirector)
ydl._opener.add_handler(handler)
_STRTABLE = '123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz'
_HKEY = b'3AF0298C219469522A313570E8583005A642E73EDD58E3EA2FB7339D3DF1597E'
def remove_opener(ydl, handler):
'''
Remove handler(s) for opening URLs
@param handler Either handler object itself or handler type.
Specifying handler type will remove all handler which isinstance returns True.
'''
# https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426
# https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605
opener = ydl._opener
assert isinstance(ydl._opener, urllib.request.OpenerDirector)
if isinstance(handler, (type, tuple)):
find_cp = lambda x: isinstance(x, handler)
else:
find_cp = lambda x: x is handler
removed = []
for meth in dir(handler):
if meth in ["redirect_request", "do_open", "proxy_open"]:
# oops, coincidental match
continue
i = meth.find("_")
protocol = meth[:i]
condition = meth[i + 1:]
if condition.startswith("error"):
j = condition.find("_") + i + 1
kind = meth[j + 1:]
try:
kind = int(kind)
except ValueError:
pass
lookup = opener.handle_error.get(protocol, {})
opener.handle_error[protocol] = lookup
elif condition == "open":
kind = protocol
lookup = opener.handle_open
elif condition == "response":
kind = protocol
lookup = opener.process_response
elif condition == "request":
kind = protocol
lookup = opener.process_request
else:
continue
handlers = lookup.setdefault(kind, [])
if handlers:
handlers[:] = [x for x in handlers if not find_cp(x)]
removed.append(x for x in handlers if find_cp(x))
if removed:
for x in opener.handlers:
if find_cp(x):
x.add_parent(None)
opener.handlers[:] = [x for x in opener.handlers if not find_cp(x)]
class AbemaLicenseHandler(urllib.request.BaseHandler):
handler_order = 499
STRTABLE = '123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz'
HKEY = b'3AF0298C219469522A313570E8583005A642E73EDD58E3EA2FB7339D3DF1597E'
def __init__(self, ie: 'AbemaTVIE'):
# the protocol that this should really handle is 'abematv-license://'
# abematv_license_open is just a placeholder for development purposes
# ref. https://github.com/python/cpython/blob/f4c03484da59049eb62a9bf7777b963e2267d187/Lib/urllib/request.py#L510
setattr(self, 'abematv-license_open', getattr(self, 'abematv_license_open'))
def __init__(self, *, ie: 'AbemaTVIE', **kwargs):
super().__init__(**kwargs)
self.ie = ie
def _send(self, request):
url = request.url
ticket = urllib.parse.urlparse(url).netloc
try:
response_data = self._get_videokey_from_ticket(ticket)
except ExtractorError as e:
raise TransportError(cause=e.cause) from e
except (IndexError, KeyError, TypeError) as e:
raise TransportError(cause=repr(e)) from e
return Response(
io.BytesIO(response_data), url,
headers={'Content-Length': str(len(response_data))})
def _get_videokey_from_ticket(self, ticket):
to_show = self.ie.get_param('verbose', False)
media_token = self.ie._get_media_token(to_show=to_show)
@@ -119,33 +65,27 @@ class AbemaLicenseHandler(urllib.request.BaseHandler):
query={'t': media_token},
data=json.dumps({
'kv': 'a',
'lt': ticket
}).encode('utf-8'),
'lt': ticket,
}).encode(),
headers={
'Content-Type': 'application/json',
})
res = decode_base_n(license_response['k'], table=self.STRTABLE)
res = decode_base_n(license_response['k'], table=self._STRTABLE)
encvideokey = bytes_to_intlist(struct.pack('>QQ', res >> 64, res & 0xffffffffffffffff))
h = hmac.new(
binascii.unhexlify(self.HKEY),
(license_response['cid'] + self.ie._DEVICE_ID).encode('utf-8'),
binascii.unhexlify(self._HKEY),
(license_response['cid'] + self.ie._DEVICE_ID).encode(),
digestmod=hashlib.sha256)
enckey = bytes_to_intlist(h.digest())
return intlist_to_bytes(aes_ecb_decrypt(encvideokey, enckey))
def abematv_license_open(self, url):
url = request_to_url(url)
ticket = urllib.parse.urlparse(url).netloc
response_data = self._get_videokey_from_ticket(ticket)
return urllib.response.addinfourl(io.BytesIO(response_data), headers={
'Content-Length': len(response_data),
}, url=url, code=200)
class AbemaTVBaseIE(InfoExtractor):
_NETRC_MACHINE = 'abematv'
_USERTOKEN = None
_DEVICE_ID = None
_MEDIATOKEN = None
@@ -154,11 +94,11 @@ class AbemaTVBaseIE(InfoExtractor):
@classmethod
def _generate_aks(cls, deviceid):
deviceid = deviceid.encode('utf-8')
deviceid = deviceid.encode()
# add 1 hour and then drop minute and secs
ts_1hour = int((time_seconds() // 3600 + 1) * 3600)
time_struct = time.gmtime(ts_1hour)
ts_1hour_str = str(ts_1hour).encode('utf-8')
ts_1hour_str = str(ts_1hour).encode()
tmp = None
@@ -170,7 +110,7 @@ class AbemaTVBaseIE(InfoExtractor):
def mix_tmp(count):
nonlocal tmp
for i in range(count):
for _ in range(count):
mix_once(tmp)
def mix_twist(nonce):
@@ -190,11 +130,15 @@ class AbemaTVBaseIE(InfoExtractor):
if self._USERTOKEN:
return self._USERTOKEN
self._downloader._request_director.add_handler(AbemaLicenseRH(ie=self, logger=None))
username, _ = self._get_login_info()
AbemaTVBaseIE._USERTOKEN = username and self.cache.load(self._NETRC_MACHINE, username)
auth_cache = username and self.cache.load(self._NETRC_MACHINE, username, min_ver='2024.01.19')
AbemaTVBaseIE._USERTOKEN = auth_cache and auth_cache.get('usertoken')
if AbemaTVBaseIE._USERTOKEN:
# try authentication with locally stored token
try:
AbemaTVBaseIE._DEVICE_ID = auth_cache.get('device_id')
self._get_media_token(True)
return
except ExtractorError as e:
@@ -207,16 +151,12 @@ class AbemaTVBaseIE(InfoExtractor):
data=json.dumps({
'deviceId': self._DEVICE_ID,
'applicationKeySecret': aks,
}).encode('utf-8'),
}).encode(),
headers={
'Content-Type': 'application/json',
})
AbemaTVBaseIE._USERTOKEN = user_data['token']
# don't allow adding it 2 times or more, though it's guarded
remove_opener(self._downloader, AbemaLicenseHandler)
add_opener(self._downloader, AbemaLicenseHandler(self))
return self._USERTOKEN
def _get_media_token(self, invalidate=False, to_show=True):
@@ -231,13 +171,44 @@ class AbemaTVBaseIE(InfoExtractor):
'osLang': 'ja_JP',
'osTimezone': 'Asia/Tokyo',
'appId': 'tv.abema',
'appVersion': '3.27.1'
'appVersion': '3.27.1',
}, headers={
'Authorization': f'bearer {self._get_device_token()}',
})['token']
return self._MEDIATOKEN
def _perform_login(self, username, password):
self._get_device_token()
if self.cache.load(self._NETRC_MACHINE, username, min_ver='2024.01.19') and self._get_media_token():
self.write_debug('Skipping logging in')
return
if '@' in username: # don't strictly check if it's email address or not
ep, method = 'user/email', 'email'
else:
ep, method = 'oneTimePassword', 'userId'
login_response = self._download_json(
f'https://api.abema.io/v1/auth/{ep}', None, note='Logging in',
data=json.dumps({
method: username,
'password': password,
}).encode(), headers={
'Authorization': f'bearer {self._get_device_token()}',
'Origin': 'https://abema.tv',
'Referer': 'https://abema.tv/',
'Content-Type': 'application/json',
})
AbemaTVBaseIE._USERTOKEN = login_response['token']
self._get_media_token(True)
auth_cache = {
'device_id': AbemaTVBaseIE._DEVICE_ID,
'usertoken': AbemaTVBaseIE._USERTOKEN,
}
self.cache.store(self._NETRC_MACHINE, username, auth_cache)
def _call_api(self, endpoint, video_id, query=None, note='Downloading JSON metadata'):
return self._download_json(
f'https://api.abema.io/{endpoint}', video_id, query=query or {},
@@ -261,14 +232,14 @@ class AbemaTVBaseIE(InfoExtractor):
class AbemaTVIE(AbemaTVBaseIE):
_VALID_URL = r'https?://abema\.tv/(?P<type>now-on-air|video/episode|channels/.+?/slots)/(?P<id>[^?/]+)'
_NETRC_MACHINE = 'abematv'
_TESTS = [{
'url': 'https://abema.tv/video/episode/194-25_s2_p1',
'info_dict': {
'id': '194-25_s2_p1',
'title': '第1話 「チーズケーキ」　「モーニング再び」',
'series': '異世界食堂２',
'series_number': 2,
'season': 'シーズン2',
'season_number': 2,
'episode': '第1話 「チーズケーキ」　「モーニング再び」',
'episode_number': 1,
},
@@ -280,7 +251,7 @@ class AbemaTVIE(AbemaTVBaseIE):
'title': 'ゆるキャン△ SEASON２ 全話一挙【無料ビデオ72時間】',
'series': 'ゆるキャン△ SEASON２',
'episode': 'ゆるキャン△ SEASON２ 全話一挙【無料ビデオ72時間】',
'series_number': 2,
'season_number': 2,
'episode_number': 1,
'description': 'md5:9c5a3172ae763278f9303922f0ea5b17',
},
@@ -309,33 +280,6 @@ class AbemaTVIE(AbemaTVBaseIE):
}]
_TIMETABLE = None
def _perform_login(self, username, password):
self._get_device_token()
if self.cache.load(self._NETRC_MACHINE, username) and self._get_media_token():
self.write_debug('Skipping logging in')
return
if '@' in username: # don't strictly check if it's email address or not
ep, method = 'user/email', 'email'
else:
ep, method = 'oneTimePassword', 'userId'
login_response = self._download_json(
f'https://api.abema.io/v1/auth/{ep}', None, note='Logging in',
data=json.dumps({
method: username,
'password': password
}).encode('utf-8'), headers={
'Authorization': f'bearer {self._get_device_token()}',
'Origin': 'https://abema.tv',
'Referer': 'https://abema.tv/',
'Content-Type': 'application/json',
})
AbemaTVBaseIE._USERTOKEN = login_response['token']
self._get_media_token(True)
self.cache.store(self._NETRC_MACHINE, username, AbemaTVBaseIE._USERTOKEN)
def _real_extract(self, url):
# starting download using infojson from this extractor is undefined behavior,
# and never be fixed in the future; you must trigger downloads by directly specifying URL.
@@ -391,7 +335,7 @@ class AbemaTVIE(AbemaTVBaseIE):
description = self._html_search_regex(
(r'<p\s+class="com-video-EpisodeDetailsBlock__content"><span\s+class=".+?">(.+?)</span></p><div',
r'<span\s+class=".+?SlotSummary.+?">(.+?)</span></div><div',),
r'<span\s+class=".+?SlotSummary.+?">(.+?)</span></div><div'),
webpage, 'description', default=None, group=1)
if not description:
og_desc = self._html_search_meta(
@@ -404,17 +348,18 @@ class AbemaTVIE(AbemaTVBaseIE):
)?
''', r'\1', og_desc)
# canonical URL may contain series and episode number
# canonical URL may contain season and episode number
mobj = re.search(r's(\d+)_p(\d+)$', canonical_url)
if mobj:
seri = int_or_none(mobj.group(1), default=float('inf'))
epis = int_or_none(mobj.group(2), default=float('inf'))
info['series_number'] = seri if seri < 100 else None
info['season_number'] = seri if seri < 100 else None
# some anime like Detective Conan (though not available in AbemaTV)
# has more than 1000 episodes (1026 as of 2021/11/15)
info['episode_number'] = epis if epis < 2000 else None
is_live, m3u8_url = False, None
availability = 'public'
if video_type == 'now-on-air':
is_live = True
channel_url = 'https://api.abema.io/v1/channels'
@@ -432,10 +377,20 @@ class AbemaTVIE(AbemaTVBaseIE):
f'https://api.abema.io/v1/video/programs/{video_id}', video_id,
note='Checking playability',
headers=headers)
ondemand_types = traverse_obj(api_response, ('terms', ..., 'onDemandType'))
if 3 not in ondemand_types:
if not traverse_obj(api_response, ('label', 'free', {bool})):
# cannot acquire decryption key for these streams
self.report_warning('This is a premium-only stream')
availability = 'premium_only'
info.update(traverse_obj(api_response, {
'series': ('series', 'title'),
'season': ('season', 'name'),
'season_number': ('season', 'sequence'),
'episode_number': ('episode', 'number'),
}))
if not title:
title = traverse_obj(api_response, ('episode', 'title'))
if not description:
description = traverse_obj(api_response, ('episode', 'content'))
m3u8_url = f'https://vod-abematv.akamaized.net/program/{video_id}/playlist.m3u8'
elif video_type == 'slots':
@@ -445,6 +400,7 @@ class AbemaTVIE(AbemaTVBaseIE):
headers=headers)
if not traverse_obj(api_response, ('slot', 'flags', 'timeshiftFree'), default=False):
self.report_warning('This is a premium-only stream')
availability = 'premium_only'
m3u8_url = f'https://vod-abematv.akamaized.net/slot/{video_id}/playlist.m3u8'
else:
@@ -462,6 +418,7 @@ class AbemaTVIE(AbemaTVBaseIE):
'description': description,
'formats': formats,
'is_live': is_live,
'availability': availability,
})
return info
yt_dlp/extractor/acast.py
+31 -9
View File
@@ -40,29 +40,34 @@ class ACastBaseIE(InfoExtractor):
class ACastIE(ACastBaseIE):
IE_NAME = 'acast'
_VALID_URL = r'''(?x)
_VALID_URL = r'''(?x:
https?://
(?:
(?:(?:embed|www)\.)?acast\.com/|
play\.acast\.com/s/
)
(?P<channel>[^/]+)/(?P<id>[^/#?]+)
'''
(?P<channel>[^/]+)/(?P<id>[^/#?"]+)
)'''
_EMBED_REGEX = [rf'(?x)<iframe[^>]+\bsrc=[\'"](?P<url>{_VALID_URL})']
_TESTS = [{
'url': 'https://www.acast.com/sparpodcast/2.raggarmordet-rosterurdetforflutna',
'md5': 'f5598f3ad1e4776fed12ec1407153e4b',
'info_dict': {
'id': '2a92b283-1a75-4ad8-8396-499c641de0d9',
'ext': 'mp3',
'title': '2. Raggarmordet - Röster ur det förflutna',
'description': 'md5:a992ae67f4d98f1c0141598f7bebbf67',
'description': 'md5:013959207e05011ad14a222cf22278cc',
'timestamp': 1477346700,
'upload_date': '20161024',
'duration': 2766,
'creator': 'Anton Berg & Martin Johnson',
'creator': 'Third Ear Studio',
'series': 'Spår',
'episode': '2. Raggarmordet - Röster ur det förflutna',
}
'thumbnail': 'https://assets.pippa.io/shows/616ebe1886d7b1398620b943/616ebe33c7e6e70013cae7da.jpg',
'episode_number': 2,
'display_id': '2.raggarmordet-rosterurdetforflutna',
'season_number': 4,
'season': 'Season 4',
},
}, {
'url': 'http://embed.acast.com/adambuxton/ep.12-adam-joeschristmaspodcast2015',
'only_matching': True,
@@ -73,11 +78,28 @@ class ACastIE(ACastBaseIE):
'url': 'https://play.acast.com/s/sparpodcast/2a92b283-1a75-4ad8-8396-499c641de0d9',
'only_matching': True,
}]
_WEBPAGE_TESTS = [{
'url': 'https://ausi.anu.edu.au/news/democracy-sausage-episode-can-labor-be-long-form-government',
'info_dict': {
'id': '646c68fb21fbf20011e9c651',
'ext': 'mp3',
'creator': 'The Australian National University',
'display_id': 'can-labor-be-a-long-form-government',
'duration': 2618,
'thumbnail': 'https://assets.pippa.io/shows/6113e8578b4903809f16f7e5/1684821529295-515b9520db9ce53275b995eb302f941c.jpeg',
'title': 'Can Labor be a long-form government?',
'episode': 'Can Labor be a long-form government?',
'upload_date': '20230523',
'series': 'Democracy Sausage with Mark Kenny',
'timestamp': 1684826362,
'description': 'md5:feabe1fc5004c78ee59c84a46bf4ba16',
},
}]
def _real_extract(self, url):
channel, display_id = self._match_valid_url(url).groups()
episode = self._call_api(
'%s/episodes/%s' % (channel, display_id),
f'{channel}/episodes/{display_id}',
display_id, {'showInfo': 'true'})
return self._extract_episode(
episode, self._extract_show_info(episode.get('show') or {}))
@@ -108,7 +130,7 @@ class ACastChannelIE(ACastBaseIE):
@classmethod
def suitable(cls, url):
return False if ACastIE.suitable(url) else super(ACastChannelIE, cls).suitable(url)
return False if ACastIE.suitable(url) else super().suitable(url)
def _real_extract(self, url):
show_slug = self._match_id(url)
yt_dlp/extractor/acfun.py
+7 -6
View File
@@ -3,9 +3,10 @@ from ..utils import (
float_or_none,
format_field,
int_or_none,
traverse_obj,
parse_codecs,
parse_qs,
str_or_none,
traverse_obj,
)
@@ -24,7 +25,7 @@ class AcFunVideoBaseIE(InfoExtractor):
'width': int_or_none(video.get('width')),
'height': int_or_none(video.get('height')),
'tbr': float_or_none(video.get('avgBitrate')),
**parse_codecs(video.get('codecs', ''))
**parse_codecs(video.get('codecs', '')),
})
return {
@@ -76,7 +77,7 @@ class AcFunVideoIE(AcFunVideoBaseIE):
'comment_count': int,
'thumbnail': r're:^https?://.*\.(jpg|jpeg)',
'description': 'md5:67583aaf3a0f933bd606bc8a2d3ebb17',
}
},
}]
def _real_extract(self, url):
@@ -129,7 +130,7 @@ class AcFunBangumiIE(AcFunVideoBaseIE):
'title': '红孩儿之趴趴蛙寻石记 第5话 ',
'duration': 760.0,
'season': '红孩儿之趴趴蛙寻石记',
'season_id': 5023171,
'season_id': '5023171',
'season_number': 1, # series has only 1 season
'episode': 'Episode 5',
'episode_number': 5,
@@ -146,7 +147,7 @@ class AcFunBangumiIE(AcFunVideoBaseIE):
'title': '叽歪老表（第二季） 第5话 坚不可摧',
'season': '叽歪老表（第二季）',
'season_number': 2,
'season_id': 6065485,
'season_id': '6065485',
'episode': '坚不可摧',
'episode_number': 5,
'upload_date': '20220324',
@@ -191,7 +192,7 @@ class AcFunBangumiIE(AcFunVideoBaseIE):
'title': json_bangumi_data.get('showTitle'),
'thumbnail': json_bangumi_data.get('image'),
'season': json_bangumi_data.get('bangumiTitle'),
'season_id': season_id,
'season_id': str_or_none(season_id),
'season_number': season_number,
'episode': json_bangumi_data.get('title'),
'episode_number': episode_number,
yt_dlp/extractor/adn.py
+117 -45
View File
@@ -3,35 +3,55 @@ import binascii
import json
import os
import random
import time
from .common import InfoExtractor
from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
from ..compat import (
compat_HTTPError,
compat_b64decode,
)
from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
ass_subtitles_timecode,
bytes_to_intlist,
bytes_to_long,
ExtractorError,
float_or_none,
int_or_none,
intlist_to_bytes,
join_nonempty,
long_to_bytes,
parse_iso8601,
pkcs1pad,
str_or_none,
strip_or_none,
try_get,
unified_strdate,
urlencode_postdata,
)
from ..utils.traversal import traverse_obj
class ADNIE(InfoExtractor):
class ADNBaseIE(InfoExtractor):
IE_DESC = 'Animation Digital Network'
_VALID_URL = r'https?://(?:www\.)?(?:animation|anime)digitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
_NETRC_MACHINE = 'animationdigitalnetwork'
_BASE = 'animationdigitalnetwork.fr'
_API_BASE_URL = f'https://gw.api.{_BASE}/'
_PLAYER_BASE_URL = f'{_API_BASE_URL}player/'
_HEADERS = {}
_LOGIN_ERR_MESSAGE = 'Unable to log in'
_RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537)
_POS_ALIGN_MAP = {
'start': 1,
'end': 3,
}
_LINE_ALIGN_MAP = {
'middle': 8,
'end': 4,
}
class ADNIE(ADNBaseIE):
_VALID_URL = r'https?://(?:www\.)?animationdigitalnetwork\.com/(?:(?P<lang>de)/)?video/[^/?#]+/(?P<id>\d+)'
_TESTS = [{
'url': 'https://animationdigitalnetwork.fr/video/fruits-basket/9841-episode-1-a-ce-soir',
'url': 'https://animationdigitalnetwork.com/video/558-fruits-basket/9841-episode-1-a-ce-soir',
'md5': '1c9ef066ceb302c86f80c2b371615261',
'info_dict': {
'id': '9841',
@@ -46,29 +66,32 @@ class ADNIE(InfoExtractor):
'season_number': 1,
'episode': 'À ce soir !',
'episode_number': 1,
'thumbnail': str,
'season': 'Season 1',
},
'skip': 'Only available in region (FR, ...)',
'skip': 'Only available in French and German speaking Europe',
}, {
'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
'only_matching': True,
'url': 'https://animationdigitalnetwork.com/de/video/973-the-eminence-in-shadow/23550-folge-1',
'md5': '5c5651bf5791fa6fcd7906012b9d94e8',
'info_dict': {
'id': '23550',
'ext': 'mp4',
'episode_number': 1,
'duration': 1417,
'release_date': '20231004',
'series': 'The Eminence in Shadow',
'season_number': 2,
'episode': str,
'title': str,
'thumbnail': str,
'season': 'Season 2',
'comment_count': int,
'average_rating': float,
'description': str,
},
# 'skip': 'Only available in French and German speaking Europe',
}]
_NETRC_MACHINE = 'animationdigitalnetwork'
_BASE = 'animationdigitalnetwork.fr'
_API_BASE_URL = 'https://gw.api.' + _BASE + '/'
_PLAYER_BASE_URL = _API_BASE_URL + 'player/'
_HEADERS = {}
_LOGIN_ERR_MESSAGE = 'Unable to log in'
_RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537)
_POS_ALIGN_MAP = {
'start': 1,
'end': 3,
}
_LINE_ALIGN_MAP = {
'middle': 8,
'end': 4,
}
def _get_subtitles(self, sub_url, video_id):
if not sub_url:
return None
@@ -85,9 +108,9 @@ class ADNIE(InfoExtractor):
# http://animationdigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
dec_subtitles = unpad_pkcs7(aes_cbc_decrypt_bytes(
compat_b64decode(enc_subtitles[24:]),
base64.b64decode(enc_subtitles[24:]),
binascii.unhexlify(self._K + '7fac1178830cfe0c'),
compat_b64decode(enc_subtitles[:24])))
base64.b64decode(enc_subtitles[:24])))
subtitles_json = self._parse_json(dec_subtitles.decode(), None, fatal=False)
if not subtitles_json:
return None
@@ -110,7 +133,7 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
if start is None or end is None or text is None:
continue
alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
ssa += os.linesep + 'Dialogue: Marked=0,{},{},Default,,0,0,0,,{}{}'.format(
ass_subtitles_timecode(start),
ass_subtitles_timecode(end),
'{\\a%d}' % alignment if alignment != 2 else '',
@@ -118,6 +141,8 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
if sub_lang == 'vostf':
sub_lang = 'fr'
elif sub_lang == 'vostde':
sub_lang = 'de'
subtitles.setdefault(sub_lang, []).extend([{
'ext': 'json',
'data': json.dumps(sub),
@@ -139,18 +164,19 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
'username': username,
})) or {}).get('accessToken')
if access_token:
self._HEADERS = {'authorization': 'Bearer ' + access_token}
self._HEADERS['Authorization'] = f'Bearer {access_token}'
except ExtractorError as e:
message = None
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
if isinstance(e.cause, HTTPError) and e.cause.status == 401:
resp = self._parse_json(
e.cause.read().decode(), None, fatal=False) or {}
e.cause.response.read().decode(), None, fatal=False) or {}
message = resp.get('message') or resp.get('code')
self.report_warning(message or self._LOGIN_ERR_MESSAGE)
def _real_extract(self, url):
video_id = self._match_id(url)
video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id
lang, video_id = self._match_valid_url(url).group('lang', 'id')
self._HEADERS['X-Target-Distribution'] = lang or 'fr'
video_base_url = self._PLAYER_BASE_URL + f'video/{video_id}/'
player = self._download_json(
video_base_url + 'configuration', video_id,
'Downloading player config JSON metadata',
@@ -159,12 +185,15 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
user = options['user']
if not user.get('hasAccess'):
self.raise_login_required()
start_date = traverse_obj(options, ('video', 'startDate', {str}))
if (parse_iso8601(start_date) or 0) > time.time():
raise ExtractorError(f'This video is not available yet. Release date: {start_date}', expected=True)
self.raise_login_required('This video requires a subscription', method='password')
token = self._download_json(
user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'),
video_id, 'Downloading access token', headers={
'x-player-refresh-token': user['refreshToken']
'X-Player-Refresh-Token': user['refreshToken'],
}, data=b'')['token']
links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link')
@@ -186,23 +215,24 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
try:
links_data = self._download_json(
links_url, video_id, 'Downloading links JSON metadata', headers={
'X-Player-Token': authorization
'X-Player-Token': authorization,
**self._HEADERS,
}, query={
'freeWithAds': 'true',
'adaptive': 'false',
'withMetadata': 'true',
'source': 'Web'
'source': 'Web',
})
break
except ExtractorError as e:
if not isinstance(e.cause, compat_HTTPError):
if not isinstance(e.cause, HTTPError):
raise e
if e.cause.code == 401:
if e.cause.status == 401:
# This usually goes away with a different random pkcs1pad, so retry
continue
error = self._parse_json(e.cause.read(), video_id)
error = self._parse_json(e.cause.response.read(), video_id)
message = error.get('message')
if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country':
self.raise_geo_restricted(msg=message)
@@ -223,7 +253,8 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
for quality, load_balancer_url in qualities.items():
load_balancer_data = self._download_json(
load_balancer_url, video_id,
'Downloading %s %s JSON metadata' % (format_id, quality),
f'Downloading {format_id} {quality} JSON metadata',
headers=self._HEADERS,
fatal=False) or {}
m3u8_url = load_balancer_data.get('location')
if not m3u8_url:
@@ -234,11 +265,17 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
if format_id == 'vf':
for f in m3u8_formats:
f['language'] = 'fr'
elif format_id == 'vde':
for f in m3u8_formats:
f['language'] = 'de'
formats.extend(m3u8_formats)
if not formats:
self.raise_login_required('This video requires a subscription', method='password')
video = (self._download_json(
self._API_BASE_URL + 'video/%s' % video_id, video_id,
'Downloading additional video metadata', fatal=False) or {}).get('video') or {}
self._API_BASE_URL + f'video/{video_id}', video_id,
'Downloading additional video metadata', fatal=False, headers=self._HEADERS) or {}).get('video') or {}
show = video.get('show') or {}
return {
@@ -257,3 +294,38 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
'average_rating': float_or_none(video.get('rating') or metas.get('rating')),
'comment_count': int_or_none(video.get('commentsCount')),
}
class ADNSeasonIE(ADNBaseIE):
_VALID_URL = r'https?://(?:www\.)?animationdigitalnetwork\.com/(?:(?P<lang>de)/)?video/(?P<id>\d+)[^/?#]*/?(?:$|[#?])'
_TESTS = [{
'url': 'https://animationdigitalnetwork.com/video/911-tokyo-mew-mew-new',
'playlist_count': 12,
'info_dict': {
'id': '911',
'title': 'Tokyo Mew Mew New',
},
# 'skip': 'Only available in French end German speaking Europe',
}]
def _real_extract(self, url):
lang, video_show_slug = self._match_valid_url(url).group('lang', 'id')
self._HEADERS['X-Target-Distribution'] = lang or 'fr'
show = self._download_json(
f'{self._API_BASE_URL}show/{video_show_slug}/', video_show_slug,
'Downloading show JSON metadata', headers=self._HEADERS)['show']
show_id = str(show['id'])
episodes = self._download_json(
f'{self._API_BASE_URL}video/show/{show_id}', video_show_slug,
'Downloading episode list', headers=self._HEADERS, query={
'order': 'asc',
'limit': '-1',
})
def entries():
for episode_id in traverse_obj(episodes, ('videos', ..., 'id', {str_or_none})):
yield self.url_result(join_nonempty(
'https://animationdigitalnetwork.com', lang, 'video',
video_show_slug, episode_id, delim='/'), ADNIE, episode_id)
return self.playlist_result(entries(), show_id, show.get('title'))
yt_dlp/extractor/adobeconnect.py
+4 -6
View File
@@ -1,8 +1,6 @@
import urllib.parse
from .common import InfoExtractor
from ..compat import (
compat_parse_qs,
compat_urlparse,
)
class AdobeConnectIE(InfoExtractor):
@@ -12,13 +10,13 @@ class AdobeConnectIE(InfoExtractor):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
title = self._html_extract_title(webpage)
qs = compat_parse_qs(self._search_regex(r"swfUrl\s*=\s*'([^']+)'", webpage, 'swf url').split('?')[1])
qs = urllib.parse.parse_qs(self._search_regex(r"swfUrl\s*=\s*'([^']+)'", webpage, 'swf url').split('?')[1])
is_live = qs.get('isLive', ['false'])[0] == 'true'
formats = []
for con_string in qs['conStrings'][0].split(','):
formats.append({
'format_id': con_string.split('://')[0],
'app': compat_urlparse.quote('?' + con_string.split('?')[1] + 'flvplayerapp/' + qs['appInstance'][0]),
'app': urllib.parse.quote('?' + con_string.split('?')[1] + 'flvplayerapp/' + qs['appInstance'][0]),
'ext': 'flv',
'play_path': 'mp4:' + qs['streamName'][0],
'rtmp_conn': 'S:' + qs['ticket'][0],
yt_dlp/extractor/adobepass.py
+444 -444
View File
File diff suppressed because it is too large Load Diff
yt_dlp/extractor/adobetv.py
+4 -5
View File
@@ -2,13 +2,12 @@ import functools
import re
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
ISO639Utils,
OnDemandPagedList,
float_or_none,
int_or_none,
ISO639Utils,
join_nonempty,
OnDemandPagedList,
parse_duration,
str_or_none,
str_to_int,
@@ -36,7 +35,7 @@ class AdobeTVBaseIE(InfoExtractor):
return subtitles
def _parse_video_data(self, video_data):
video_id = compat_str(video_data['id'])
video_id = str(video_data['id'])
title = video_data['title']
s3_extracted = False
@@ -151,7 +150,7 @@ class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
page += 1
query['page'] = page
for element_data in self._call_api(
self._RESOURCE, display_id, query, 'Download Page %d' % page):
self._RESOURCE, display_id, query, f'Download Page {page}'):
yield self._process_data(element_data)
def _extract_playlist_entries(self, display_id, query):
yt_dlp/extractor/adultswim.py
+7 -6
View File
@@ -91,7 +91,7 @@ class AdultSwimIE(TurnerBaseIE):
getShowBySlug(slug:"%s") {
%%s
}
}''' % show_path
}''' % show_path # noqa: UP031
if episode_path:
query = query % '''title
getVideoBySlug(slug:"%s") {
@@ -107,7 +107,6 @@ class AdultSwimIE(TurnerBaseIE):
title
tvRating
}''' % episode_path
['getVideoBySlug']
else:
query = query % '''metaDescription
title
@@ -129,7 +128,7 @@ class AdultSwimIE(TurnerBaseIE):
episode_title = title = video_data['title']
series = show_data.get('title')
if series:
title = '%s - %s' % (series, title)
title = f'{series} - {title}'
info = {
'id': video_id,
'title': title,
@@ -170,8 +169,10 @@ class AdultSwimIE(TurnerBaseIE):
continue
ext = determine_ext(asset_url, mimetype2ext(asset.get('mime_type')))
if ext == 'm3u8':
info['formats'].extend(self._extract_m3u8_formats(
asset_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
fmts, subs = self._extract_m3u8_formats_and_subtitles(
asset_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
info['formats'].extend(fmts)
self._merge_subtitles(subs, target=info['subtitles'])
elif ext == 'f4m':
continue
# info['formats'].extend(self._extract_f4m_formats(
@@ -190,7 +191,7 @@ class AdultSwimIE(TurnerBaseIE):
if not slug:
continue
entries.append(self.url_result(
'http://adultswim.com/videos/%s/%s' % (show_path, slug),
f'http://adultswim.com/videos/{show_path}/{slug}',
'AdultSwim', video.get('_id')))
return self.playlist_result(
entries, show_path, show_data.get('title'),
yt_dlp/extractor/aenetworks.py
+51 -20
View File
@@ -3,6 +3,8 @@ from ..utils import (
ExtractorError,
GeoRestrictedError,
int_or_none,
remove_start,
traverse_obj,
update_url_query,
urlencode_postdata,
)
@@ -71,8 +73,15 @@ class AENetworksBaseIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
def _extract_aetn_info(self, domain, filter_key, filter_value, url):
requestor_id, brand = self._DOMAIN_MAP[domain]
result = self._download_json(
'https://feeds.video.aetnd.com/api/v2/%s/videos' % brand,
filter_value, query={'filter[%s]' % filter_key: filter_value})['results'][0]
f'https://feeds.video.aetnd.com/api/v2/{brand}/videos',
filter_value, query={f'filter[{filter_key}]': filter_value})
result = traverse_obj(
result, ('results',
lambda k, v: k == 0 and v[filter_key] == filter_value),
get_all=False)
if not result:
raise ExtractorError('Show not found in A&E feed (too new?)', expected=True,
video_id=remove_start(filter_value, '/'))
title = result['title']
video_id = result['id']
media_url = result['publicUrl']
@@ -84,7 +93,7 @@ class AENetworksBaseIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
resource = self._get_mvpd_resource(
requestor_id, theplatform_metadata['title'],
theplatform_metadata.get('AETN$PPL_pplProgramId') or theplatform_metadata.get('AETN$PPL_pplProgramId_OLD'),
theplatform_metadata['ratings'][0]['rating'])
traverse_obj(theplatform_metadata, ('ratings', 0, 'rating')))
auth = self._extract_mvpd_auth(
url, video_id, requestor_id, resource)
info.update(self._extract_aen_smil(media_url, video_id, auth))
@@ -112,18 +121,28 @@ class AENetworksIE(AENetworksBaseIE):
'info_dict': {
'id': '22253814',
'ext': 'mp4',
'title': 'Winter is Coming',
'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
'title': 'Winter Is Coming',
'description': 'md5:a40e370925074260b1c8a633c632c63a',
'timestamp': 1338306241,
'upload_date': '20120529',
'uploader': 'AENE-NEW',
'duration': 2592.0,
'thumbnail': r're:^https?://.*\.jpe?g$',
'chapters': 'count:5',
'tags': 'count:14',
'categories': ['Mountain Men'],
'episode_number': 1,
'episode': 'Episode 1',
'season': 'Season 1',
'season_number': 1,
'series': 'Mountain Men',
},
'params': {
# m3u8 download
'skip_download': True,
},
'add_ie': ['ThePlatform'],
'skip': 'This video is only available for users of participating TV providers.',
'skip': 'Geo-restricted - This content is not available in your location.',
}, {
'url': 'http://www.aetv.com/shows/duck-dynasty/season-9/episode-1',
'info_dict': {
@@ -134,36 +153,46 @@ class AENetworksIE(AENetworksBaseIE):
'timestamp': 1452634428,
'upload_date': '20160112',
'uploader': 'AENE-NEW',
'duration': 1277.695,
'thumbnail': r're:^https?://.*\.jpe?g$',
'chapters': 'count:4',
'tags': 'count:23',
'episode': 'Episode 1',
'episode_number': 1,
'season': 'Season 9',
'season_number': 9,
'series': 'Duck Dynasty',
},
'params': {
# m3u8 download
'skip_download': True,
},
'add_ie': ['ThePlatform'],
'skip': 'This video is only available for users of participating TV providers.',
}, {
'url': 'http://www.fyi.tv/shows/tiny-house-nation/season-1/episode-8',
'only_matching': True
'only_matching': True,
}, {
'url': 'http://www.mylifetime.com/shows/project-runway-junior/season-1/episode-6',
'only_matching': True
'only_matching': True,
}, {
'url': 'http://www.mylifetime.com/movies/center-stage-on-pointe/full-movie',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://watch.lifetimemovieclub.com/movies/10-year-reunion/full-movie',
'only_matching': True
'only_matching': True,
}, {
'url': 'http://www.history.com/specials/sniper-into-the-kill-zone/full-special',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://www.aetv.com/specials/hunting-jonbenets-killer-the-untold-story/preview-hunting-jonbenets-killer-the-untold-story',
'only_matching': True
'only_matching': True,
}, {
'url': 'http://www.history.com/videos/history-of-valentines-day',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://play.aetv.com/shows/duck-dynasty/videos/best-of-duck-dynasty-getting-quack-in-shape',
'only_matching': True
'only_matching': True,
}]
def _real_extract(self, url):
@@ -180,14 +209,14 @@ class AENetworksListBaseIE(AENetworksBaseIE):
%s(slug: "%s") {
%s
}
}''' % (resource, slug, fields),
}''' % (resource, slug, fields), # noqa: UP031
}))['data'][resource]
def _real_extract(self, url):
domain, slug = self._match_valid_url(url).groups()
_, brand = self._DOMAIN_MAP[domain]
playlist = self._call_api(self._RESOURCE, slug, brand, self._FIELDS)
base_url = 'http://watch.%s' % domain
base_url = f'http://watch.{domain}'
entries = []
for item in (playlist.get(self._ITEMS_KEY) or []):
@@ -219,10 +248,10 @@ class AENetworksCollectionIE(AENetworksListBaseIE):
'playlist_mincount': 12,
}, {
'url': 'https://watch.historyvault.com/shows/america-the-story-of-us-2/season-1/list/america-the-story-of-us',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://www.historyvault.com/collections/mysteryquest',
'only_matching': True
'only_matching': True,
}]
_RESOURCE = 'list'
_ITEMS_KEY = 'items'
@@ -280,7 +309,7 @@ class HistoryTopicIE(AENetworksBaseIE):
'info_dict': {
'id': '40700995724',
'ext': 'mp4',
'title': "History of Valentine’s Day",
'title': 'History of Valentine’s Day',
'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7',
'timestamp': 1375819729,
'upload_date': '20130806',
@@ -303,6 +332,7 @@ class HistoryTopicIE(AENetworksBaseIE):
class HistoryPlayerIE(AENetworksBaseIE):
IE_NAME = 'history:player'
_VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:history|biography)\.com)/player/(?P<id>\d+)'
_TESTS = []
def _real_extract(self, url):
domain, video_id = self._match_valid_url(url).groups()
@@ -327,12 +357,13 @@ class BiographyIE(AENetworksBaseIE):
'skip_download': True,
},
'add_ie': ['ThePlatform'],
'skip': '404 Not Found',
}]
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
player_url = self._search_regex(
r'<phoenix-iframe[^>]+src="(%s)' % HistoryPlayerIE._VALID_URL,
rf'<phoenix-iframe[^>]+src="({HistoryPlayerIE._VALID_URL})',
webpage, 'player URL')
return self.url_result(player_url, HistoryPlayerIE.ie_key())
yt_dlp/extractor/aeonco.py
+47 -13
View File
@@ -1,5 +1,6 @@
from .common import InfoExtractor
from .vimeo import VimeoIE
from ..utils import ExtractorError, traverse_obj, url_or_none
class AeonCoIE(InfoExtractor):
@@ -15,26 +16,59 @@ class AeonCoIE(InfoExtractor):
'uploader': 'Semiconductor',
'uploader_id': 'semiconductor',
'uploader_url': 'https://vimeo.com/semiconductor',
'duration': 348
}
'duration': 348,
},
}, {
'url': 'https://aeon.co/videos/dazzling-timelapse-shows-how-microbes-spoil-our-food-and-sometimes-enrich-it',
'md5': '4e5f3dad9dbda0dbfa2da41a851e631e',
'md5': '03582d795382e49f2fd0b427b55de409',
'info_dict': {
'id': '728595228',
'id': '759576926',
'ext': 'mp4',
'title': 'Wrought',
'thumbnail': 'https://i.vimeocdn.com/video/1484618528-c91452611f9a4e4497735a533da60d45b2fe472deb0c880f0afaab0cd2efb22a-d_1280',
'uploader': 'Biofilm Productions',
'uploader_id': 'user140352216',
'uploader_url': 'https://vimeo.com/user140352216',
'duration': 1344
}
'thumbnail': 'https://i.vimeocdn.com/video/1525599692-84614af88e446612f49ca966cf8f80eab2c73376bedd80555741c521c26f9a3e-d_1280',
'uploader': 'Aeon Video',
'uploader_id': 'aeonvideo',
'uploader_url': 'https://vimeo.com/aeonvideo',
'duration': 1344,
},
}, {
'url': 'https://aeon.co/videos/chew-over-the-prisoners-dilemma-and-see-if-you-can-find-the-rational-path-out',
'md5': '1cfda0bf3ae24df17d00f2c0cb6cc21b',
'info_dict': {
'id': 'emyi4z-O0ls',
'ext': 'mp4',
'title': 'How to outsmart the Prisoner’s Dilemma - Lucas Husted',
'thumbnail': 'https://i.ytimg.com/vi_webp/emyi4z-O0ls/maxresdefault.webp',
'uploader': 'TED-Ed',
'uploader_id': '@TEDEd',
'uploader_url': 'https://www.youtube.com/@TEDEd',
'duration': 344,
'upload_date': '20200827',
'channel_id': 'UCsooa4yRKGN_zEE8iknghZA',
'playable_in_embed': True,
'description': 'md5:c0959524f08cb60f96fd010f3dfb17f3',
'categories': ['Education'],
'like_count': int,
'channel': 'TED-Ed',
'chapters': 'count:7',
'channel_url': 'https://www.youtube.com/channel/UCsooa4yRKGN_zEE8iknghZA',
'tags': 'count:26',
'availability': 'public',
'channel_follower_count': int,
'view_count': int,
'age_limit': 0,
'live_status': 'not_live',
'comment_count': int,
},
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
vimeo_id = self._search_regex(r'hosterId":\s*"(?P<id>[0-9]+)', webpage, 'vimeo id')
vimeo_url = VimeoIE._smuggle_referrer(f'https://player.vimeo.com/video/{vimeo_id}', 'https://aeon.co')
return self.url_result(vimeo_url, VimeoIE)
embed_url = traverse_obj(self._yield_json_ld(webpage, video_id), (
lambda _, v: v['@type'] == 'VideoObject', 'embedUrl', {url_or_none}), get_all=False)
if not embed_url:
raise ExtractorError('No embed URL found in webpage')
if 'player.vimeo.com' in embed_url:
embed_url = VimeoIE._smuggle_referrer(embed_url, 'https://aeon.co/')
return self.url_result(embed_url)
yt_dlp/extractor/afreecatv.py
+252 -312
View File
@@ -1,25 +1,73 @@
import functools
import re
from .common import InfoExtractor
from ..networking import Request
from ..utils import (
ExtractorError,
OnDemandPagedList,
date_from_str,
UserNotLive,
determine_ext,
filter_dict,
int_or_none,
qualities,
traverse_obj,
unified_strdate,
orderedSet,
unified_timestamp,
update_url_query,
url_or_none,
urlencode_postdata,
xpath_text,
urljoin,
)
from ..utils.traversal import traverse_obj
class AfreecaTVIE(InfoExtractor):
class AfreecaTVBaseIE(InfoExtractor):
_NETRC_MACHINE = 'afreecatv'
def _perform_login(self, username, password):
login_form = {
'szWork': 'login',
'szType': 'json',
'szUid': username,
'szPassword': password,
'isSaveId': 'false',
'szScriptVar': 'oLoginRet',
'szAction': '',
}
response = self._download_json(
'https://login.afreecatv.com/app/LoginAction.php', None,
'Logging in', data=urlencode_postdata(login_form))
_ERRORS = {
-4: 'Your account has been suspended due to a violation of our terms and policies.',
-5: 'https://member.afreecatv.com/app/user_delete_progress.php',
-6: 'https://login.afreecatv.com/membership/changeMember.php',
-8: "Hello! AfreecaTV here.\nThe username you have entered belongs to \n an account that requires a legal guardian's consent. \nIf you wish to use our services without restriction, \nplease make sure to go through the necessary verification process.",
-9: 'https://member.afreecatv.com/app/pop_login_block.php',
-11: 'https://login.afreecatv.com/afreeca/second_login.php',
-12: 'https://member.afreecatv.com/app/user_security.php',
0: 'The username does not exist or you have entered the wrong password.',
-1: 'The username does not exist or you have entered the wrong password.',
-3: 'You have entered your username/password incorrectly.',
-7: 'You cannot use your Global AfreecaTV account to access Korean AfreecaTV.',
-10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.',
-32008: 'You have failed to log in. Please contact our Help Center.',
}
result = int_or_none(response.get('RESULT'))
if result != 1:
error = _ERRORS.get(result, 'You have failed to log in.')
raise ExtractorError(
f'Unable to login: {self.IE_NAME} said: {error}',
expected=True)
def _call_api(self, endpoint, display_id, data=None, headers=None, query=None):
return self._download_json(Request(
f'https://api.m.afreecatv.com/{endpoint}',
data=data, headers=headers, query=query,
extensions={'legacy_ssl': True}), display_id,
'Downloading API JSON', 'Unable to download API JSON')
class AfreecaTVIE(AfreecaTVBaseIE):
IE_NAME = 'afreecatv'
IE_DESC = 'afreecatv.com'
_VALID_URL = r'''(?x)
@@ -32,9 +80,8 @@ class AfreecaTVIE(InfoExtractor):
)\?.*?\bnTitleNo=|
vod\.afreecatv\.com/(PLAYER/STATION|player)/
)
(?P<id>\d+)
(?P<id>\d+)/?(?:$|[?#&])
'''
_NETRC_MACHINE = 'afreecatv'
_TESTS = [{
'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
@@ -76,59 +123,6 @@ class AfreecaTVIE(InfoExtractor):
},
}],
'skip': 'Video is gone',
}, {
'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
'info_dict': {
'id': '18650793',
'ext': 'mp4',
'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
'thumbnail': r're:^https?://.*\.jpg$',
'uploader': '윈아디',
'uploader_id': 'badkids',
'duration': 107,
},
'params': {
'skip_download': True,
},
}, {
'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652',
'info_dict': {
'id': '10481652',
'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
'uploader': 'dailyapril',
'uploader_id': 'dailyapril',
'duration': 6492,
},
'playlist_count': 2,
'playlist': [{
'md5': 'd8b7c174568da61d774ef0203159bf97',
'info_dict': {
'id': '20160502_c4c62b9d_174361386_1',
'ext': 'mp4',
'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)",
'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
'uploader': 'dailyapril',
'uploader_id': 'dailyapril',
'upload_date': '20160502',
'duration': 3601,
},
}, {
'md5': '58f2ce7f6044e34439ab2d50612ab02b',
'info_dict': {
'id': '20160502_39e739bb_174361386_2',
'ext': 'mp4',
'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)",
'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
'uploader': 'dailyapril',
'uploader_id': 'dailyapril',
'upload_date': '20160502',
'duration': 2891,
},
}],
'params': {
'skip_download': True,
},
}, {
# non standard key
'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
@@ -140,14 +134,15 @@ class AfreecaTVIE(InfoExtractor):
'uploader': '♥이슬이',
'uploader_id': 'dasl8121',
'upload_date': '20170411',
'timestamp': 1491929865,
'duration': 213,
},
'params': {
'skip_download': True,
},
}, {
# PARTIAL_ADULT
'url': 'http://vod.afreecatv.com/PLAYER/STATION/32028439',
# adult content
'url': 'https://vod.afreecatv.com/player/97267690',
'info_dict': {
'id': '20180327_27901457_202289533_1',
'ext': 'mp4',
@@ -161,227 +156,151 @@ class AfreecaTVIE(InfoExtractor):
'params': {
'skip_download': True,
},
'expected_warnings': ['adult content'],
'skip': 'The VOD does not exist',
}, {
'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
'only_matching': True,
}, {
'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
'url': 'https://vod.afreecatv.com/player/96753363',
'info_dict': {
'id': '20230108_9FF5BEE1_244432674_1',
'ext': 'mp4',
'uploader_id': 'rlantnghks',
'uploader': '페이즈으',
'duration': 10840,
'thumbnail': r're:https?://videoimg\.afreecatv\.com/.+',
'upload_date': '20230108',
'timestamp': 1673218805,
'title': '젠지 페이즈',
},
'params': {
'skip_download': True,
},
}, {
# adult content
'url': 'https://vod.afreecatv.com/player/70395877',
'only_matching': True,
}, {
'url': 'http://vod.afreecatv.com/player/15055030',
# subscribers only
'url': 'https://vod.afreecatv.com/player/104647403',
'only_matching': True,
}, {
# private
'url': 'https://vod.afreecatv.com/player/81669846',
'only_matching': True,
}]
@staticmethod
def parse_video_key(key):
video_key = {}
m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
if m:
video_key['upload_date'] = m.group('upload_date')
video_key['part'] = int(m.group('part'))
return video_key
def _real_extract(self, url):
video_id = self._match_id(url)
data = self._call_api(
'station/video/a/view', video_id, headers={'Referer': url},
data=urlencode_postdata({
'nTitleNo': video_id,
'nApiLevel': 10,
}))['data']
def _perform_login(self, username, password):
login_form = {
'szWork': 'login',
'szType': 'json',
'szUid': username,
'szPassword': password,
'isSaveId': 'false',
'szScriptVar': 'oLoginRet',
'szAction': '',
}
error_code = traverse_obj(data, ('code', {int}))
if error_code == -6221:
raise ExtractorError('The VOD does not exist', expected=True)
elif error_code == -6205:
raise ExtractorError('This VOD is private', expected=True)
response = self._download_json(
'https://login.afreecatv.com/app/LoginAction.php', None,
'Logging in', data=urlencode_postdata(login_form))
common_info = traverse_obj(data, {
'title': ('title', {str}),
'uploader': ('writer_nick', {str}),
'uploader_id': ('bj_id', {str}),
'duration': ('total_file_duration', {functools.partial(int_or_none, scale=1000)}),
'thumbnail': ('thumb', {url_or_none}),
})
_ERRORS = {
-4: 'Your account has been suspended due to a violation of our terms and policies.',
-5: 'https://member.afreecatv.com/app/user_delete_progress.php',
-6: 'https://login.afreecatv.com/membership/changeMember.php',
-8: "Hello! AfreecaTV here.\nThe username you have entered belongs to \n an account that requires a legal guardian's consent. \nIf you wish to use our services without restriction, \nplease make sure to go through the necessary verification process.",
-9: 'https://member.afreecatv.com/app/pop_login_block.php',
-11: 'https://login.afreecatv.com/afreeca/second_login.php',
-12: 'https://member.afreecatv.com/app/user_security.php',
0: 'The username does not exist or you have entered the wrong password.',
-1: 'The username does not exist or you have entered the wrong password.',
-3: 'You have entered your username/password incorrectly.',
-7: 'You cannot use your Global AfreecaTV account to access Korean AfreecaTV.',
-10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.',
-32008: 'You have failed to log in. Please contact our Help Center.',
}
entries = []
for file_num, file_element in enumerate(
traverse_obj(data, ('files', lambda _, v: url_or_none(v['file']))), start=1):
file_url = file_element['file']
if determine_ext(file_url) == 'm3u8':
formats = self._extract_m3u8_formats(
file_url, video_id, 'mp4', m3u8_id='hls',
note=f'Downloading part {file_num} m3u8 information')
else:
formats = [{
'url': file_url,
'format_id': 'http',
}]
result = int_or_none(response.get('RESULT'))
if result != 1:
error = _ERRORS.get(result, 'You have failed to log in.')
raise ExtractorError(
'Unable to login: %s said: %s' % (self.IE_NAME, error),
expected=True)
entries.append({
**common_info,
'id': file_element.get('file_info_key') or f'{video_id}_{file_num}',
'title': f'{common_info.get("title") or "Untitled"} (part {file_num})',
'formats': formats,
**traverse_obj(file_element, {
'duration': ('duration', {functools.partial(int_or_none, scale=1000)}),
'timestamp': ('file_start', {unified_timestamp}),
}),
})
if traverse_obj(data, ('adult_status', {str})) == 'notLogin':
if not entries:
self.raise_login_required(
'Only users older than 19 are able to watch this video', method='password')
self.report_warning(
'In accordance with local laws and regulations, underage users are '
'restricted from watching adult content. Only content suitable for all '
f'ages will be downloaded. {self._login_hint("password")}')
if not entries and traverse_obj(data, ('sub_upload_type', {str})):
self.raise_login_required('This VOD is for subscribers only', method='password')
if len(entries) == 1:
return {
**entries[0],
'title': common_info.get('title'),
}
common_info['timestamp'] = traverse_obj(entries, (..., 'timestamp'), get_all=False)
return self.playlist_result(entries, video_id, multi_video=True, **common_info)
class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
IE_NAME = 'afreecatv:catchstory'
IE_DESC = 'afreecatv.com catch story'
_VALID_URL = r'https?://vod\.afreecatv\.com/player/(?P<id>\d+)/catchstory'
_TESTS = [{
'url': 'https://vod.afreecatv.com/player/103247/catchstory',
'info_dict': {
'id': '103247',
},
'playlist_count': 2,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
data = self._call_api(
'catchstory/a/view', video_id, headers={'Referer': url},
query={'aStoryListIdx': '', 'nStoryIdx': video_id})
webpage = self._download_webpage(url, video_id)
return self.playlist_result(self._entries(data), video_id)
if re.search(r'alert\(["\']This video has been deleted', webpage):
raise ExtractorError(
'Video %s has been deleted' % video_id, expected=True)
station_id = self._search_regex(
r'nStationNo\s*=\s*(\d+)', webpage, 'station')
bbs_id = self._search_regex(
r'nBbsNo\s*=\s*(\d+)', webpage, 'bbs')
video_id = self._search_regex(
r'nTitleNo\s*=\s*(\d+)', webpage, 'title', default=video_id)
partial_view = False
adult_view = False
for _ in range(2):
query = {
'nTitleNo': video_id,
'nStationNo': station_id,
'nBbsNo': bbs_id,
}
if partial_view:
query['partialView'] = 'SKIP_ADULT'
if adult_view:
query['adultView'] = 'ADULT_VIEW'
video_xml = self._download_xml(
'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
video_id, 'Downloading video info XML%s'
% (' (skipping adult)' if partial_view else ''),
video_id, headers={
'Referer': url,
}, query=query)
flag = xpath_text(video_xml, './track/flag', 'flag', default=None)
if flag and flag == 'SUCCEED':
break
if flag == 'PARTIAL_ADULT':
self.report_warning(
'In accordance with local laws and regulations, underage users are restricted from watching adult content. '
'Only content suitable for all ages will be downloaded. '
'Provide account credentials if you wish to download restricted content.')
partial_view = True
continue
elif flag == 'ADULT':
if not adult_view:
adult_view = True
continue
error = 'Only users older than 19 are able to watch this video. Provide account credentials to download this content.'
else:
error = flag
raise ExtractorError(
'%s said: %s' % (self.IE_NAME, error), expected=True)
else:
raise ExtractorError('Unable to download video info')
video_element = video_xml.findall('./track/video')[-1]
if video_element is None or video_element.text is None:
raise ExtractorError(
'Video %s does not exist' % video_id, expected=True)
video_url = video_element.text.strip()
title = xpath_text(video_xml, './track/title', 'title', fatal=True)
uploader = xpath_text(video_xml, './track/nickname', 'uploader')
uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
duration = int_or_none(xpath_text(
video_xml, './track/duration', 'duration'))
thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
common_entry = {
'uploader': uploader,
'uploader_id': uploader_id,
'thumbnail': thumbnail,
}
info = common_entry.copy()
info.update({
'id': video_id,
'title': title,
'duration': duration,
})
if not video_url:
entries = []
file_elements = video_element.findall('./file')
one = len(file_elements) == 1
for file_num, file_element in enumerate(file_elements, start=1):
file_url = url_or_none(file_element.text)
if not file_url:
continue
key = file_element.get('key', '')
upload_date = unified_strdate(self._search_regex(
r'^(\d{8})_', key, 'upload date', default=None))
if upload_date is not None:
# sometimes the upload date isn't included in the file name
# instead, another random ID is, which may parse as a valid
# date but be wildly out of a reasonable range
parsed_date = date_from_str(upload_date)
if parsed_date.year < 2000 or parsed_date.year >= 2100:
upload_date = None
file_duration = int_or_none(file_element.get('duration'))
format_id = key if key else '%s_%s' % (video_id, file_num)
if determine_ext(file_url) == 'm3u8':
formats = self._extract_m3u8_formats(
file_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls',
note='Downloading part %d m3u8 information' % file_num)
else:
formats = [{
'url': file_url,
'format_id': 'http',
}]
if not formats and not self.get_param('ignore_no_formats'):
continue
file_info = common_entry.copy()
file_info.update({
'id': format_id,
'title': title if one else '%s (part %d)' % (title, file_num),
'upload_date': upload_date,
'duration': file_duration,
'formats': formats,
})
entries.append(file_info)
entries_info = info.copy()
entries_info.update({
'_type': 'multi_video',
'entries': entries,
})
return entries_info
info = {
'id': video_id,
'title': title,
'uploader': uploader,
'uploader_id': uploader_id,
'duration': duration,
'thumbnail': thumbnail,
}
if determine_ext(video_url) == 'm3u8':
info['formats'] = self._extract_m3u8_formats(
video_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls')
else:
app, playpath = video_url.split('mp4:')
info.update({
'url': app,
'ext': 'flv',
'play_path': 'mp4:' + playpath,
'rtmp_live': True, # downloading won't end without this
})
return info
@staticmethod
def _entries(data):
# 'files' is always a list with 1 element
yield from traverse_obj(data, (
'data', lambda _, v: v['story_type'] == 'catch',
'catch_list', lambda _, v: v['files'][0]['file'], {
'id': ('files', 0, 'file_info_key', {str}),
'url': ('files', 0, 'file', {url_or_none}),
'duration': ('files', 0, 'duration', {functools.partial(int_or_none, scale=1000)}),
'title': ('title', {str}),
'uploader': ('writer_nick', {str}),
'uploader_id': ('writer_id', {str}),
'thumbnail': ('thumb', {url_or_none}),
'timestamp': ('write_timestamp', {int_or_none}),
}))
class AfreecaTVLiveIE(AfreecaTVIE): # XXX: Do not subclass from concrete IE
class AfreecaTVLiveIE(AfreecaTVBaseIE):
IE_NAME = 'afreecatv:live'
IE_DESC = 'afreecatv.com livestreams'
_VALID_URL = r'https?://play\.afreeca(?:tv)?\.com/(?P<id>[^/]+)(?:/(?P<bno>\d+))?'
_TESTS = [{
'url': 'https://play.afreecatv.com/pyh3646/237852185',
@@ -396,77 +315,97 @@ class AfreecaTVLiveIE(AfreecaTVIE): # XXX: Do not subclass from concrete IE
},
'skip': 'Livestream has ended',
}, {
'url': 'http://play.afreeca.com/pyh3646/237852185',
'url': 'https://play.afreecatv.com/pyh3646/237852185',
'only_matching': True,
}, {
'url': 'http://play.afreeca.com/pyh3646',
'url': 'https://play.afreecatv.com/pyh3646',
'only_matching': True,
}]
_LIVE_API_URL = 'https://live.afreecatv.com/afreeca/player_live_api.php'
_WORKING_CDNS = [
'gcp_cdn', # live-global-cdn-v02.afreecatv.com
'gs_cdn_pc_app', # pc-app.stream.afreecatv.com
'gs_cdn_mobile_web', # mobile-web.stream.afreecatv.com
'gs_cdn_pc_web', # pc-web.stream.afreecatv.com
]
_BAD_CDNS = [
'gs_cdn', # chromecast.afreeca.gscdn.com (cannot resolve)
'gs_cdn_chromecast', # chromecast.stream.afreecatv.com (HTTP Error 400)
'azure_cdn', # live-global-cdn-v01.afreecatv.com (cannot resolve)
'aws_cf', # live-global-cdn-v03.afreecatv.com (cannot resolve)
'kt_cdn', # kt.stream.afreecatv.com (HTTP Error 400)
]
_QUALITIES = ('sd', 'hd', 'hd2k', 'original')
def _extract_formats(self, channel_info, broadcast_no, aid):
stream_base_url = channel_info.get('RMD') or 'https://livestream-manager.afreecatv.com'
# If user has not passed CDN IDs, try API-provided CDN ID followed by other working CDN IDs
default_cdn_ids = orderedSet([
*traverse_obj(channel_info, ('CDN', {str}, all, lambda _, v: v not in self._BAD_CDNS)),
*self._WORKING_CDNS,
])
cdn_ids = self._configuration_arg('cdn', default_cdn_ids)
for attempt, cdn_id in enumerate(cdn_ids, start=1):
m3u8_url = traverse_obj(self._download_json(
urljoin(stream_base_url, 'broad_stream_assign.html'), broadcast_no,
f'Downloading {cdn_id} stream info', f'Unable to download {cdn_id} stream info',
fatal=False, query={
'return_type': cdn_id,
'broad_key': f'{broadcast_no}-common-master-hls',
}), ('view_url', {url_or_none}))
try:
return self._extract_m3u8_formats(
m3u8_url, broadcast_no, 'mp4', m3u8_id='hls', query={'aid': aid},
headers={'Referer': 'https://play.afreecatv.com/'})
except ExtractorError as e:
if attempt == len(cdn_ids):
raise
self.report_warning(
f'{e.cause or e.msg}. Retrying... (attempt {attempt} of {len(cdn_ids)})')
def _real_extract(self, url):
broadcaster_id, broadcast_no = self._match_valid_url(url).group('id', 'bno')
password = self.get_param('videopassword')
channel_info = traverse_obj(self._download_json(
self._LIVE_API_URL, broadcaster_id, data=urlencode_postdata({'bid': broadcaster_id})),
('CHANNEL', {dict})) or {}
info = self._download_json(self._LIVE_API_URL, broadcaster_id, fatal=False,
data=urlencode_postdata({'bid': broadcaster_id})) or {}
channel_info = info.get('CHANNEL') or {}
broadcaster_id = channel_info.get('BJID') or broadcaster_id
broadcast_no = channel_info.get('BNO') or broadcast_no
password_protected = channel_info.get('BPWD')
if not broadcast_no:
raise ExtractorError(f'Unable to extract broadcast number ({broadcaster_id} may not be live)', expected=True)
if password_protected == 'Y' and password is None:
raise UserNotLive(video_id=broadcaster_id)
password = self.get_param('videopassword')
if channel_info.get('BPWD') == 'Y' and password is None:
raise ExtractorError(
'This livestream is protected by a password, use the --video-password option',
expected=True)
formats = []
quality_key = qualities(self._QUALITIES)
for quality_str in self._QUALITIES:
params = {
token_info = traverse_obj(self._download_json(
self._LIVE_API_URL, broadcast_no, 'Downloading access token for stream',
'Unable to download access token for stream', data=urlencode_postdata(filter_dict({
'bno': broadcast_no,
'stream_type': 'common',
'type': 'aid',
'quality': quality_str,
}
if password is not None:
params['pwd'] = password
aid_response = self._download_json(
self._LIVE_API_URL, broadcast_no, fatal=False,
data=urlencode_postdata(params),
note=f'Downloading access token for {quality_str} stream',
errnote=f'Unable to download access token for {quality_str} stream')
aid = traverse_obj(aid_response, ('CHANNEL', 'AID'))
if not aid:
continue
'quality': 'master',
'pwd': password,
}))), ('CHANNEL', {dict})) or {}
aid = token_info.get('AID')
if not aid:
result = token_info.get('RESULT')
if result == 0:
raise ExtractorError('This livestream has ended', expected=True)
elif result == -6:
self.raise_login_required('This livestream is for subscribers only', method='password')
raise ExtractorError('Unable to extract access token')
stream_base_url = channel_info.get('RMD') or 'https://livestream-manager.afreecatv.com'
stream_info = self._download_json(
f'{stream_base_url}/broad_stream_assign.html', broadcast_no, fatal=False,
query={
'return_type': channel_info.get('CDN', 'gcp_cdn'),
'broad_key': f'{broadcast_no}-common-{quality_str}-hls',
},
note=f'Downloading metadata for {quality_str} stream',
errnote=f'Unable to download metadata for {quality_str} stream') or {}
formats = self._extract_formats(channel_info, broadcast_no, aid)
if stream_info.get('view_url'):
formats.append({
'format_id': quality_str,
'url': update_url_query(stream_info['view_url'], {'aid': aid}),
'ext': 'mp4',
'protocol': 'm3u8',
'quality': quality_key(quality_str),
})
station_info = self._download_json(
station_info = traverse_obj(self._download_json(
'https://st.afreecatv.com/api/get_station_status.php', broadcast_no,
query={'szBjId': broadcaster_id}, fatal=False,
note='Downloading channel metadata', errnote='Unable to download channel metadata') or {}
'Downloading channel metadata', 'Unable to download channel metadata',
query={'szBjId': broadcaster_id}, fatal=False), {dict}) or {}
return {
'id': broadcast_no,
@@ -476,6 +415,7 @@ class AfreecaTVLiveIE(AfreecaTVIE): # XXX: Do not subclass from concrete IE
'timestamp': unified_timestamp(station_info.get('broad_start')),
'formats': formats,
'is_live': True,
'http_headers': {'Referer': url},
}
yt_dlp/extractor/agora.py
+3 -3
View File
@@ -168,7 +168,7 @@ class TokFMPodcastIE(InfoExtractor):
for ext in ('aac', 'mp3'):
url_data = self._download_json(
f'https://api.podcast.radioagora.pl/api4/getSongUrl?podcast_id={media_id}&device_id={uuid.uuid4()}&ppre=false&audio={ext}',
media_id, 'Downloading podcast %s URL' % ext)
media_id, f'Downloading podcast {ext} URL')
# prevents inserting the mp3 (default) multiple times
if 'link_ssl' in url_data and f'.{ext}' in url_data['link_ssl']:
formats.append({
@@ -206,8 +206,8 @@ class TokFMAuditionIE(InfoExtractor):
}
@staticmethod
def _create_url(id):
return f'https://audycje.tokfm.pl/audycja/{id}'
def _create_url(video_id):
return f'https://audycje.tokfm.pl/audycja/{video_id}'
def _real_extract(self, url):
audition_id = self._match_id(url)
yt_dlp/extractor/airmozilla.py
-63
View File
@@ -1,63 +0,0 @@
import re
from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_duration,
parse_iso8601,
)
class AirMozillaIE(InfoExtractor):
_VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
_TEST = {
'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
'md5': '8d02f53ee39cf006009180e21df1f3ba',
'info_dict': {
'id': '6x4q2w',
'ext': 'mp4',
'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
'thumbnail': r're:https?://.*/poster\.jpg',
'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
'timestamp': 1422487800,
'upload_date': '20150128',
'location': 'SFO Commons',
'duration': 3780,
'view_count': int,
'categories': ['Main', 'Privacy'],
}
}
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
video_id = self._html_search_regex(r'//vid\.ly/(.*?)/embed', webpage, 'id')
embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
jwconfig = self._parse_json(self._search_regex(
r'initCallback\((.*)\);', embed_script, 'metadata'), video_id)['config']
info_dict = self._parse_jwplayer_data(jwconfig, video_id)
view_count = int_or_none(self._html_search_regex(
r'Views since archived: ([0-9]+)',
webpage, 'view count', fatal=False))
timestamp = parse_iso8601(self._html_search_regex(
r'<time datetime="(.*?)"', webpage, 'timestamp', fatal=False))
duration = parse_duration(self._search_regex(
r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
webpage, 'duration', fatal=False))
info_dict.update({
'id': video_id,
'title': self._og_search_title(webpage),
'url': self._og_search_url(webpage),
'display_id': display_id,
'description': self._og_search_description(webpage),
'timestamp': timestamp,
'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
'duration': duration,
'view_count': view_count,
'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
})
return info_dict
yt_dlp/extractor/airtv.py
+3 -3
View File
@@ -5,7 +5,7 @@ from ..utils import (
int_or_none,
mimetype2ext,
parse_iso8601,
traverse_obj
traverse_obj,
)
@@ -26,7 +26,7 @@ class AirTVIE(InfoExtractor):
'view_count': int,
'thumbnail': 'https://cdn-sp-gcs.air.tv/videos/W/8/W87jcWleSn2hXZN47zJZsQ/b13fc56464f47d9d62a36d110b9b5a72-4096x2160_9.jpg',
'timestamp': 1664792603,
}
},
}, {
# with youtube_id
'url': 'https://www.air.tv/watch?v=sv57EC8tRXG6h8dNXFUU1Q',
@@ -54,7 +54,7 @@ class AirTVIE(InfoExtractor):
'channel': 'Newsflare',
'duration': 37,
'upload_date': '20180511',
}
},
}]
def _get_formats_and_subtitle(self, json_data, video_id):
yt_dlp/extractor/aitube.py
+1 -1
View File
@@ -22,7 +22,7 @@ class AitubeKZVideoIE(InfoExtractor):
'timestamp': 1667370519,
'title': 'Ангел хранитель 1 серия',
'channel_follower_count': int,
}
},
}, {
# embed url
'url': 'https://aitube.kz/embed/?id=9291d29b-c038-49a1-ad42-3da2051d353c',
yt_dlp/extractor/aliexpress.py
+1 -2
View File
@@ -1,5 +1,4 @@
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
float_or_none,
try_get,
@@ -44,7 +43,7 @@ class AliExpressLiveIE(InfoExtractor):
'title': title,
'thumbnail': data.get('coverUrl'),
'uploader': try_get(
data, lambda x: x['followBar']['name'], compat_str),
data, lambda x: x['followBar']['name'], str),
'timestamp': float_or_none(data.get('startTimeLong'), scale=1000),
'formats': formats,
}
yt_dlp/extractor/aljazeera.py
+7 -7
View File
@@ -18,7 +18,7 @@ class AlJazeeraIE(InfoExtractor):
'timestamp': 1636219149,
'description': 'U sarajevskim naseljima Rajlovac i Reljevo stambeni objekti, ali i industrijska postrojenja i dalje su pod vodom.',
'upload_date': '20211106',
}
},
}, {
'url': 'https://balkans.aljazeera.net/videos/2021/11/6/djokovic-usao-u-finale-mastersa-u-parizu',
'info_dict': {
@@ -33,7 +33,7 @@ class AlJazeeraIE(InfoExtractor):
BRIGHTCOVE_URL_RE = r'https?://players.brightcove.net/(?P<account>\d+)/(?P<player_id>[a-zA-Z0-9]+)_(?P<embed>[^/]+)/index.html\?videoId=(?P<id>\d+)'
def _real_extract(self, url):
base, post_type, id = self._match_valid_url(url).groups()
base, post_type, display_id = self._match_valid_url(url).groups()
wp = {
'balkans.aljazeera.net': 'ajb',
'chinese.aljazeera.net': 'chinese',
@@ -47,11 +47,11 @@ class AlJazeeraIE(InfoExtractor):
'news': 'news',
}[post_type.split('/')[0]]
video = self._download_json(
f'https://{base}/graphql', id, query={
f'https://{base}/graphql', display_id, query={
'wp-site': wp,
'operationName': 'ArchipelagoSingleArticleQuery',
'variables': json.dumps({
'name': id,
'name': display_id,
'postType': post_type,
}),
}, headers={
@@ -64,7 +64,7 @@ class AlJazeeraIE(InfoExtractor):
embed = 'default'
if video_id is None:
webpage = self._download_webpage(url, id)
webpage = self._download_webpage(url, display_id)
account, player_id, embed, video_id = self._search_regex(self.BRIGHTCOVE_URL_RE, webpage, 'video id',
group=(1, 2, 3, 4), default=(None, None, None, None))
@@ -73,11 +73,11 @@ class AlJazeeraIE(InfoExtractor):
return {
'_type': 'url_transparent',
'url': url,
'ie_key': 'Generic'
'ie_key': 'Generic',
}
return {
'_type': 'url_transparent',
'url': f'https://players.brightcove.net/{account}/{player_id}_{embed}/index.html?videoId={video_id}',
'ie_key': 'BrightcoveNew'
'ie_key': 'BrightcoveNew',
}
yt_dlp/extractor/allocine.py
+2 -3
View File
@@ -1,5 +1,4 @@
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
int_or_none,
qualities,
@@ -95,11 +94,11 @@ class AllocineIE(InfoExtractor):
duration = int_or_none(video.get('duration'))
view_count = int_or_none(video.get('view_count'))
timestamp = unified_timestamp(try_get(
video, lambda x: x['added_at']['date'], compat_str))
video, lambda x: x['added_at']['date'], str))
else:
video_id = display_id
media_data = self._download_json(
'http://www.allocine.fr/ws/AcVisiondataV5.ashx?media=%s' % video_id, display_id)
f'http://www.allocine.fr/ws/AcVisiondataV5.ashx?media={video_id}', display_id)
title = remove_end(strip_or_none(self._html_extract_title(webpage), ' - AlloCiné'))
for key, value in media_data['video'].items():
if not key.endswith('Path'):
yt_dlp/extractor/allstar.py
+252
View File
@@ -0,0 +1,252 @@
import functools
import json
from .common import InfoExtractor
from ..utils import (
ExtractorError,
OnDemandPagedList,
int_or_none,
join_nonempty,
parse_qs,
urljoin,
)
from ..utils.traversal import traverse_obj
_FIELDS = '''
_id
clipImageSource
clipImageThumb
clipLink
clipTitle
createdDate
shareId
user { _id }
username
views'''
_EXTRA_FIELDS = '''
clipLength
clipSizeBytes'''
_QUERIES = {
'clip': '''query ($id: String!) {
video: getClip(clipIdentifier: $id) {
%s %s
}
}''' % (_FIELDS, _EXTRA_FIELDS), # noqa: UP031
'montage': '''query ($id: String!) {
video: getMontage(clipIdentifier: $id) {
%s
}
}''' % _FIELDS, # noqa: UP031
'Clips': '''query ($page: Int!, $user: String!, $game: Int) {
videos: clips(search: createdDate, page: $page, user: $user, mobile: false, game: $game) {
data { %s %s }
}
}''' % (_FIELDS, _EXTRA_FIELDS), # noqa: UP031
'Montages': '''query ($page: Int!, $user: String!) {
videos: montages(search: createdDate, page: $page, user: $user) {
data { %s }
}
}''' % _FIELDS, # noqa: UP031
'Mobile Clips': '''query ($page: Int!, $user: String!) {
videos: clips(search: createdDate, page: $page, user: $user, mobile: true) {
data { %s %s }
}
}''' % (_FIELDS, _EXTRA_FIELDS), # noqa: UP031
}
class AllstarBaseIE(InfoExtractor):
@staticmethod
def _parse_video_data(video_data):
def media_url_or_none(path):
return urljoin('https://media.allstar.gg/', path)
info = traverse_obj(video_data, {
'id': ('_id', {str}),
'display_id': ('shareId', {str}),
'title': ('clipTitle', {str}),
'url': ('clipLink', {media_url_or_none}),
'thumbnails': (('clipImageThumb', 'clipImageSource'), {'url': {media_url_or_none}}),
'duration': ('clipLength', {int_or_none}),
'filesize': ('clipSizeBytes', {int_or_none}),
'timestamp': ('createdDate', {functools.partial(int_or_none, scale=1000)}),
'uploader': ('username', {str}),
'uploader_id': ('user', '_id', {str}),
'view_count': ('views', {int_or_none}),
})
if info.get('id') and info.get('url'):
basename = 'clip' if '/clips/' in info['url'] else 'montage'
info['webpage_url'] = f'https://allstar.gg/{basename}?{basename}={info["id"]}'
info.update({
'extractor_key': AllstarIE.ie_key(),
'extractor': AllstarIE.IE_NAME,
'uploader_url': urljoin('https://allstar.gg/u/', info.get('uploader_id')),
})
return info
def _call_api(self, query, variables, path, video_id=None, note=None):
response = self._download_json(
'https://a1.allstar.gg/graphql', video_id, note=note,
headers={'content-type': 'application/json'},
data=json.dumps({'variables': variables, 'query': query}).encode())
errors = traverse_obj(response, ('errors', ..., 'message', {str}))
if errors:
raise ExtractorError('; '.join(errors))
return traverse_obj(response, path)
class AllstarIE(AllstarBaseIE):
_VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?P<type>(?:clip|montage))\?(?P=type)=(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://allstar.gg/clip?clip=64482c2da9eec30008a67d1b',
'info_dict': {
'id': '64482c2da9eec30008a67d1b',
'title': '4K on Inferno',
'url': 'md5:66befb5381eef0c9456026386c25fa55',
'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
'uploader': 'chrk.',
'ext': 'mp4',
'duration': 20,
'filesize': 21199257,
'timestamp': 1682451501,
'uploader_id': '62b8bdfc9021052f7905882d',
'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
'upload_date': '20230425',
'view_count': int,
},
}, {
'url': 'https://allstar.gg/clip?clip=8LJLY4JKB',
'info_dict': {
'id': '64a1ec6b887f4c0008dc50b8',
'display_id': '8LJLY4JKB',
'title': 'AK-47 3K on Mirage',
'url': 'md5:dde224fd12f035c0e2529a4ae34c4283',
'ext': 'mp4',
'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
'duration': 16,
'filesize': 30175859,
'timestamp': 1688333419,
'uploader': 'cherokee',
'uploader_id': '62b8bdfc9021052f7905882d',
'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
'upload_date': '20230702',
'view_count': int,
},
}, {
'url': 'https://allstar.gg/montage?montage=643e64089da7e9363e1fa66c',
'info_dict': {
'id': '643e64089da7e9363e1fa66c',
'display_id': 'APQLGM2IMXW',
'title': 'cherokee Rapid Fire Snipers Montage',
'url': 'md5:a3ee356022115db2b27c81321d195945',
'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
'ext': 'mp4',
'timestamp': 1681810448,
'uploader': 'cherokee',
'uploader_id': '62b8bdfc9021052f7905882d',
'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
'upload_date': '20230418',
'view_count': int,
},
}, {
'url': 'https://allstar.gg/montage?montage=RILJMH6QOS',
'info_dict': {
'id': '64a2697372ce3703de29e868',
'display_id': 'RILJMH6QOS',
'title': 'cherokee Rapid Fire Snipers Montage',
'url': 'md5:d5672e6f88579730c2310a80fdbc4030',
'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
'ext': 'mp4',
'timestamp': 1688365434,
'uploader': 'cherokee',
'uploader_id': '62b8bdfc9021052f7905882d',
'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
'upload_date': '20230703',
'view_count': int,
},
}]
def _real_extract(self, url):
query_id, video_id = self._match_valid_url(url).group('type', 'id')
return self._parse_video_data(
self._call_api(
_QUERIES.get(query_id), {'id': video_id}, ('data', 'video'), video_id))
class AllstarProfileIE(AllstarBaseIE):
_VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?:profile\?user=|u/)(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://allstar.gg/profile?user=62b8bdfc9021052f7905882d',
'info_dict': {
'id': '62b8bdfc9021052f7905882d-clips',
'title': 'cherokee - Clips',
},
'playlist_mincount': 15,
}, {
'url': 'https://allstar.gg/u/cherokee?game=730&view=Clips',
'info_dict': {
'id': '62b8bdfc9021052f7905882d-clips-730',
'title': 'cherokee - Clips - 730',
},
'playlist_mincount': 15,
}, {
'url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d?view=Montages',
'info_dict': {
'id': '62b8bdfc9021052f7905882d-montages',
'title': 'cherokee - Montages',
},
'playlist_mincount': 4,
}, {
'url': 'https://allstar.gg/profile?user=cherokee&view=Mobile Clips',
'info_dict': {
'id': '62b8bdfc9021052f7905882d-mobile',
'title': 'cherokee - Mobile Clips',
},
'playlist_mincount': 1,
}]
_PAGE_SIZE = 10
def _get_page(self, user_id, display_id, game, query, page_num):
page_num += 1
for video_data in self._call_api(
query, {
'user': user_id,
'page': page_num,
'game': game,
}, ('data', 'videos', 'data'), display_id, f'Downloading page {page_num}'):
yield self._parse_video_data(video_data)
def _real_extract(self, url):
display_id = self._match_id(url)
profile_data = self._download_json(
urljoin('https://api.allstar.gg/v1/users/profile/', display_id), display_id)
user_id = traverse_obj(profile_data, ('data', ('_id'), {str}))
if not user_id:
raise ExtractorError('Unable to extract the user id')
username = traverse_obj(profile_data, ('data', 'profile', ('username'), {str}))
url_query = parse_qs(url)
game = traverse_obj(url_query, ('game', 0, {int_or_none}))
query_id = traverse_obj(url_query, ('view', 0), default='Clips')
if query_id not in ('Clips', 'Montages', 'Mobile Clips'):
raise ExtractorError(f'Unsupported playlist URL type {query_id!r}')
return self.playlist_result(
OnDemandPagedList(
functools.partial(
self._get_page, user_id, display_id, game, _QUERIES.get(query_id)), self._PAGE_SIZE),
playlist_id=join_nonempty(user_id, query_id.lower().split()[0], game),
playlist_title=join_nonempty((username or display_id), query_id, game, delim=' - '))
yt_dlp/extractor/alphaporno.py
+3 -3
View File
@@ -1,9 +1,9 @@
from .common import InfoExtractor
from ..utils import (
parse_iso8601,
int_or_none,
parse_duration,
parse_filesize,
int_or_none,
parse_iso8601,
)
@@ -25,7 +25,7 @@ class AlphaPornoIE(InfoExtractor):
'tbr': 1145,
'categories': list,
'age_limit': 18,
}
},
}
def _real_extract(self, url):
yt_dlp/extractor/alsace20tv.py
+3 -3
View File
@@ -12,7 +12,7 @@ from ..utils import (
class Alsace20TVBaseIE(InfoExtractor):
def _extract_video(self, video_id, url=None):
info = self._download_json(
'https://www.alsace20.tv/visionneuse/visio_v9_js.php?key=%s&habillage=0&mode=html' % (video_id, ),
f'https://www.alsace20.tv/visionneuse/visio_v9_js.php?key={video_id}&habillage=0&mode=html',
video_id) or {}
title = info.get('titre')
@@ -24,9 +24,9 @@ class Alsace20TVBaseIE(InfoExtractor):
else self._extract_mpd_formats(fmt_url, video_id, mpd_id=res, fatal=False))
webpage = (url and self._download_webpage(url, video_id, fatal=False)) or ''
thumbnail = url_or_none(dict_get(info, ('image', 'preview', )) or self._og_search_thumbnail(webpage))
thumbnail = url_or_none(dict_get(info, ('image', 'preview')) or self._og_search_thumbnail(webpage))
upload_date = self._search_regex(r'/(\d{6})_', thumbnail, 'upload_date', default=None)
upload_date = unified_strdate('20%s-%s-%s' % (upload_date[:2], upload_date[2:4], upload_date[4:])) if upload_date else None
upload_date = unified_strdate(f'20{upload_date[:2]}-{upload_date[2:4]}-{upload_date[4:]}') if upload_date else None
return {
'id': video_id,
'title': title,
yt_dlp/extractor/altcensored.py
+104
View File
@@ -0,0 +1,104 @@
import re
from .archiveorg import ArchiveOrgIE
from .common import InfoExtractor
from ..utils import (
InAdvancePagedList,
clean_html,
int_or_none,
orderedSet,
str_to_int,
urljoin,
)
class AltCensoredIE(InfoExtractor):
IE_NAME = 'altcensored'
_VALID_URL = r'https?://(?:www\.)?altcensored\.com/(?:watch\?v=|embed/)(?P<id>[^/?#]+)'
_TESTS = [{
'url': 'https://www.altcensored.com/watch?v=k0srjLSkga8',
'info_dict': {
'id': 'youtube-k0srjLSkga8',
'ext': 'webm',
'title': "QUELLES SONT LES CONSÉQUENCES DE L'HYPERSEXUALISATION DE LA SOCIÉTÉ ?",
'display_id': 'k0srjLSkga8.webm',
'release_date': '20180403',
'creators': ['Virginie Vota'],
'release_year': 2018,
'upload_date': '20230318',
'uploader': 'admin@altcensored.com',
'description': 'md5:0b38a8fc04103579d5c1db10a247dc30',
'timestamp': 1679161343,
'track': 'k0srjLSkga8',
'duration': 926.09,
'thumbnail': 'https://archive.org/download/youtube-k0srjLSkga8/youtube-k0srjLSkga8.thumbs/k0srjLSkga8_000925.jpg',
'view_count': int,
'categories': ['News & Politics'],
},
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
category = clean_html(self._html_search_regex(
r'<a href="/category/\d+">([^<]+)</a>', webpage, 'category', default=None))
return {
'_type': 'url_transparent',
'url': f'https://archive.org/details/youtube-{video_id}',
'ie_key': ArchiveOrgIE.ie_key(),
'view_count': str_to_int(self._html_search_regex(
r'YouTube Views:(?:\s|&nbsp;)*([\d,]+)', webpage, 'view count', default=None)),
'categories': [category] if category else None,
}
class AltCensoredChannelIE(InfoExtractor):
IE_NAME = 'altcensored:channel'
_VALID_URL = r'https?://(?:www\.)?altcensored\.com/channel/(?!page|table)(?P<id>[^/?#]+)'
_PAGE_SIZE = 24
_TESTS = [{
'url': 'https://www.altcensored.com/channel/UCFPTO55xxHqFqkzRZHu4kcw',
'info_dict': {
'title': 'Virginie Vota',
'id': 'UCFPTO55xxHqFqkzRZHu4kcw',
},
'playlist_count': 85,
}, {
'url': 'https://altcensored.com/channel/UC9CcJ96HKMWn0LZlcxlpFTw',
'info_dict': {
'title': 'yukikaze775',
'id': 'UC9CcJ96HKMWn0LZlcxlpFTw',
},
'playlist_count': 4,
}, {
'url': 'https://altcensored.com/channel/UCfYbb7nga6-icsFWWgS-kWw',
'info_dict': {
'title': 'Mister Metokur',
'id': 'UCfYbb7nga6-icsFWWgS-kWw',
},
'playlist_count': 121,
}]
def _real_extract(self, url):
channel_id = self._match_id(url)
webpage = self._download_webpage(
url, channel_id, 'Download channel webpage', 'Unable to get channel webpage')
title = self._html_search_meta('altcen_title', webpage, 'title', fatal=False)
page_count = int_or_none(self._html_search_regex(
r'<a[^>]+href="/channel/[\w-]+/page/(\d+)">(?:\1)</a>',
webpage, 'page count', default='1'))
def page_func(page_num):
page_num += 1
webpage = self._download_webpage(
f'https://altcensored.com/channel/{channel_id}/page/{page_num}',
channel_id, note=f'Downloading page {page_num}')
items = re.findall(r'<a[^>]+href="(/watch\?v=[^"]+)', webpage)
return [self.url_result(urljoin('https://www.altcensored.com', path), AltCensoredIE)
for path in orderedSet(items)]
return self.playlist_result(
InAdvancePagedList(page_func, page_count, self._PAGE_SIZE),
playlist_id=channel_id, playlist_title=title)
yt_dlp/extractor/alura.py
+13 -17
View File
@@ -1,17 +1,13 @@
import re
import urllib.parse
from .common import InfoExtractor
from ..compat import (
compat_urlparse,
)
from ..utils import (
ExtractorError,
clean_html,
int_or_none,
urlencode_postdata,
urljoin,
int_or_none,
clean_html,
ExtractorError
)
@@ -25,7 +21,7 @@ class AluraIE(InfoExtractor):
'info_dict': {
'id': '60095',
'ext': 'mp4',
'title': 'Referências, ref-set e alter'
'title': 'Referências, ref-set e alter',
},
'skip': 'Requires alura account credentials'},
{
@@ -34,12 +30,12 @@ class AluraIE(InfoExtractor):
'only_matching': True},
{
'url': 'https://cursos.alura.com.br/course/fundamentos-market-digital/task/55219',
'only_matching': True}
'only_matching': True},
]
def _real_extract(self, url):
course, video_id = self._match_valid_url(url)
course, video_id = self._match_valid_url(url).group('course_name', 'id')
video_url = self._VIDEO_URL % (course, video_id)
video_dict = self._download_json(video_url, video_id, 'Searching for videos')
@@ -52,7 +48,7 @@ class AluraIE(InfoExtractor):
formats = []
for video_obj in video_dict:
video_url_m3u8 = video_obj.get('link')
video_url_m3u8 = video_obj.get('mp4')
video_format = self._extract_m3u8_formats(
video_url_m3u8, None, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False)
@@ -66,7 +62,7 @@ class AluraIE(InfoExtractor):
return {
'id': video_id,
'title': video_title,
"formats": formats
'formats': formats,
}
def _perform_login(self, username, password):
@@ -95,7 +91,7 @@ class AluraIE(InfoExtractor):
'post url', default=self._LOGIN_URL, group='url')
if not post_url.startswith('http'):
post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
post_url = urllib.parse.urljoin(self._LOGIN_URL, post_url)
response = self._download_webpage(
post_url, None, 'Logging in',
@@ -107,7 +103,7 @@ class AluraIE(InfoExtractor):
r'(?s)<p[^>]+class="alert-message[^"]*">(.+?)</p>',
response, 'error message', default=None)
if error:
raise ExtractorError('Unable to login: %s' % error, expected=True)
raise ExtractorError(f'Unable to login: {error}', expected=True)
raise ExtractorError('Unable to log in')
@@ -123,7 +119,7 @@ class AluraCourseIE(AluraIE): # XXX: Do not subclass from concrete IE
@classmethod
def suitable(cls, url):
return False if AluraIE.suitable(url) else super(AluraCourseIE, cls).suitable(url)
return False if AluraIE.suitable(url) else super().suitable(url)
def _real_extract(self, url):
@@ -161,7 +157,7 @@ class AluraCourseIE(AluraIE): # XXX: Do not subclass from concrete IE
'url': video_url,
'id_key': self.ie_key(),
'chapter': chapter,
'chapter_number': chapter_number
'chapter_number': chapter_number,
}
entries.append(entry)
return self.playlist_result(entries, course_path, course_title)
yt_dlp/extractor/amadeustv.py
+77
View File
@@ -0,0 +1,77 @@
from .common import InfoExtractor
from ..utils import (
ExtractorError,
float_or_none,
int_or_none,
parse_iso8601,
url_or_none,
)
from ..utils.traversal import traverse_obj
class AmadeusTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?amadeus\.tv/library/(?P<id>[\da-f]+)'
_TESTS = [{
'url': 'http://www.amadeus.tv/library/65091a87ff85af59d9fc54c3',
'info_dict': {
'id': '5576678021301411311',
'ext': 'mp4',
'title': 'Jieon Park - 第五届珠海莫扎特国际青少年音乐周小提琴C组第三轮',
'thumbnail': 'http://1253584441.vod2.myqcloud.com/a0046a27vodtransbj1253584441/7db4af535576678021301411311/coverBySnapshot_10_0.jpg',
'duration': 1264.8,
'upload_date': '20230918',
'timestamp': 1695034800,
'display_id': '65091a87ff85af59d9fc54c3',
'view_count': int,
'description': 'md5:a0357b9c215489e2067cbae0b777bb95',
},
}]
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
nuxt_data = self._search_nuxt_data(webpage, display_id, traverse=('fetch', '0'))
video_id = traverse_obj(nuxt_data, ('item', 'video', {str}))
if not video_id:
raise ExtractorError('Unable to extract actual video ID')
video_data = self._download_json(
f'http://playvideo.qcloud.com/getplayinfo/v2/1253584441/{video_id}',
video_id, headers={'Referer': 'http://www.amadeus.tv/'})
formats = []
for video in traverse_obj(video_data, ('videoInfo', ('sourceVideo', ('transcodeList', ...)), {dict})):
if not url_or_none(video.get('url')):
continue
formats.append({
**traverse_obj(video, {
'url': 'url',
'format_id': ('definition', {lambda x: f'http-{x or "0"}'}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
'filesize': (('totalSize', 'size'), {int_or_none}),
'vcodec': ('videoStreamList', 0, 'codec'),
'acodec': ('audioStreamList', 0, 'codec'),
'fps': ('videoStreamList', 0, 'fps', {float_or_none}),
}, get_all=False),
'http_headers': {'Referer': 'http://www.amadeus.tv/'},
})
return {
'id': video_id,
'display_id': display_id,
'formats': formats,
**traverse_obj(video_data, {
'title': ('videoInfo', 'basicInfo', 'name', {str}),
'thumbnail': ('coverInfo', 'coverUrl', {url_or_none}),
'duration': ('videoInfo', 'sourceVideo', ('floatDuration', 'duration'), {float_or_none}),
}, get_all=False),
**traverse_obj(nuxt_data, ('item', {
'title': (('title', 'title_en', 'title_cn'), {str}),
'description': (('description', 'description_en', 'description_cn'), {str}),
'timestamp': ('date', {parse_iso8601}),
'view_count': ('view', {int_or_none}),
}), get_all=False),
}
yt_dlp/extractor/amara.py
+6 -6
View File
@@ -1,6 +1,6 @@
from .common import InfoExtractor
from .youtube import YoutubeIE
from .vimeo import VimeoIE
from .youtube import YoutubeIE
from ..utils import (
int_or_none,
parse_iso8601,
@@ -25,7 +25,7 @@ class AmaraIE(InfoExtractor):
'uploader': 'PBS NewsHour',
'uploader_id': 'PBSNewsHour',
'timestamp': 1549639570,
}
},
}, {
# Vimeo
'url': 'https://amara.org/en/videos/kYkK1VUTWW5I/info/vimeo-at-ces-2011',
@@ -40,8 +40,8 @@ class AmaraIE(InfoExtractor):
'timestamp': 1294763658,
'upload_date': '20110111',
'uploader': 'Sam Morrill',
'uploader_id': 'sammorrill'
}
'uploader_id': 'sammorrill',
},
}, {
# Direct Link
'url': 'https://amara.org/en/videos/s8KL7I3jLmh6/info/the-danger-of-a-single-story/',
@@ -55,13 +55,13 @@ class AmaraIE(InfoExtractor):
'subtitles': dict,
'upload_date': '20091007',
'timestamp': 1254942511,
}
},
}]
def _real_extract(self, url):
video_id = self._match_id(url)
meta = self._download_json(
'https://amara.org/api/videos/%s/' % video_id,
f'https://amara.org/api/videos/{video_id}/',
video_id, query={'format': 'json'})
title = meta['title']
video_url = meta['all_urls'][0]
yt_dlp/extractor/amazon.py
+4 -4
View File
@@ -61,13 +61,13 @@ class AmazonStoreIE(InfoExtractor):
}]
def _real_extract(self, url):
id = self._match_id(url)
playlist_id = self._match_id(url)
for retry in self.RetryManager():
webpage = self._download_webpage(url, id)
webpage = self._download_webpage(url, playlist_id)
try:
data_json = self._search_json(
r'var\s?obj\s?=\s?jQuery\.parseJSON\(\'', webpage, 'data', id,
r'var\s?obj\s?=\s?jQuery\.parseJSON\(\'', webpage, 'data', playlist_id,
transform_source=js_to_json)
except ExtractorError as e:
retry.error = e
@@ -81,7 +81,7 @@ class AmazonStoreIE(InfoExtractor):
'height': int_or_none(video.get('videoHeight')),
'width': int_or_none(video.get('videoWidth')),
} for video in (data_json.get('videos') or []) if video.get('isVideo') and video.get('url')]
return self.playlist_result(entries, playlist_id=id, playlist_title=data_json.get('title'))
return self.playlist_result(entries, playlist_id=playlist_id, playlist_title=data_json.get('title'))
class AmazonReviewsIE(InfoExtractor):
yt_dlp/extractor/amazonminitv.py
+7 -4
View File
@@ -22,8 +22,11 @@ class AmazonMiniTVBaseIE(InfoExtractor):
resp = self._download_json(
f'https://www.amazon.in/minitv/api/web/{"graphql" if data else "prs"}',
asin, note=note, headers={'Content-Type': 'application/json'},
data=json.dumps(data).encode() if data else None,
asin, note=note, headers={
'Content-Type': 'application/json',
'currentpageurl': '/',
'currentplatform': 'dWeb',
}, data=json.dumps(data).encode() if data else None,
query=None if data else {
'deviceType': 'A1WMMUXPCUJL4N',
'contentId': asin,
@@ -46,7 +49,7 @@ class AmazonMiniTVIE(AmazonMiniTVBaseIE):
'ext': 'mp4',
'title': 'May I Kiss You?',
'language': 'Hindi',
'thumbnail': r're:^https?://.*\.jpg$',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'description': 'md5:a549bfc747973e04feb707833474e59d',
'release_timestamp': 1644710400,
'release_date': '20220213',
@@ -68,7 +71,7 @@ class AmazonMiniTVIE(AmazonMiniTVBaseIE):
'ext': 'mp4',
'title': 'Jahaan',
'language': 'Hindi',
'thumbnail': r're:^https?://.*\.jpg',
'thumbnail': r're:^https?://.*\.(?:jpg|png)',
'description': 'md5:05eb765a77bf703f322f120ec6867339',
'release_timestamp': 1647475200,
'release_date': '20220317',
yt_dlp/extractor/amcnetworks.py
+7 -6
View File
@@ -26,6 +26,7 @@ class AMCNetworksIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
# m3u8 download
'skip_download': True,
},
'skip': '404 Not Found',
}, {
'url': 'http://www.bbcamerica.com/shows/the-hunt/full-episodes/season-1/episode-01-the-hardest-challenge',
'only_matching': True,
@@ -63,8 +64,8 @@ class AMCNetworksIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
site, display_id = self._match_valid_url(url).groups()
requestor_id = self._REQUESTOR_ID_MAP[site]
page_data = self._download_json(
'https://content-delivery-gw.svc.ds.amcn.com/api/v2/content/amcn/%s/url/%s'
% (requestor_id.lower(), display_id), display_id)['data']
f'https://content-delivery-gw.svc.ds.amcn.com/api/v2/content/amcn/{requestor_id.lower()}/url/{display_id}',
display_id)['data']
properties = page_data.get('properties') or {}
query = {
'mbr': 'true',
@@ -75,15 +76,15 @@ class AMCNetworksIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
try:
for v in page_data['children']:
if v.get('type') == 'video-player':
releasePid = v['properties']['currentVideo']['meta']['releasePid']
tp_path = 'M_UwQC/' + releasePid
release_pid = v['properties']['currentVideo']['meta']['releasePid']
tp_path = 'M_UwQC/' + release_pid
media_url = 'https://link.theplatform.com/s/' + tp_path
video_player_count += 1
except KeyError:
pass
if video_player_count > 1:
self.report_warning(
'The JSON data has %d video players. Only one will be extracted' % video_player_count)
f'The JSON data has {video_player_count} video players. Only one will be extracted')
# Fall back to videoPid if releasePid not found.
# TODO: Fall back to videoPid if releasePid manifest uses DRM.
@@ -130,7 +131,7 @@ class AMCNetworksIE(ThePlatformIE): # XXX: Do not subclass from concrete IE
})
ns_keys = theplatform_metadata.get('$xmlns', {}).keys()
if ns_keys:
ns = list(ns_keys)[0]
ns = next(iter(ns_keys))
episode = theplatform_metadata.get(ns + '$episodeTitle') or None
episode_number = int_or_none(
theplatform_metadata.get(ns + '$episode'))
yt_dlp/extractor/americastestkitchen.py
+7 -7
View File
@@ -87,13 +87,13 @@ class AmericasTestKitchenIE(InfoExtractor):
resource_type = 'episodes'
resource = self._download_json(
'https://www.americastestkitchen.com/api/v6/%s/%s' % (resource_type, video_id), video_id)
f'https://www.americastestkitchen.com/api/v6/{resource_type}/{video_id}', video_id)
video = resource['video'] if is_episode else resource
episode = resource if is_episode else resource.get('episode') or {}
return {
'_type': 'url_transparent',
'url': 'https://player.zype.com/embed/%s.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ' % video['zypeId'],
'url': 'https://player.zype.com/embed/{}.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ'.format(video['zypeId']),
'ie_key': 'Zype',
'description': clean_html(video.get('description')),
'timestamp': unified_timestamp(video.get('publishDate')),
@@ -174,22 +174,22 @@ class AmericasTestKitchenSeasonIE(InfoExtractor):
]
if season_number:
playlist_id = 'season_%d' % season_number
playlist_title = 'Season %d' % season_number
playlist_id = f'season_{season_number}'
playlist_title = f'Season {season_number}'
facet_filters.append('search_season_list:' + playlist_title)
else:
playlist_id = show
playlist_title = title
season_search = self._download_json(
'https://y1fnzxui30-dsn.algolia.net/1/indexes/everest_search_%s_season_desc_production' % slug,
f'https://y1fnzxui30-dsn.algolia.net/1/indexes/everest_search_{slug}_season_desc_production',
playlist_id, headers={
'Origin': 'https://www.americastestkitchen.com',
'X-Algolia-API-Key': '8d504d0099ed27c1b73708d22871d805',
'X-Algolia-Application-Id': 'Y1FNZXUI30',
}, query={
'facetFilters': json.dumps(facet_filters),
'attributesToRetrieve': 'description,search_%s_episode_number,search_document_date,search_url,title,search_atk_episode_season' % slug,
'attributesToRetrieve': f'description,search_{slug}_episode_number,search_document_date,search_url,title,search_atk_episode_season',
'attributesToHighlight': '',
'hitsPerPage': 1000,
})
@@ -207,7 +207,7 @@ class AmericasTestKitchenSeasonIE(InfoExtractor):
'description': episode.get('description'),
'timestamp': unified_timestamp(episode.get('search_document_date')),
'season_number': season_number,
'episode_number': int_or_none(episode.get('search_%s_episode_number' % slug)),
'episode_number': int_or_none(episode.get(f'search_{slug}_episode_number')),
'ie_key': AmericasTestKitchenIE.ie_key(),
}
yt_dlp/extractor/amp.py
+9 -6
View File
@@ -1,10 +1,11 @@
from .common import InfoExtractor
from ..utils import (
determine_ext,
ExtractorError,
determine_ext,
int_or_none,
mimetype2ext,
parse_iso8601,
strip_jsonp,
unified_timestamp,
url_or_none,
)
@@ -15,15 +16,15 @@ class AMPIE(InfoExtractor): # XXX: Conventionally, base classes should end with
def _extract_feed_info(self, url):
feed = self._download_json(
url, None, 'Downloading Akamai AMP feed',
'Unable to download Akamai AMP feed')
'Unable to download Akamai AMP feed', transform_source=strip_jsonp)
item = feed.get('channel', {}).get('item')
if not item:
raise ExtractorError('%s said: %s' % (self.IE_NAME, feed['error']))
raise ExtractorError('{} said: {}'.format(self.IE_NAME, feed['error']))
video_id = item['guid']
def get_media_node(name, default=None):
media_name = 'media-%s' % name
media_name = f'media-{name}'
media_group = item.get('media-group') or item
return media_group.get(media_name) or item.get(media_name) or item.get(name, default)
@@ -73,8 +74,10 @@ class AMPIE(InfoExtractor): # XXX: Conventionally, base classes should end with
media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
video_id, f4m_id='hds', fatal=False))
elif ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
fmts, subs = self._extract_m3u8_formats_and_subtitles(
media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
formats.extend(fmts)
self._merge_subtitles(subs, target=subtitles)
else:
formats.append({
'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
yt_dlp/extractor/anchorfm.py
+4 -4
View File
@@ -5,7 +5,7 @@ from ..utils import (
int_or_none,
str_or_none,
traverse_obj,
unified_timestamp
unified_timestamp,
)
@@ -29,7 +29,7 @@ class AnchorFMEpisodeIE(InfoExtractor):
'release_date': '20230121',
'release_timestamp': 1674285179,
'episode_id': 'e1tpt3d',
}
},
}, {
# embed url
'url': 'https://anchor.fm/apakatatempo/embed/episodes/S2E75-Perang-Bintang-di-Balik-Kasus-Ferdy-Sambo-dan-Ismail-Bolong-e1shjqd',
@@ -50,7 +50,7 @@ class AnchorFMEpisodeIE(InfoExtractor):
'season': 'Season 2',
'season_number': 2,
'episode_id': 'e1shjqd',
}
},
}]
_WEBPAGE_TESTS = [{
@@ -72,7 +72,7 @@ class AnchorFMEpisodeIE(InfoExtractor):
'thumbnail': 'https://s3-us-west-2.amazonaws.com/anchor-generated-image-bank/production/podcast_uploaded_episode400/2627805/2627805-1671590688729-4db3882ac9e4b.jpg',
'uploader': 'Podcast Tempo',
'channel': 'apakatatempo',
}
},
}]
def _real_extract(self, url):
yt_dlp/extractor/angel.py
+6 -6
View File
@@ -1,7 +1,7 @@
import re
from .common import InfoExtractor
from ..utils import url_or_none, merge_dicts
from ..utils import merge_dicts, url_or_none
class AngelIE(InfoExtractor):
@@ -15,8 +15,8 @@ class AngelIE(InfoExtractor):
'title': 'Tuttle Twins Season 1, Episode 1: When Laws Give You Lemons',
'description': 'md5:73b704897c20ab59c433a9c0a8202d5e',
'thumbnail': r're:^https?://images.angelstudios.com/image/upload/angel-app/.*$',
'duration': 1359.0
}
'duration': 1359.0,
},
}, {
'url': 'https://www.angel.com/watch/the-chosen/episode/8dfb714d-bca5-4812-8125-24fb9514cd10/season-1/episode-1/i-have-called-you-by-name',
'md5': 'e4774bad0a5f0ad2e90d175cafdb797d',
@@ -26,8 +26,8 @@ class AngelIE(InfoExtractor):
'title': 'The Chosen Season 1, Episode 1: I Have Called You By Name',
'description': 'md5:aadfb4827a94415de5ff6426e6dee3be',
'thumbnail': r're:^https?://images.angelstudios.com/image/upload/angel-app/.*$',
'duration': 3276.0
}
'duration': 3276.0,
},
}]
def _real_extract(self, url):
@@ -44,7 +44,7 @@ class AngelIE(InfoExtractor):
'title': self._og_search_title(webpage),
'description': self._og_search_description(webpage),
'formats': formats,
'subtitles': subtitles
'subtitles': subtitles,
}
# Angel uses cloudinary in the background and supports image transformations.
yt_dlp/extractor/ant1newsgr.py → yt_dlp/extractor/antenna.py
+42 -27
View File
@@ -1,26 +1,30 @@
import urllib.parse
from .common import InfoExtractor
from ..networking import HEADRequest
from ..utils import (
HEADRequest,
ExtractorError,
determine_ext,
make_archive_id,
scale_thumbnails_to_max_format_width,
)
class Ant1NewsGrBaseIE(InfoExtractor):
class AntennaBaseIE(InfoExtractor):
def _download_and_extract_api_data(self, video_id, netloc, cid=None):
url = f'{self.http_scheme()}//{netloc}{self._API_PATH}'
info = self._download_json(url, video_id, query={'cid': cid or video_id})
try:
source = info['url']
except KeyError:
raise ExtractorError('no source found for %s' % video_id)
formats, subs = (self._extract_m3u8_formats_and_subtitles(source, video_id, 'mp4')
if determine_ext(source) == 'm3u8' else ([{'url': source}], {}))
info = self._download_json(f'{self.http_scheme()}//{netloc}{self._API_PATH}',
video_id, query={'cid': cid or video_id})
if not info.get('url'):
raise ExtractorError(f'No source found for {video_id}')
ext = determine_ext(info['url'])
if ext == 'm3u8':
formats, subs = self._extract_m3u8_formats_and_subtitles(info['url'], video_id, 'mp4')
else:
formats, subs = [{'url': info['url'], 'format_id': ext}], {}
thumbnails = scale_thumbnails_to_max_format_width(
formats, [{'url': info['thumb']}], r'(?<=/imgHandler/)\d+')
formats, [{'url': info['thumb']}], r'(?<=/imgHandler/)\d+') if info.get('thumb') else []
return {
'id': video_id,
'title': info.get('title'),
@@ -30,21 +34,31 @@ class Ant1NewsGrBaseIE(InfoExtractor):
}
class Ant1NewsGrWatchIE(Ant1NewsGrBaseIE):
IE_NAME = 'ant1newsgr:watch'
IE_DESC = 'ant1news.gr videos'
_VALID_URL = r'https?://(?P<netloc>(?:www\.)?ant1news\.gr)/watch/(?P<id>\d+)/'
class AntennaGrWatchIE(AntennaBaseIE):
IE_NAME = 'antenna:watch'
IE_DESC = 'antenna.gr and ant1news.gr videos'
_VALID_URL = r'https?://(?P<netloc>(?:www\.)?(?:antenna|ant1news)\.gr)/watch/(?P<id>\d+)/'
_API_PATH = '/templates/data/player'
_TESTS = [{
'url': 'https://www.ant1news.gr/watch/1506168/ant1-news-09112021-stis-18-45',
'md5': '95925e6b32106754235f2417e0d2dfab',
'md5': 'c472d9dd7cd233c63aff2ea42201cda6',
'info_dict': {
'id': '1506168',
'ext': 'mp4',
'title': 'md5:0ad00fa66ecf8aa233d26ab0dba7514a',
'description': 'md5:18665af715a6dcfeac1d6153a44f16b0',
'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/26d46bf6-8158-4f02-b197-7096c714b2de.jpg',
'thumbnail': r're:https://ant1media\.azureedge\.net/imgHandler/\d+/26d46bf6-8158-4f02-b197-7096c714b2de\.jpg',
},
}, {
'url': 'https://www.antenna.gr/watch/1643812/oi-prodotes-epeisodio-01',
'md5': '8f6f7dd3b1dba4d835ba990e25f31243',
'info_dict': {
'id': '1643812',
'ext': 'mp4',
'format_id': 'mp4',
'title': 'ΟΙ ΠΡΟΔΟΤΕΣ – ΕΠΕΙΣΟΔΙΟ 01',
'thumbnail': r're:https://ant1media\.azureedge\.net/imgHandler/\d+/b3d63096-e72d-43c4-87a0-00d4363d242f\.jpg',
},
}]
@@ -52,25 +66,26 @@ class Ant1NewsGrWatchIE(Ant1NewsGrBaseIE):
video_id, netloc = self._match_valid_url(url).group('id', 'netloc')
webpage = self._download_webpage(url, video_id)
info = self._download_and_extract_api_data(video_id, netloc)
info['description'] = self._og_search_description(webpage)
info['description'] = self._og_search_description(webpage, default=None)
info['_old_archive_ids'] = [make_archive_id('Ant1NewsGrWatch', video_id)]
return info
class Ant1NewsGrArticleIE(Ant1NewsGrBaseIE):
class Ant1NewsGrArticleIE(AntennaBaseIE):
IE_NAME = 'ant1newsgr:article'
IE_DESC = 'ant1news.gr articles'
_VALID_URL = r'https?://(?:www\.)?ant1news\.gr/[^/]+/article/(?P<id>\d+)/'
_TESTS = [{
'url': 'https://www.ant1news.gr/afieromata/article/549468/o-tzeims-mpont-sta-meteora-oi-apeiles-kai-o-xesikomos-ton-kalogeron',
'md5': '294f18331bb516539d72d85a82887dcc',
'md5': '57eb8d12181f0fa2b14b0b138e1de9b6',
'info_dict': {
'id': '_xvg/m_cmbatw=',
'ext': 'mp4',
'title': 'md5:a93e8ecf2e4073bfdffcb38f59945411',
'timestamp': 1603092840,
'upload_date': '20201019',
'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/756206d2-d640-40e2-b201-3555abdfc0db.jpg',
'timestamp': 1666166520,
'upload_date': '20221019',
'thumbnail': 'https://ant1media.azureedge.net/imgHandler/1920/756206d2-d640-40e2-b201-3555abdfc0db.jpg',
},
}, {
'url': 'https://ant1news.gr/Society/article/620286/symmoria-anilikon-dikigoros-thymaton-ithelan-na-toys-apoteleiosoyn',
@@ -90,19 +105,19 @@ class Ant1NewsGrArticleIE(Ant1NewsGrBaseIE):
info = self._search_json_ld(webpage, video_id, expected_type='NewsArticle')
embed_urls = list(Ant1NewsGrEmbedIE._extract_embed_urls(url, webpage))
if not embed_urls:
raise ExtractorError('no videos found for %s' % video_id, expected=True)
raise ExtractorError(f'no videos found for {video_id}', expected=True)
return self.playlist_from_matches(
embed_urls, video_id, info.get('title'), ie=Ant1NewsGrEmbedIE.ie_key(),
video_kwargs={'url_transparent': True, 'timestamp': info.get('timestamp')})
class Ant1NewsGrEmbedIE(Ant1NewsGrBaseIE):
class Ant1NewsGrEmbedIE(AntennaBaseIE):
IE_NAME = 'ant1newsgr:embed'
IE_DESC = 'ant1news.gr embedded videos'
_BASE_PLAYER_URL_RE = r'(?:https?:)?//(?:[a-zA-Z0-9\-]+\.)?(?:antenna|ant1news)\.gr/templates/pages/player'
_VALID_URL = rf'{_BASE_PLAYER_URL_RE}\?([^#]+&)?cid=(?P<id>[^#&]+)'
_EMBED_REGEX = [rf'<iframe[^>]+?src=(?P<_q1>["\'])(?P<url>{_BASE_PLAYER_URL_RE}\?(?:(?!(?P=_q1)).)+)(?P=_q1)']
_API_PATH = '/news/templates/data/jsonPlayer'
_API_PATH = '/templates/data/jsonPlayer'
_TESTS = [{
'url': 'https://www.antenna.gr/templates/pages/player?cid=3f_li_c_az_jw_y_u=&w=670&h=377',
@@ -121,7 +136,7 @@ class Ant1NewsGrEmbedIE(Ant1NewsGrBaseIE):
canonical_url = self._request_webpage(
HEADRequest(url), video_id,
note='Resolve canonical player URL',
errnote='Could not resolve canonical player URL').geturl()
errnote='Could not resolve canonical player URL').url
_, netloc, _, _, query, _ = urllib.parse.urlparse(canonical_url)
cid = urllib.parse.parse_qs(query)['cid'][0]
yt_dlp/extractor/anvato.py
+5 -13
View File
@@ -238,7 +238,7 @@ class AnvatoIE(InfoExtractor):
'gray': 'anvato_mcp_gray_web_prod_4c10f067c393ed8fc453d3930f8ab2b159973900',
'hearst': 'anvato_mcp_hearst_web_prod_5356c3de0fc7c90a3727b4863ca7fec3a4524a99',
'cbs': 'anvato_mcp_cbs_web_prod_02f26581ff80e5bda7aad28226a8d369037f2cbe',
'telemundo': 'anvato_mcp_telemundo_web_prod_c5278d51ad46fda4b6ca3d0ea44a7846a054f582'
'telemundo': 'anvato_mcp_telemundo_web_prod_c5278d51ad46fda4b6ca3d0ea44a7846a054f582',
}
def _generate_nfl_token(self, anvack, mcp_id):
@@ -255,7 +255,7 @@ class AnvatoIE(InfoExtractor):
token
}
}
}''' % (anvack, mcp_id),
}''' % (anvack, mcp_id), # noqa: UP031
}).encode(), headers={
'Authorization': auth_token,
'Content-Type': 'application/json',
@@ -299,7 +299,7 @@ class AnvatoIE(InfoExtractor):
return self._download_json(
video_data_url, video_id, transform_source=strip_jsonp, query=query,
data=json.dumps({'api': api}, separators=(',', ':')).encode('utf-8'))
data=json.dumps({'api': api}, separators=(',', ':')).encode())
def _get_anvato_videos(self, access_key, video_id, token):
video_data = self._get_video_json(access_key, video_id, token)
@@ -336,7 +336,7 @@ class AnvatoIE(InfoExtractor):
elif media_format == 'm3u8-variant' or ext == 'm3u8':
# For some videos the initial m3u8 URL returns JSON instead
manifest_json = self._download_json(
video_url, video_id, note='Downloading manifest JSON', errnote=False)
video_url, video_id, note='Downloading manifest JSON', fatal=False)
if manifest_json:
video_url = manifest_json.get('master_m3u8')
if not video_url:
@@ -358,7 +358,7 @@ class AnvatoIE(InfoExtractor):
for caption in video_data.get('captions', []):
a_caption = {
'url': caption['url'],
'ext': 'tt' if caption.get('format') == 'SMPTE-TT' else None
'ext': 'tt' if caption.get('format') == 'SMPTE-TT' else None,
}
subtitles.setdefault(caption['language'], []).append(a_caption)
subtitles = self._merge_subtitles(subtitles, hls_subs, vtt_subs)
@@ -392,14 +392,6 @@ class AnvatoIE(InfoExtractor):
url = smuggle_url(url, {'token': anvplayer_data['token']})
yield cls.url_result(url, AnvatoIE, video_id)
def _extract_anvato_videos(self, webpage, video_id):
anvplayer_data = self._parse_json(
self._html_search_regex(
self._ANVP_RE, webpage, 'Anvato player data', group='anvp'),
video_id)
return self._get_anvato_videos(
anvplayer_data['accessKey'], anvplayer_data['video'], 'default') # cbslocal token = 'default'
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
self._initialize_geo_bypass({
yt_dlp/extractor/aol.py
+5 -4
View File
@@ -10,6 +10,7 @@ from ..utils import (
class AolIE(YahooIE): # XXX: Do not subclass from concrete IE
_WORKING = False
IE_NAME = 'aol.com'
_VALID_URL = r'(?:aol-video:|https?://(?:www\.)?aol\.(?:com|ca|co\.uk|de|jp)/video/(?:[^/]+/)*)(?P<id>\d{9}|[0-9a-f]{24}|[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12})'
@@ -29,7 +30,7 @@ class AolIE(YahooIE): # XXX: Do not subclass from concrete IE
'params': {
# m3u8 download
'skip_download': True,
}
},
}, {
# video with vidible ID
'url': 'https://www.aol.com/video/view/netflix-is-raising-rates/5707d6b8e4b090497b04f706/',
@@ -45,7 +46,7 @@ class AolIE(YahooIE): # XXX: Do not subclass from concrete IE
'params': {
# m3u8 download
'skip_download': True,
}
},
}, {
'url': 'https://www.aol.com/video/view/park-bench-season-2-trailer/559a1b9be4b0c3bfad3357a7/',
'only_matching': True,
@@ -82,10 +83,10 @@ class AolIE(YahooIE): # XXX: Do not subclass from concrete IE
return self._extract_yahoo_video(video_id, 'us')
response = self._download_json(
'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id,
f'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/{video_id}/details',
video_id)['response']
if response['statusText'] != 'Ok':
raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusText']), expected=True)
raise ExtractorError('{} said: {}'.format(self.IE_NAME, response['statusText']), expected=True)
video_data = response['data']
formats = []
yt_dlp/extractor/apa.py
+2 -2
View File
@@ -34,7 +34,7 @@ class APAIE(InfoExtractor):
video_id, base_url = mobj.group('id', 'base_url')
webpage = self._download_webpage(
'%s/player/%s' % (base_url, video_id), video_id)
f'{base_url}/player/{video_id}', video_id)
jwplatform_id = self._search_regex(
r'media[iI]d\s*:\s*["\'](?P<id>[a-zA-Z0-9]{8})', webpage,
@@ -47,7 +47,7 @@ class APAIE(InfoExtractor):
def extract(field, name=None):
return self._search_regex(
r'\b%s["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1' % field,
rf'\b{field}["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
webpage, name or field, default=None, group='value')
title = extract('title') or video_id
yt_dlp/extractor/appleconnect.py
+1 -4
View File
@@ -1,8 +1,5 @@
from .common import InfoExtractor
from ..utils import (
str_to_int,
ExtractorError
)
from ..utils import ExtractorError, str_to_int
class AppleConnectIE(InfoExtractor):
yt_dlp/extractor/applepodcasts.py
+1 -1
View File
@@ -24,7 +24,7 @@ class ApplePodcastsIE(InfoExtractor):
'duration': 6454,
'series': 'The Tim Dillon Show',
'thumbnail': 're:.+[.](png|jpe?g|webp)',
}
},
}, {
'url': 'https://podcasts.apple.com/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777',
'only_matching': True,
yt_dlp/extractor/appletrailers.py
+16 -17
View File
@@ -1,8 +1,8 @@
import re
import json
import re
import urllib.parse
from .common import InfoExtractor
from ..compat import compat_urlparse
from ..utils import (
int_or_none,
parse_duration,
@@ -64,7 +64,7 @@ class AppleTrailersIE(InfoExtractor):
'uploader_id': 'wb',
},
},
]
],
}, {
'url': 'http://trailers.apple.com/trailers/magnolia/blackthorn/',
'info_dict': {
@@ -99,7 +99,7 @@ class AppleTrailersIE(InfoExtractor):
webpage = self._download_webpage(url, movie)
film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
film_data = self._download_json(
'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
f'http://trailers.apple.com/trailers/feeds/data/{film_id}.json',
film_id, fatal=False)
if film_data:
@@ -114,7 +114,7 @@ class AppleTrailersIE(InfoExtractor):
if not src:
continue
formats.append({
'format_id': '%s-%s' % (version, size),
'format_id': f'{version}-{size}',
'url': re.sub(r'_(\d+p\.mov)', r'_h\1', src),
'width': int_or_none(size_data.get('width')),
'height': int_or_none(size_data.get('height')),
@@ -134,7 +134,7 @@ class AppleTrailersIE(InfoExtractor):
page_data = film_data.get('page', {})
return self.playlist_result(entries, film_id, page_data.get('movie_title'))
playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
playlist_url = urllib.parse.urljoin(url, 'includes/playlists/itunes.inc')
def fix_html(s):
s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
@@ -143,10 +143,9 @@ class AppleTrailersIE(InfoExtractor):
# like: http://trailers.apple.com/trailers/wb/gravity/
def _clean_json(m):
return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
return 'iTunes.playURL({});'.format(m.group(1).replace('\'', '&#39;'))
s = re.sub(self._JSON_RE, _clean_json, s)
s = '<html>%s</html>' % s
return s
return f'<html>{s}</html>'
doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
playlist = []
@@ -170,18 +169,18 @@ class AppleTrailersIE(InfoExtractor):
duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
settings_json_url = urllib.parse.urljoin(url, f'includes/settings/{trailer_id}.json')
settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
formats = []
for format in settings['metadata']['sizes']:
for fmt in settings['metadata']['sizes']:
# The src is a file pointing to the real video file
format_url = re.sub(r'_(\d*p\.mov)', r'_h\1', format['src'])
format_url = re.sub(r'_(\d*p\.mov)', r'_h\1', fmt['src'])
formats.append({
'url': format_url,
'format': format['type'],
'width': int_or_none(format['width']),
'height': int_or_none(format['height']),
'format': fmt['type'],
'width': int_or_none(fmt['width']),
'height': int_or_none(fmt['height']),
})
playlist.append({
@@ -229,7 +228,7 @@ class AppleTrailersSectionIE(InfoExtractor):
'title': 'Movie Studios',
},
}
_VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/#section=(?P<id>%s)' % '|'.join(_SECTIONS)
_VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/#section=(?P<id>{})'.format('|'.join(_SECTIONS))
_TESTS = [{
'url': 'http://trailers.apple.com/#section=justadded',
'info_dict': {
@@ -270,7 +269,7 @@ class AppleTrailersSectionIE(InfoExtractor):
def _real_extract(self, url):
section = self._match_id(url)
section_data = self._download_json(
'http://trailers.apple.com/trailers/home/feeds/%s.json' % self._SECTIONS[section]['feed_path'],
'http://trailers.apple.com/trailers/home/feeds/{}.json'.format(self._SECTIONS[section]['feed_path']),
section)
entries = [
self.url_result('http://trailers.apple.com' + e['location'])
yt_dlp/extractor/archiveorg.py
+65 -298
View File
@@ -1,16 +1,16 @@
from __future__ import annotations
import json
import re
import urllib.error
import urllib.parse
from .common import InfoExtractor
from .naver import NaverBaseIE
from .youtube import YoutubeBaseInfoExtractor, YoutubeIE
from ..compat import compat_HTTPError, compat_urllib_parse_unquote
from ..networking import HEADRequest
from ..networking.exceptions import HTTPError
from ..utils import (
KNOWN_EXTENSIONS,
ExtractorError,
HEADRequest,
bug_reports_message,
clean_html,
dict_get,
@@ -32,6 +32,7 @@ from ..utils import (
unified_timestamp,
url_or_none,
urlhandle_detect_ext,
variadic,
)
@@ -50,10 +51,9 @@ class ArchiveOrgIE(InfoExtractor):
'release_date': '19681210',
'timestamp': 1268695290,
'upload_date': '20100315',
'creator': 'SRI International',
'creators': ['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',
@@ -111,7 +111,7 @@ class ArchiveOrgIE(InfoExtractor):
'title': 'Turning',
'ext': 'flac',
'track': 'Turning',
'creator': 'Grateful Dead',
'creators': ['Grateful Dead'],
'display_id': 'gd1977-05-08d01t01.flac',
'track_number': 1,
'album': '1977-05-08 - Barton Hall - Cornell University',
@@ -131,11 +131,10 @@ class ArchiveOrgIE(InfoExtractor):
'location': 'Barton Hall - Cornell University',
'duration': 438.68,
'track': 'Deal',
'creator': 'Grateful Dead',
'creators': ['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,
},
}, {
@@ -147,7 +146,7 @@ class ArchiveOrgIE(InfoExtractor):
'title': 'Bells Of Rostov',
'ext': 'mp3',
},
'skip': 'restricted'
'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',
@@ -160,7 +159,7 @@ class ArchiveOrgIE(InfoExtractor):
'description': 'md5:012b2d668ae753be36896f343d12a236',
'upload_date': '20190928',
},
'skip': 'restricted'
'skip': 'restricted',
}, {
# Original formats are private
'url': 'https://archive.org/details/irelandthemakingofarepublic',
@@ -170,7 +169,7 @@ class ArchiveOrgIE(InfoExtractor):
'upload_date': '20160610',
'description': 'md5:f70956a156645a658a0dc9513d9e78b7',
'uploader': 'dimitrios@archive.org',
'creator': ['British Broadcasting Corporation', 'Time-Life Films'],
'creators': ['British Broadcasting Corporation', 'Time-Life Films'],
'timestamp': 1465594947,
},
'playlist': [
@@ -204,8 +203,8 @@ class ArchiveOrgIE(InfoExtractor):
'thumbnail': 'https://archive.org/download/irelandthemakingofarepublic/irelandthemakingofarepublic.thumbs/irelandthemakingofarepublicreel2_001554.jpg',
'display_id': 'irelandthemakingofarepublicreel2.mov',
},
}
]
},
],
}]
@staticmethod
@@ -222,7 +221,7 @@ class ArchiveOrgIE(InfoExtractor):
def _real_extract(self, url):
video_id = urllib.parse.unquote_plus(self._match_id(url))
identifier, entry_id = (video_id.split('/', 1) + [None])[:2]
identifier, _, entry_id = video_id.partition('/')
# Archive.org metadata API doesn't clearly demarcate playlist entries
# or subtitle tracks, so we get them from the embeddable player.
@@ -248,7 +247,7 @@ class ArchiveOrgIE(InfoExtractor):
if track['kind'] != 'subtitles':
continue
entries[p['orig']][track['label']] = {
'url': 'https://archive.org/' + track['file'].lstrip('/')
'url': 'https://archive.org/' + track['file'].lstrip('/'),
}
metadata = self._download_json('http://archive.org/metadata/' + identifier, identifier)
@@ -260,7 +259,7 @@ class ArchiveOrgIE(InfoExtractor):
'title': m['title'],
'description': clean_html(m.get('description')),
'uploader': dict_get(m, ['uploader', 'adder']),
'creator': m.get('creator'),
'creators': traverse_obj(m, ('creator', {variadic}, {lambda x: x[0] and list(x)})),
'license': m.get('licenseurl'),
'release_date': unified_strdate(m.get('date')),
'timestamp': unified_timestamp(dict_get(m, ['publicdate', 'addeddate'])),
@@ -275,7 +274,7 @@ class ArchiveOrgIE(InfoExtractor):
'title': f.get('title') or f['name'],
'display_id': f['name'],
'description': clean_html(f.get('description')),
'creator': f.get('creator'),
'creators': traverse_obj(f, ('creator', {variadic}, {lambda x: x[0] and list(x)})),
'duration': parse_duration(f.get('length')),
'track_number': int_or_none(f.get('track')),
'album': f.get('album'),
@@ -295,7 +294,9 @@ class ArchiveOrgIE(InfoExtractor):
'height': int_or_none(f.get('width')),
'filesize': int_or_none(f.get('size'))})
extension = (f['name'].rsplit('.', 1) + [None])[1]
_, has_ext, extension = f['name'].rpartition('.')
if not has_ext:
extension = None
# 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.
@@ -303,14 +304,14 @@ class ArchiveOrgIE(InfoExtractor):
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'],
'url': 'https://archive.org/download/' + identifier + '/' + urllib.parse.quote(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',
'source_preference': 0 if f.get('source') == 'original' else -1,
'format_note': f.get('source')
'format_note': f.get('source'),
})
for entry in entries.values():
@@ -373,7 +374,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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
'url': 'https://web.archive.org/web/2oe/http://wayback-fakeurl.archive.org/yt/97t7Xj_iBv0',
@@ -390,7 +391,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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.
# Should use the date in the link. Title ends with '- Youtube'. Capture has description in eow-description
@@ -405,8 +406,8 @@ class YoutubeWebArchiveIE(InfoExtractor):
'uploader_id': 'machinima',
'uploader_url': 'https://www.youtube.com/user/machinima',
'thumbnail': r're:https?://.*\.(jpg|webp)',
'uploader': 'machinima'
}
'uploader': 'machinima',
},
}, {
# FLV video. Video file URL does not provide itag information
'url': 'https://web.archive.org/web/20081211103536/http://www.youtube.com/watch?v=jNQXAC9IVRw',
@@ -423,7 +424,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
'info_dict': {
@@ -439,7 +440,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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.
'url': 'https://web.archive.org/https://www.youtube.com/watch?v=1JYutPM8O6E',
@@ -456,7 +457,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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.
'url': 'https://web.archive.org/web/20180803221945/https://www.youtube.com/watch?v=6FPhZJGvf4E',
@@ -475,15 +476,15 @@ class YoutubeWebArchiveIE(InfoExtractor):
'uploader': 'ETC News',
},
'expected_warnings': [
r'unable to download capture webpage \(it may not be archived\)'
]
r'unable to download capture webpage \(it may not be archived\)',
],
}, { # Very old YouTube page, has - YouTube in title.
'url': 'http://web.archive.org/web/20070302011044/http://youtube.com/watch?v=-06-KB9XTzg',
'info_dict': {
'id': '-06-KB9XTzg',
'ext': 'flv',
'title': 'New Coin Hack!! 100% Safe!!'
}
'title': 'New Coin Hack!! 100% Safe!!',
},
}, {
'url': 'web.archive.org/https://www.youtube.com/watch?v=dWW7qP423y8',
'info_dict': {
@@ -497,7 +498,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'description': 'md5:7b567f898d8237b256f36c1a07d6d7bc',
'thumbnail': r're:https?://.*\.(jpg|webp)',
'uploader': 'DankPods',
}
},
}, {
# player response contains '};' See: https://github.com/ytdl-org/youtube-dl/issues/27093
'url': 'https://web.archive.org/web/20200827003909if_/http://www.youtube.com/watch?v=6Dh-RL__uN4',
@@ -514,7 +515,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'uploader_id': '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',
@@ -529,7 +530,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -544,7 +545,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -559,7 +560,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -573,7 +574,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -589,7 +590,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -606,7 +607,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -621,7 +622,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
'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',
@@ -636,35 +637,35 @@ class YoutubeWebArchiveIE(InfoExtractor):
'duration': 200,
'upload_date': '20120407',
'uploader_id': 'thecomputernerd01',
}
},
}, {
'url': 'https://web.archive.org/web/http://www.youtube.com/watch?v=kH-G_aIBlFw',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://web.archive.org/web/20050214000000_if/http://www.youtube.com/watch?v=0altSZ96U4M',
'only_matching': True
'only_matching': True,
}, {
# Video not archived, only capture is unavailable video page
'url': 'https://web.archive.org/web/20210530071008/https://www.youtube.com/watch?v=lHJTf93HL1s&spfreload=10',
'only_matching': True
'only_matching': True,
}, { # Encoded url
'url': 'https://web.archive.org/web/20120712231619/http%3A//www.youtube.com/watch%3Fgl%3DUS%26v%3DAkhihxRKcrs%26hl%3Den',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://web.archive.org/web/20120712231619/http%3A//www.youtube.com/watch%3Fv%3DAkhihxRKcrs%26gl%3DUS%26hl%3Den',
'only_matching': True
'only_matching': True,
}, {
'url': 'https://web.archive.org/web/20060527081937/http://www.youtube.com:80/watch.php?v=ELTFsLT73fA&amp;search=soccer',
'only_matching': True
'only_matching': True,
}, {
'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
'only_matching': True,
}, {
'url': 'ytarchive:BaW_jenozKc',
'only_matching': True
'only_matching': True,
},
]
_YT_INITIAL_DATA_RE = YoutubeBaseInfoExtractor._YT_INITIAL_DATA_RE
@@ -675,13 +676,13 @@ class YoutubeWebArchiveIE(InfoExtractor):
_YT_DEFAULT_THUMB_SERVERS = ['i.ytimg.com'] # thumbnails most likely archived on these servers
_YT_ALL_THUMB_SERVERS = orderedSet(
_YT_DEFAULT_THUMB_SERVERS + ['img.youtube.com', *[f'{c}{n or ""}.ytimg.com' for c in ('i', 's') for n in (*range(0, 5), 9)]])
[*_YT_DEFAULT_THUMB_SERVERS, 'img.youtube.com', *[f'{c}{n or ""}.ytimg.com' for c in ('i', 's') for n in (*range(5), 9)]])
_WAYBACK_BASE_URL = 'https://web.archive.org/web/%sif_/'
_OLDEST_CAPTURE_DATE = 20050214000000
_NEWEST_CAPTURE_DATE = 20500101000000
def _call_cdx_api(self, item_id, url, filters: list = None, collapse: list = None, query: dict = None, note=None, fatal=False):
def _call_cdx_api(self, item_id, url, filters: list | None = None, collapse: list | None = None, query: dict | None = None, note=None, fatal=False):
# CDX docs: https://github.com/internetarchive/wayback/blob/master/wayback-cdx-server/README.md
query = {
'url': url,
@@ -690,14 +691,14 @@ class YoutubeWebArchiveIE(InfoExtractor):
'limit': 500,
'filter': ['statuscode:200'] + (filters or []),
'collapse': collapse or [],
**(query or {})
**(query or {}),
}
res = self._download_json(
'https://web.archive.org/cdx/search/cdx', item_id,
note or 'Downloading CDX API JSON', query=query, fatal=fatal)
if isinstance(res, list) and len(res) >= 2:
# format response to make it easier to use
return list(dict(zip(res[0], v)) for v in res[1:])
return [dict(zip(res[0], v)) for v in res[1:]]
elif not isinstance(res, list) or len(res) != 0:
self.report_warning('Error while parsing CDX API response' + bug_reports_message())
@@ -854,7 +855,7 @@ class YoutubeWebArchiveIE(InfoExtractor):
{
'url': (self._WAYBACK_BASE_URL % (int_or_none(thumbnail_dict.get('timestamp')) or self._OLDEST_CAPTURE_DATE)) + thumbnail_dict.get('original'),
'filesize': int_or_none(thumbnail_dict.get('length')),
'preference': int_or_none(thumbnail_dict.get('length'))
'preference': int_or_none(thumbnail_dict.get('length')),
} for thumbnail_dict in response)
if not try_all:
break
@@ -895,11 +896,11 @@ class YoutubeWebArchiveIE(InfoExtractor):
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),
HEADRequest(f'https://web.archive.org/web/2oe_/http://wayback-fakeurl.archive.org/yt/{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:
if isinstance(e.cause, HTTPError) and e.cause.status == 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:
@@ -926,258 +927,24 @@ class YoutubeWebArchiveIE(InfoExtractor):
info['thumbnails'] = self._extract_thumbnails(video_id)
if urlh:
url = compat_urllib_parse_unquote(urlh.geturl())
url = urllib.parse.unquote(urlh.url)
video_file_url_qs = parse_qs(url)
# 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'))}
fmt = {'url': url, 'filesize': int_or_none(urlh.headers.get('x-archive-orig-content-length'))}
itag = try_get(video_file_url_qs, lambda x: x['itag'][0])
if itag and itag in YoutubeIE._formats:
format.update(YoutubeIE._formats[itag])
format.update({'format_id': itag})
fmt.update(YoutubeIE._formats[itag])
fmt.update({'format_id': itag})
else:
mime = try_get(video_file_url_qs, lambda x: x['mime'][0])
ext = (mimetype2ext(mime)
or urlhandle_detect_ext(urlh)
or mimetype2ext(urlh.headers.get('x-archive-guessed-content-type')))
format.update({'ext': ext})
info['formats'] = [format]
fmt.update({'ext': ext})
info['formats'] = [fmt]
if not info.get('duration'):
info['duration'] = str_to_int(try_get(video_file_url_qs, lambda x: x['dur'][0]))
if not info.get('title'):
info['title'] = video_id
return info
class VLiveWebArchiveIE(InfoExtractor):
IE_NAME = 'web.archive:vlive'
IE_DESC = 'web.archive.org saved vlive videos'
_VALID_URL = r'''(?x)
(?:https?://)?web\.archive\.org/
(?:web/)?(?:(?P<date>[0-9]{14})?[0-9A-Za-z_*]*/)? # /web and the version index is optional
(?:https?(?::|%3[Aa])//)?(?:
(?:(?:www|m)\.)?vlive\.tv(?::(?:80|443))?/(?:video|embed)/(?P<id>[0-9]+) # VLive URL
)
'''
_TESTS = [{
'url': 'https://web.archive.org/web/20221221144331/http://www.vlive.tv/video/1326',
'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': {
'id': '1326',
'ext': 'mp4',
'title': "Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
'uploader_id': 'muploader_a',
'uploader_url': None,
'uploader': None,
'upload_date': '20150817',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1439816449,
'like_count': int,
'channel': 'Girl\'s Day',
'channel_id': 'FDF27',
'comment_count': int,
'release_timestamp': 1439818140,
'release_date': '20150817',
'duration': 1014,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://web.archive.org/web/20221221182103/http://www.vlive.tv/video/16937',
'info_dict': {
'id': '16937',
'ext': 'mp4',
'title': '첸백시 걍방',
'creator': 'EXO',
'view_count': int,
'subtitles': 'mincount:12',
'uploader_id': 'muploader_j',
'uploader_url': 'http://vlive.tv',
'uploader': None,
'upload_date': '20161112',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1478923074,
'like_count': int,
'channel': 'EXO',
'channel_id': 'F94BD',
'comment_count': int,
'release_timestamp': 1478924280,
'release_date': '20161112',
'duration': 906,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://web.archive.org/web/20221127190050/http://www.vlive.tv/video/101870',
'info_dict': {
'id': '101870',
'ext': 'mp4',
'title': '[ⓓ xV] “레벨이들 매력에 반해? 안 반해?” 움직이는 HD 포토 (레드벨벳:Red Velvet)',
'creator': 'Dispatch',
'view_count': int,
'subtitles': 'mincount:6',
'uploader_id': 'V__FRA08071',
'uploader_url': 'http://vlive.tv',
'uploader': None,
'upload_date': '20181130',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1543601327,
'like_count': int,
'channel': 'Dispatch',
'channel_id': 'C796F3',
'comment_count': int,
'release_timestamp': 1543601040,
'release_date': '20181130',
'duration': 279,
},
'params': {
'skip_download': True,
},
}]
# The wayback machine has special timestamp and "mode" values:
# timestamp:
# 1 = the first capture
# 2 = the last capture
# mode:
# id_ = Identity - perform no alterations of the original resource, return it as it was archived.
_WAYBACK_BASE_URL = 'https://web.archive.org/web/2id_/'
def _download_archived_page(self, url, video_id, *, timestamp='2', **kwargs):
for retry in self.RetryManager():
try:
return self._download_webpage(f'https://web.archive.org/web/{timestamp}id_/{url}', video_id, **kwargs)
except ExtractorError as e:
if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 404:
raise ExtractorError('Page was not archived', expected=True)
retry.error = e
continue
def _download_archived_json(self, url, video_id, **kwargs):
page = self._download_archived_page(url, video_id, **kwargs)
if not page:
raise ExtractorError('Page was not archived', expected=True)
else:
return self._parse_json(page, video_id)
def _extract_formats_from_m3u8(self, m3u8_url, params, video_id):
m3u8_doc = self._download_archived_page(m3u8_url, video_id, note='Downloading m3u8', query=params, fatal=False)
if not m3u8_doc:
return
# M3U8 document should be changed to archive domain
m3u8_doc = m3u8_doc.splitlines()
url_base = m3u8_url.rsplit('/', 1)[0]
first_segment = None
for i, line in enumerate(m3u8_doc):
if not line.startswith('#'):
m3u8_doc[i] = f'{self._WAYBACK_BASE_URL}{url_base}/{line}?{urllib.parse.urlencode(params)}'
first_segment = first_segment or m3u8_doc[i]
# Segments may not have been archived. See https://web.archive.org/web/20221127190050/http://www.vlive.tv/video/101870
urlh = self._request_webpage(HEADRequest(first_segment), video_id, errnote=False,
fatal=False, note='Check first segment availablity')
if urlh:
formats, subtitles = self._parse_m3u8_formats_and_subtitles('\n'.join(m3u8_doc), ext='mp4', video_id=video_id)
if subtitles:
self._report_ignoring_subs('m3u8')
return formats
# Closely follows the logic of the ArchiveTeam grab script
# See: https://github.com/ArchiveTeam/vlive-grab/blob/master/vlive.lua
def _real_extract(self, url):
video_id, url_date = self._match_valid_url(url).group('id', 'date')
webpage = self._download_archived_page(f'https://www.vlive.tv/video/{video_id}', video_id, timestamp=url_date)
player_info = self._search_json(r'__PRELOADED_STATE__\s*=', webpage, 'player info', video_id)
user_country = traverse_obj(player_info, ('common', 'userCountry'))
main_script_url = self._search_regex(r'<script\s+src="([^"]+/js/main\.[^"]+\.js)"', webpage, 'main script url')
main_script = self._download_archived_page(main_script_url, video_id, note='Downloading main script')
app_id = self._search_regex(r'appId\s*=\s*"([^"]+)"', main_script, 'app id')
inkey = self._download_archived_json(
f'https://www.vlive.tv/globalv-web/vam-web/video/v1.0/vod/{video_id}/inkey', video_id, note='Fetching inkey', query={
'appId': app_id,
'platformType': 'PC',
'gcc': user_country,
'locale': 'en_US',
}, fatal=False)
vod_id = traverse_obj(player_info, ('postDetail', 'post', 'officialVideo', 'vodId'))
vod_data = self._download_archived_json(
f'https://apis.naver.com/rmcnmv/rmcnmv/vod/play/v2.0/{vod_id}', video_id, note='Fetching vod data', query={
'key': inkey.get('inkey'),
'pid': 'rmcPlayer_16692457559726800', # partially unix time and partially random. Fixed value used by archiveteam project
'sid': '2024',
'ver': '2.0',
'devt': 'html5_pc',
'doct': 'json',
'ptc': 'https',
'sptc': 'https',
'cpt': 'vtt',
'ctls': '%7B%22visible%22%3A%7B%22fullscreen%22%3Atrue%2C%22logo%22%3Afalse%2C%22playbackRate%22%3Afalse%2C%22scrap%22%3Afalse%2C%22playCount%22%3Atrue%2C%22commentCount%22%3Atrue%2C%22title%22%3Atrue%2C%22writer%22%3Atrue%2C%22expand%22%3Afalse%2C%22subtitles%22%3Atrue%2C%22thumbnails%22%3Atrue%2C%22quality%22%3Atrue%2C%22setting%22%3Atrue%2C%22script%22%3Afalse%2C%22logoDimmed%22%3Atrue%2C%22badge%22%3Atrue%2C%22seekingTime%22%3Atrue%2C%22muted%22%3Atrue%2C%22muteButton%22%3Afalse%2C%22viewerNotice%22%3Afalse%2C%22linkCount%22%3Afalse%2C%22createTime%22%3Afalse%2C%22thumbnail%22%3Atrue%7D%2C%22clicked%22%3A%7B%22expand%22%3Afalse%2C%22subtitles%22%3Afalse%7D%7D',
'pv': '4.26.9',
'dr': '1920x1080',
'cpl': 'en_US',
'lc': 'en_US',
'adi': '%5B%7B%22type%22%3A%22pre%22%2C%22exposure%22%3Afalse%2C%22replayExposure%22%3Afalse%7D%5D',
'adu': '%2F',
'videoId': vod_id,
'cc': user_country,
})
formats = []
streams = traverse_obj(vod_data, ('streams', ...))
if len(streams) > 1:
self.report_warning('Multiple streams found. Only the first stream will be downloaded.')
stream = streams[0]
max_stream = max(
stream.get('videos') or [],
key=lambda v: traverse_obj(v, ('bitrate', 'video'), default=0), default=None)
if max_stream is not None:
params = {arg.get('name'): arg.get('value') for arg in stream.get('keys', []) if arg.get('type') == 'param'}
formats = self._extract_formats_from_m3u8(max_stream.get('source'), params, video_id) or []
# For parts of the project MP4 files were archived
max_video = max(
traverse_obj(vod_data, ('videos', 'list', ...)),
key=lambda v: traverse_obj(v, ('bitrate', 'video'), default=0), default=None)
if max_video is not None:
video_url = self._WAYBACK_BASE_URL + max_video.get('source')
urlh = self._request_webpage(HEADRequest(video_url), video_id, errnote=False,
fatal=False, note='Check video availablity')
if urlh:
formats.append({'url': video_url})
return {
'id': video_id,
'formats': formats,
**traverse_obj(player_info, ('postDetail', 'post', {
'title': ('officialVideo', 'title', {str}),
'creator': ('author', 'nickname', {str}),
'channel': ('channel', 'channelName', {str}),
'channel_id': ('channel', 'channelCode', {str}),
'duration': ('officialVideo', 'playTime', {int_or_none}),
'view_count': ('officialVideo', 'playCount', {int_or_none}),
'like_count': ('officialVideo', 'likeCount', {int_or_none}),
'comment_count': ('officialVideo', 'commentCount', {int_or_none}),
'timestamp': ('officialVideo', 'createdAt', {lambda x: int_or_none(x, scale=1000)}),
'release_timestamp': ('officialVideo', 'willStartAt', {lambda x: int_or_none(x, scale=1000)}),
})),
**traverse_obj(vod_data, ('meta', {
'uploader_id': ('user', 'id', {str}),
'uploader': ('user', 'name', {str}),
'uploader_url': ('user', 'url', {url_or_none}),
'thumbnail': ('cover', 'source', {url_or_none}),
}), expected_type=lambda x: x or None),
**NaverBaseIE.process_subtitles(vod_data, lambda x: [self._WAYBACK_BASE_URL + x]),
}
yt_dlp/extractor/arcpublishing.py
+6 -5
View File
@@ -4,6 +4,7 @@ from .common import InfoExtractor
from ..utils import (
extract_attributes,
int_or_none,
join_nonempty,
parse_iso8601,
try_get,
)
@@ -11,7 +12,7 @@ from ..utils import (
class ArcPublishingIE(InfoExtractor):
_UUID_REGEX = r'[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12}'
_VALID_URL = r'arcpublishing:(?P<org>[a-z]+):(?P<id>%s)' % _UUID_REGEX
_VALID_URL = rf'arcpublishing:(?P<org>[a-z]+):(?P<id>{_UUID_REGEX})'
_TESTS = [{
# https://www.adn.com/politics/2020/11/02/video-senate-candidates-campaign-in-anchorage-on-eve-of-election-day/
'url': 'arcpublishing:adn:8c99cb6e-b29c-4bc9-9173-7bf9979225ab',
@@ -74,12 +75,12 @@ class ArcPublishingIE(InfoExtractor):
def _extract_embed_urls(cls, url, webpage):
entries = []
# https://arcpublishing.atlassian.net/wiki/spaces/POWA/overview
for powa_el in re.findall(r'(<div[^>]+class="[^"]*\bpowa\b[^"]*"[^>]+data-uuid="%s"[^>]*>)' % ArcPublishingIE._UUID_REGEX, webpage):
for powa_el in re.findall(rf'(<div[^>]+class="[^"]*\bpowa\b[^"]*"[^>]+data-uuid="{ArcPublishingIE._UUID_REGEX}"[^>]*>)', webpage):
powa = extract_attributes(powa_el) or {}
org = powa.get('data-org')
uuid = powa.get('data-uuid')
if org and uuid:
entries.append('arcpublishing:%s:%s' % (org, uuid))
entries.append(f'arcpublishing:{org}:{uuid}')
return entries
def _real_extract(self, url):
@@ -122,7 +123,7 @@ class ArcPublishingIE(InfoExtractor):
elif stream_type in ('ts', 'hls'):
m3u8_formats = self._extract_m3u8_formats(
s_url, uuid, 'mp4', live=is_live, m3u8_id='hls', fatal=False)
if all([f.get('acodec') == 'none' for f in m3u8_formats]):
if all(f.get('acodec') == 'none' for f in m3u8_formats):
continue
for f in m3u8_formats:
height = f.get('height')
@@ -136,7 +137,7 @@ class ArcPublishingIE(InfoExtractor):
else:
vbr = int_or_none(s.get('bitrate'))
formats.append({
'format_id': '%s-%d' % (stream_type, vbr) if vbr else stream_type,
'format_id': join_nonempty(stream_type, vbr),
'vbr': vbr,
'width': int_or_none(s.get('width')),
'height': int_or_none(s.get('height')),
yt_dlp/extractor/ard.py
+301 -367
View File
@@ -1,23 +1,25 @@
import json
import functools
import re
from .common import InfoExtractor
from .generic import GenericIE
from ..utils import (
OnDemandPagedList,
bug_reports_message,
determine_ext,
ExtractorError,
int_or_none,
join_nonempty,
jwt_decode_hs256,
make_archive_id,
parse_duration,
qualities,
parse_iso8601,
remove_start,
str_or_none,
try_get,
unified_strdate,
unified_timestamp,
update_url_query,
url_or_none,
xpath_text,
)
from ..compat import compat_etree_fromstring
from ..utils.traversal import traverse_obj
class ARDMediathekBaseIE(InfoExtractor):
@@ -60,45 +62,6 @@ class ARDMediathekBaseIE(InfoExtractor):
'subtitles': subtitles,
}
def _ARD_extract_episode_info(self, title):
"""Try to extract season/episode data from the title."""
res = {}
if not title:
return res
for pattern in [
# Pattern for title like "Homo sapiens (S06/E07) - Originalversion"
# from: https://www.ardmediathek.de/one/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw
r'.*(?P<ep_info> \(S(?P<season_number>\d+)/E(?P<episode_number>\d+)\)).*',
# E.g.: title="Fritjof aus Norwegen (2) (AD)"
# from: https://www.ardmediathek.de/ard/sammlung/der-krieg-und-ich/68cMkqJdllm639Skj4c7sS/
r'.*(?P<ep_info> \((?:Folge |Teil )?(?P<episode_number>\d+)(?:/\d+)?\)).*',
r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:\:| -|) )\"(?P<episode>.+)\".*',
# E.g.: title="Folge 25/42: Symmetrie"
# from: https://www.ardmediathek.de/ard/video/grips-mathe/folge-25-42-symmetrie/ard-alpha/Y3JpZDovL2JyLmRlL3ZpZGVvLzMyYzI0ZjczLWQ1N2MtNDAxNC05ZmZhLTFjYzRkZDA5NDU5OQ/
# E.g.: title="Folge 1063 - Vertrauen"
# from: https://www.ardmediathek.de/ard/sendung/die-fallers/Y3JpZDovL3N3ci5kZS8yMzAyMDQ4/
r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:/\d+)?(?:\:| -|) ).*',
]:
m = re.match(pattern, title)
if m:
groupdict = m.groupdict()
res['season_number'] = int_or_none(groupdict.get('season_number'))
res['episode_number'] = int_or_none(groupdict.get('episode_number'))
res['episode'] = str_or_none(groupdict.get('episode'))
# Build the episode title by removing numeric episode information:
if groupdict.get('ep_info') and not res['episode']:
res['episode'] = str_or_none(
title.replace(groupdict.get('ep_info'), ''))
if res['episode']:
res['episode'] = res['episode'].strip()
break
# As a fallback use the whole title as the episode name:
if not res.get('episode'):
res['episode'] = title.strip()
return res
def _extract_formats(self, media_info, video_id):
type_ = media_info.get('_type')
media_array = media_info.get('_mediaArray', [])
@@ -122,7 +85,7 @@ class ARDMediathekBaseIE(InfoExtractor):
formats.extend(self._extract_f4m_formats(
update_url_query(stream_url, {
'hdcore': '3.1.1',
'plugin': 'aasp-3.1.1.69.124'
'plugin': 'aasp-3.1.1.69.124',
}), video_id, f4m_id='hds', fatal=False))
elif ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
@@ -133,12 +96,12 @@ class ARDMediathekBaseIE(InfoExtractor):
f = {
'url': server,
'play_path': stream_url,
'format_id': 'a%s-rtmp-%s' % (num, quality),
'format_id': f'a{num}-rtmp-{quality}',
}
else:
f = {
'url': stream_url,
'format_id': 'a%s-%s-%s' % (num, ext, quality)
'format_id': f'a{num}-{ext}-{quality}',
}
m = re.search(
r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$',
@@ -154,144 +117,12 @@ class ARDMediathekBaseIE(InfoExtractor):
return formats
class ARDMediathekIE(ARDMediathekBaseIE):
IE_NAME = 'ARD:mediathek'
_VALID_URL = r'^https?://(?:(?:(?:www|classic)\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
_TESTS = [{
# available till 26.07.2022
'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
'info_dict': {
'id': '44726822',
'ext': 'mp4',
'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
'duration': 1740,
},
'params': {
# m3u8 download
'skip_download': True,
}
}, {
'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
'only_matching': True,
}, {
# audio
'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
'only_matching': True,
}, {
'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
'only_matching': True,
}, {
# audio
'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
'only_matching': True,
}, {
'url': 'https://classic.ardmediathek.de/tv/Panda-Gorilla-Co/Panda-Gorilla-Co-Folge-274/Das-Erste/Video?bcastId=16355486&documentId=58234698',
'only_matching': True,
}]
@classmethod
def suitable(cls, url):
return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
def _real_extract(self, url):
# determine video id from url
m = self._match_valid_url(url)
document_id = None
numid = re.search(r'documentId=([0-9]+)', url)
if numid:
document_id = video_id = numid.group(1)
else:
video_id = m.group('video_id')
webpage = self._download_webpage(url, video_id)
ERRORS = (
('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
'Video %s is no longer available'),
)
for pattern, message in ERRORS:
if pattern in webpage:
raise ExtractorError(message % video_id, expected=True)
if re.search(r'[\?&]rss($|[=&])', url):
doc = compat_etree_fromstring(webpage.encode('utf-8'))
if doc.tag == 'rss':
return GenericIE()._extract_rss(url, video_id, doc)
title = self._og_search_title(webpage, default=None) or self._html_search_regex(
[r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
r'<meta name="dcterms\.title" content="(.*?)"/>',
r'<h4 class="headline">(.*?)</h4>',
r'<title[^>]*>(.*?)</title>'],
webpage, 'title')
description = self._og_search_description(webpage, default=None) or self._html_search_meta(
'dcterms.abstract', webpage, 'description', default=None)
if description is None:
description = self._html_search_meta(
'description', webpage, 'meta description', default=None)
if description is None:
description = self._html_search_regex(
r'<p\s+class="teasertext">(.+?)</p>',
webpage, 'teaser text', default=None)
# Thumbnail is sometimes not present.
# It is in the mobile version, but that seems to use a different URL
# structure altogether.
thumbnail = self._og_search_thumbnail(webpage, default=None)
media_streams = re.findall(r'''(?x)
mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
"([^"]+)"''', webpage)
if media_streams:
QUALITIES = qualities(['lo', 'hi', 'hq'])
formats = []
for furl in set(media_streams):
if furl.endswith('.f4m'):
fid = 'f4m'
else:
fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
fid = fid_m.group(1) if fid_m else None
formats.append({
'quality': QUALITIES(fid),
'format_id': fid,
'url': furl,
})
info = {
'formats': formats,
}
else: # request JSON file
if not document_id:
video_id = self._search_regex(
(r'/play/(?:config|media|sola)/(\d+)', r'contentId["\']\s*:\s*(\d+)'),
webpage, 'media id', default=None)
info = self._extract_media_info(
'http://www.ardmediathek.de/play/media/%s' % video_id,
webpage, video_id)
info.update({
'id': video_id,
'title': title,
'description': description,
'thumbnail': thumbnail,
})
info.update(self._ARD_extract_episode_info(info['title']))
return info
class ARDIE(InfoExtractor):
_VALID_URL = r'(?P<mainurl>https?://(?:www\.)?daserste\.de/(?:[^/?#&]+/)+(?P<id>[^/?#&]+))\.html'
_TESTS = [{
# available till 7.12.2023
'url': 'https://www.daserste.de/information/talk/maischberger/videos/maischberger-video-424.html',
'md5': 'a438f671e87a7eba04000336a119ccc4',
'md5': '94812e6438488fb923c361a44469614b',
'info_dict': {
'id': 'maischberger-video-424',
'display_id': 'maischberger-video-424',
@@ -398,16 +229,38 @@ class ARDIE(InfoExtractor):
}
class ARDBetaMediathekIE(ARDMediathekBaseIE):
class ARDBetaMediathekIE(InfoExtractor):
IE_NAME = 'ARDMediathek'
_VALID_URL = r'''(?x)https://
(?:(?:beta|www)\.)?ardmediathek\.de/
(?:(?P<client>[^/]+)/)?
(?:player|live|video|(?P<playlist>sendung|sammlung))/
(?:(?P<display_id>(?(playlist)[^?#]+?|[^?#]+))/)?
(?P<id>(?(playlist)|Y3JpZDovL)[a-zA-Z0-9]+)
(?(playlist)/(?P<season>\d+)?/?(?:[?#]|$))'''
(?:[^/]+/)?
(?:player|live|video)/
(?:[^?#]+/)?
(?P<id>[a-zA-Z0-9]+)
/?(?:[?#]|$)'''
_GEO_COUNTRIES = ['DE']
_TOKEN_URL = 'https://sso.ardmediathek.de/sso/token'
_TESTS = [{
'url': 'https://www.ardmediathek.de/video/filme-im-mdr/liebe-auf-vier-pfoten/mdr-fernsehen/Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
'md5': 'b6e8ab03f2bcc6e1f9e6cef25fcc03c4',
'info_dict': {
'display_id': 'Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
'id': '12939099',
'title': 'Liebe auf vier Pfoten',
'description': r're:^Claudia Schmitt, Anwältin in Salzburg',
'duration': 5222,
'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:aee7cbf8f06de976?w=960&ch=ae4d0f2ee47d8b9b',
'timestamp': 1701343800,
'upload_date': '20231130',
'ext': 'mp4',
'episode': 'Liebe auf vier Pfoten',
'series': 'Filme im MDR',
'age_limit': 0,
'channel': 'MDR',
'_old_archive_ids': ['ardbetamediathek Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0'],
},
}, {
'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/',
'md5': 'a1dc75a39c61601b980648f7c9f9f71d',
'info_dict': {
@@ -424,7 +277,7 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE):
'skip': 'Error',
}, {
'url': 'https://www.ardmediathek.de/video/tagesschau-oder-tagesschau-20-00-uhr/das-erste/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
'md5': 'f1837e563323b8a642a8ddeff0131f51',
'md5': '1e73ded21cb79bac065117e80c81dc88',
'info_dict': {
'id': '10049223',
'ext': 'mp4',
@@ -432,13 +285,31 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE):
'timestamp': 1636398000,
'description': 'md5:39578c7b96c9fe50afdf5674ad985e6b',
'upload_date': '20211108',
'display_id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
'duration': 915,
'episode': 'tagesschau, 20:00 Uhr',
'series': 'tagesschau',
'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:fbb21142783b0a49?w=960&ch=ee69108ae344f678',
'channel': 'ARD-Aktuell',
'_old_archive_ids': ['ardbetamediathek Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll'],
},
}, {
'url': 'https://www.ardmediathek.de/sendung/beforeigners/beforeigners/staffel-1/Y3JpZDovL2Rhc2Vyc3RlLmRlL2JlZm9yZWlnbmVycw/1',
'playlist_count': 6,
'url': 'https://www.ardmediathek.de/video/7-tage/7-tage-unter-harten-jungs/hr-fernsehen/N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
'md5': 'c428b9effff18ff624d4f903bda26315',
'info_dict': {
'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL2JlZm9yZWlnbmVycw',
'title': 'beforeigners/beforeigners/staffel-1',
'id': '94834686',
'ext': 'mp4',
'duration': 2700,
'episode': '7 Tage ... unter harten Jungs',
'description': 'md5:0f215470dcd2b02f59f4bd10c963f072',
'upload_date': '20231005',
'timestamp': 1696491171,
'display_id': 'N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
'series': '7 Tage ...',
'channel': 'HR',
'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:f6e6d5ffac41925c?w=960&ch=fa32ba69bc87989a',
'title': '7 Tage ... unter harten Jungs',
'_old_archive_ids': ['ardbetamediathek N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3'],
},
}, {
'url': 'https://beta.ardmediathek.de/ard/video/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
@@ -455,191 +326,254 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE):
}, {
'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
'only_matching': True,
}, {
'url': 'https://www.ardmediathek.de/video/coronavirus-update-ndr-info/astrazeneca-kurz-lockdown-und-pims-syndrom-81/ndr/Y3JpZDovL25kci5kZS84NzE0M2FjNi0wMWEwLTQ5ODEtOTE5NS1mOGZhNzdhOTFmOTI/',
'only_matching': True,
}]
def _extract_episode_info(self, title):
patterns = [
# Pattern for title like "Homo sapiens (S06/E07) - Originalversion"
# from: https://www.ardmediathek.de/one/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw
r'.*(?P<ep_info> \(S(?P<season_number>\d+)/E(?P<episode_number>\d+)\)).*',
# E.g.: title="Fritjof aus Norwegen (2) (AD)"
# from: https://www.ardmediathek.de/ard/sammlung/der-krieg-und-ich/68cMkqJdllm639Skj4c7sS/
r'.*(?P<ep_info> \((?:Folge |Teil )?(?P<episode_number>\d+)(?:/\d+)?\)).*',
r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:\:| -|) )\"(?P<episode>.+)\".*',
# E.g.: title="Folge 25/42: Symmetrie"
# from: https://www.ardmediathek.de/ard/video/grips-mathe/folge-25-42-symmetrie/ard-alpha/Y3JpZDovL2JyLmRlL3ZpZGVvLzMyYzI0ZjczLWQ1N2MtNDAxNC05ZmZhLTFjYzRkZDA5NDU5OQ/
# E.g.: title="Folge 1063 - Vertrauen"
# from: https://www.ardmediathek.de/ard/sendung/die-fallers/Y3JpZDovL3N3ci5kZS8yMzAyMDQ4/
r'.*(?P<ep_info>Folge (?P<episode_number>\d+)(?:/\d+)?(?:\:| -|) ).*',
# As a fallback use the full title
r'(?P<title>.*)',
]
return traverse_obj(patterns, (..., {functools.partial(re.match, string=title)}, {
'season_number': ('season_number', {int_or_none}),
'episode_number': ('episode_number', {int_or_none}),
'episode': ((
('episode', {str_or_none}),
('ep_info', {lambda x: title.replace(x, '')}),
('title', {str}),
), {str.strip}),
}), get_all=False)
def _real_extract(self, url):
display_id = self._match_id(url)
query = {'embedded': 'false', 'mcV6': 'true'}
headers = {}
if self._get_cookies(self._TOKEN_URL).get('ams'):
token = self._download_json(
self._TOKEN_URL, display_id, 'Fetching token for age verification',
'Unable to fetch age verification token', fatal=False)
id_token = traverse_obj(token, ('idToken', {str}))
decoded_token = traverse_obj(id_token, ({jwt_decode_hs256}, {dict}))
user_id = traverse_obj(decoded_token, (('user_id', 'sub'), {str}), get_all=False)
if not user_id:
self.report_warning('Unable to extract token, continuing without authentication')
else:
headers['x-authorization'] = f'Bearer {id_token}'
query['userId'] = user_id
if decoded_token.get('age_rating') != 18:
self.report_warning('Account is not verified as 18+; video may be unavailable')
page_data = self._download_json(
f'https://api.ardmediathek.de/page-gateway/pages/ard/item/{display_id}',
display_id, query=query, headers=headers)
# For user convenience we use the old contentId instead of the longer crid
# Ref: https://github.com/yt-dlp/yt-dlp/issues/8731#issuecomment-1874398283
old_id = traverse_obj(page_data, ('tracking', 'atiCustomVars', 'contentId', {int}))
if old_id is not None:
video_id = str(old_id)
archive_ids = [make_archive_id(ARDBetaMediathekIE, display_id)]
else:
self.report_warning(f'Could not extract contentId{bug_reports_message()}')
video_id = display_id
archive_ids = None
player_data = traverse_obj(
page_data, ('widgets', lambda _, v: v['type'] in ('player_ondemand', 'player_live'), {dict}), get_all=False)
is_live = player_data.get('type') == 'player_live'
media_data = traverse_obj(player_data, ('mediaCollection', 'embedded', {dict}))
if player_data.get('blockedByFsk'):
self.raise_login_required('This video is only available for age verified users or after 22:00')
formats = []
subtitles = {}
for stream in traverse_obj(media_data, ('streams', ..., {dict})):
kind = stream.get('kind')
# Prioritize main stream over sign language and others
preference = 1 if kind == 'main' else None
for media in traverse_obj(stream, ('media', lambda _, v: url_or_none(v['url']))):
media_url = media['url']
audio_kind = traverse_obj(media, (
'audios', 0, 'kind', {str}), default='').replace('standard', '')
lang_code = traverse_obj(media, ('audios', 0, 'languageCode', {str})) or 'deu'
lang = join_nonempty(lang_code, audio_kind)
language_preference = 10 if lang == 'deu' else -10
if determine_ext(media_url) == 'm3u8':
fmts, subs = self._extract_m3u8_formats_and_subtitles(
media_url, video_id, m3u8_id=f'hls-{kind}', preference=preference, fatal=False, live=is_live)
for f in fmts:
f['language'] = lang
f['language_preference'] = language_preference
formats.extend(fmts)
self._merge_subtitles(subs, target=subtitles)
else:
formats.append({
'url': media_url,
'format_id': f'http-{kind}',
'preference': preference,
'language': lang,
'language_preference': language_preference,
**traverse_obj(media, {
'format_note': ('forcedLabel', {str}),
'width': ('maxHResolutionPx', {int_or_none}),
'height': ('maxVResolutionPx', {int_or_none}),
'vcodec': ('videoCodec', {str}),
}),
})
for sub in traverse_obj(media_data, ('subtitles', ..., {dict})):
for sources in traverse_obj(sub, ('sources', lambda _, v: url_or_none(v['url']))):
subtitles.setdefault(sub.get('languageCode') or 'deu', []).append({
'url': sources['url'],
'ext': {'webvtt': 'vtt', 'ebutt': 'ttml'}.get(sources.get('kind')),
})
age_limit = traverse_obj(page_data, ('fskRating', {lambda x: remove_start(x, 'FSK')}, {int_or_none}))
return {
'id': video_id,
'display_id': display_id,
'formats': formats,
'subtitles': subtitles,
'is_live': is_live,
'age_limit': age_limit,
**traverse_obj(media_data, ('meta', {
'title': 'title',
'description': 'synopsis',
'timestamp': ('broadcastedOnDateTime', {parse_iso8601}),
'series': 'seriesTitle',
'thumbnail': ('images', 0, 'url', {url_or_none}),
'duration': ('durationSeconds', {int_or_none}),
'channel': 'clipSourceName',
})),
**self._extract_episode_info(page_data.get('title')),
'_old_archive_ids': archive_ids,
}
class ARDMediathekCollectionIE(InfoExtractor):
_VALID_URL = r'''(?x)https://
(?:(?:beta|www)\.)?ardmediathek\.de/
(?:[^/?#]+/)?
(?P<playlist>sendung|serie|sammlung)/
(?:(?P<display_id>[^?#]+?)/)?
(?P<id>[a-zA-Z0-9]+)
(?:/(?P<season>\d+)(?:/(?P<version>OV|AD))?)?/?(?:[?#]|$)'''
_GEO_COUNTRIES = ['DE']
_TESTS = [{
'url': 'https://www.ardmediathek.de/serie/quiz/staffel-1-originalversion/Y3JpZDovL3dkci5kZS9vbmUvcXVpeg/1/OV',
'info_dict': {
'id': 'Y3JpZDovL3dkci5kZS9vbmUvcXVpeg_1_OV',
'display_id': 'quiz/staffel-1-originalversion',
'title': 'Staffel 1 Originalversion',
},
'playlist_count': 3,
}, {
'url': 'https://www.ardmediathek.de/serie/babylon-berlin/staffel-4-mit-audiodeskription/Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu/4/AD',
'info_dict': {
'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu_4_AD',
'display_id': 'babylon-berlin/staffel-4-mit-audiodeskription',
'title': 'Staffel 4 mit Audiodeskription',
},
'playlist_count': 12,
}, {
'url': 'https://www.ardmediathek.de/serie/babylon-berlin/staffel-1/Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu/1/',
'info_dict': {
'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL2JhYnlsb24tYmVybGlu_1',
'display_id': 'babylon-berlin/staffel-1',
'title': 'Staffel 1',
},
'playlist_count': 8,
}, {
'url': 'https://www.ardmediathek.de/sendung/tatort/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydA',
'info_dict': {
'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydA',
'display_id': 'tatort',
'title': 'Tatort',
},
'playlist_mincount': 500,
}, {
'url': 'https://www.ardmediathek.de/sammlung/die-kirche-bleibt-im-dorf/5eOHzt8XB2sqeFXbIoJlg2',
'info_dict': {
'id': '5eOHzt8XB2sqeFXbIoJlg2',
'display_id': 'die-kirche-bleibt-im-dorf',
'title': 'Die Kirche bleibt im Dorf',
'description': 'Die Kirche bleibt im Dorf',
},
'playlist_count': 4,
}, {
# playlist of type 'sendung'
'url': 'https://www.ardmediathek.de/ard/sendung/doctor-who/Y3JpZDovL3dkci5kZS9vbmUvZG9jdG9yIHdobw/',
'only_matching': True,
}, {
# playlist of type 'serie'
'url': 'https://www.ardmediathek.de/serie/nachtstreife/staffel-1/Y3JpZDovL3N3ci5kZS9zZGIvc3RJZC8xMjQy/1',
'only_matching': True,
}, {
# playlist of type 'sammlung'
'url': 'https://www.ardmediathek.de/ard/sammlung/team-muenster/5JpTzLSbWUAK8184IOvEir/',
'only_matching': True,
}, {
'url': 'https://www.ardmediathek.de/video/coronavirus-update-ndr-info/astrazeneca-kurz-lockdown-und-pims-syndrom-81/ndr/Y3JpZDovL25kci5kZS84NzE0M2FjNi0wMWEwLTQ5ODEtOTE5NS1mOGZhNzdhOTFmOTI/',
'only_matching': True,
}, {
'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3dkci5kZS9CZWl0cmFnLWQ2NDJjYWEzLTMwZWYtNGI4NS1iMTI2LTU1N2UxYTcxOGIzOQ/tatort-duo-koeln-leipzig-ihr-kinderlein-kommet',
'only_matching': True,
}]
def _ARD_load_playlist_snipped(self, playlist_id, display_id, client, mode, pageNumber):
""" Query the ARD server for playlist information
and returns the data in "raw" format """
if mode == 'sendung':
graphQL = json.dumps({
'query': '''{
showPage(
client: "%s"
showId: "%s"
pageNumber: %d
) {
pagination {
pageSize
totalElements
}
teasers { # Array
mediumTitle
links { target { id href title } }
type
}
}}''' % (client, playlist_id, pageNumber),
}).encode()
else: # mode == 'sammlung'
graphQL = json.dumps({
'query': '''{
morePage(
client: "%s"
compilationId: "%s"
pageNumber: %d
) {
widget {
pagination {
pageSize
totalElements
}
teasers { # Array
mediumTitle
links { target { id href title } }
type
}
}
}}''' % (client, playlist_id, pageNumber),
}).encode()
# Ressources for ARD graphQL debugging:
# https://api-test.ardmediathek.de/public-gateway
show_page = self._download_json(
'https://api.ardmediathek.de/public-gateway',
'[Playlist] %s' % display_id,
data=graphQL,
headers={'Content-Type': 'application/json'})['data']
# align the structure of the returned data:
if mode == 'sendung':
show_page = show_page['showPage']
else: # mode == 'sammlung'
show_page = show_page['morePage']['widget']
return show_page
def _ARD_extract_playlist(self, url, playlist_id, display_id, client, mode):
""" Collects all playlist entries and returns them as info dict.
Supports playlists of mode 'sendung' and 'sammlung', and also nested
playlists. """
entries = []
pageNumber = 0
while True: # iterate by pageNumber
show_page = self._ARD_load_playlist_snipped(
playlist_id, display_id, client, mode, pageNumber)
for teaser in show_page['teasers']: # process playlist items
if '/compilation/' in teaser['links']['target']['href']:
# alternativ cond.: teaser['type'] == "compilation"
# => This is an nested compilation, e.g. like:
# https://www.ardmediathek.de/ard/sammlung/die-kirche-bleibt-im-dorf/5eOHzt8XB2sqeFXbIoJlg2/
link_mode = 'sammlung'
else:
link_mode = 'video'
item_url = 'https://www.ardmediathek.de/%s/%s/%s/%s/%s' % (
client, link_mode, display_id,
# perform HTLM quoting of episode title similar to ARD:
re.sub('^-|-$', '', # remove '-' from begin/end
re.sub('[^a-zA-Z0-9]+', '-', # replace special chars by -
teaser['links']['target']['title'].lower()
.replace('ä', 'ae').replace('ö', 'oe')
.replace('ü', 'ue').replace('ß', 'ss'))),
teaser['links']['target']['id'])
entries.append(self.url_result(
item_url,
ie=ARDBetaMediathekIE.ie_key()))
if (show_page['pagination']['pageSize'] * (pageNumber + 1)
>= show_page['pagination']['totalElements']):
# we've processed enough pages to get all playlist entries
break
pageNumber = pageNumber + 1
return self.playlist_result(entries, playlist_id, playlist_title=display_id)
_PAGE_SIZE = 100
def _real_extract(self, url):
video_id, display_id, playlist_type, client, season_number = self._match_valid_url(url).group(
'id', 'display_id', 'playlist', 'client', 'season')
display_id, client = display_id or video_id, client or 'ard'
playlist_id, display_id, playlist_type, season_number, version = self._match_valid_url(url).group(
'id', 'display_id', 'playlist', 'season', 'version')
if playlist_type:
# TODO: Extract only specified season
return self._ARD_extract_playlist(url, video_id, display_id, client, playlist_type)
def call_api(page_num):
api_path = 'compilations/ard' if playlist_type == 'sammlung' else 'widgets/ard/asset'
return self._download_json(
f'https://api.ardmediathek.de/page-gateway/{api_path}/{playlist_id}', playlist_id,
f'Downloading playlist page {page_num}', query={
'pageNumber': page_num,
'pageSize': self._PAGE_SIZE,
**({
'seasoned': 'true',
'seasonNumber': season_number,
'withOriginalversion': 'true' if version == 'OV' else 'false',
'withAudiodescription': 'true' if version == 'AD' else 'false',
} if season_number else {}),
})
player_page = self._download_json(
'https://api.ardmediathek.de/public-gateway',
display_id, data=json.dumps({
'query': '''{
playerPage(client:"%s", clipId: "%s") {
blockedByFsk
broadcastedOn
maturityContentRating
mediaCollection {
_duration
_geoblocked
_isLive
_mediaArray {
_mediaStreamArray {
_quality
_server
_stream
}
}
_previewImage
_subtitleUrl
_type
}
show {
title
}
synopsis
title
tracking {
atiCustomVars {
contentId
}
}
}
}''' % (client, video_id),
}).encode(), headers={
'Content-Type': 'application/json'
})['data']['playerPage']
title = player_page['title']
content_id = str_or_none(try_get(
player_page, lambda x: x['tracking']['atiCustomVars']['contentId']))
media_collection = player_page.get('mediaCollection') or {}
if not media_collection and content_id:
media_collection = self._download_json(
'https://www.ardmediathek.de/play/media/' + content_id,
content_id, fatal=False) or {}
info = self._parse_media_info(
media_collection, content_id or video_id,
player_page.get('blockedByFsk'))
age_limit = None
description = player_page.get('synopsis')
maturity_content_rating = player_page.get('maturityContentRating')
if maturity_content_rating:
age_limit = int_or_none(maturity_content_rating.lstrip('FSK'))
if not age_limit and description:
age_limit = int_or_none(self._search_regex(
r'\(FSK\s*(\d+)\)\s*$', description, 'age limit', default=None))
info.update({
'age_limit': age_limit,
'display_id': display_id,
'title': title,
'description': description,
'timestamp': unified_timestamp(player_page.get('broadcastedOn')),
'series': try_get(player_page, lambda x: x['show']['title']),
})
info.update(self._ARD_extract_episode_info(info['title']))
return info
def fetch_page(page_num):
for item in traverse_obj(call_api(page_num), ('teasers', ..., {dict})):
item_id = traverse_obj(item, ('links', 'target', ('urlId', 'id')), 'id', get_all=False)
if not item_id or item_id == playlist_id:
continue
item_mode = 'sammlung' if item.get('type') == 'compilation' else 'video'
yield self.url_result(
f'https://www.ardmediathek.de/{item_mode}/{item_id}',
ie=(ARDMediathekCollectionIE if item_mode == 'sammlung' else ARDBetaMediathekIE),
**traverse_obj(item, {
'id': ('id', {str}),
'title': ('longTitle', {str}),
'duration': ('duration', {int_or_none}),
'timestamp': ('broadcastedOn', {parse_iso8601}),
}))
page_data = call_api(0)
full_id = join_nonempty(playlist_id, season_number, version, delim='_')
return self.playlist_result(
OnDemandPagedList(fetch_page, self._PAGE_SIZE), full_id, display_id=display_id,
title=page_data.get('title'), description=page_data.get('synopsis'))
yt_dlp/extractor/arkena.py
+3 -3
View File
@@ -64,7 +64,7 @@ class ArkenaIE(InfoExtractor):
raise ExtractorError('Invalid URL', expected=True)
media = self._download_json(
'https://video.qbrick.com/api/v1/public/accounts/%s/medias/%s' % (account_id, video_id),
f'https://video.qbrick.com/api/v1/public/accounts/{account_id}/medias/{video_id}',
video_id, query={
# https://video.qbrick.com/docs/api/examples/library-api.html
'fields': 'asset/resources/*/renditions/*(height,id,language,links/*(href,mimeType),type,size,videos/*(audios/*(codec,sampleRate),bitrate,codec,duration,height,width),width),created,metadata/*(title,description),tags',
@@ -131,8 +131,8 @@ class ArkenaIE(InfoExtractor):
formats.extend(self._extract_f4m_formats(
href, video_id, f4m_id='hds', fatal=False))
elif mime_type == 'application/dash+xml':
formats.extend(self._extract_f4m_formats(
href, video_id, f4m_id='hds', fatal=False))
formats.extend(self._extract_mpd_formats(
href, video_id, mpd_id='dash', fatal=False))
elif mime_type == 'application/vnd.ms-sstr+xml':
formats.extend(self._extract_ism_formats(
href, video_id, ism_id='mss', fatal=False))
yt_dlp/extractor/arnes.py
+6 -8
View File
@@ -1,11 +1,9 @@
import urllib.parse
from .common import InfoExtractor
from ..compat import (
compat_parse_qs,
compat_urllib_parse_urlparse,
)
from ..utils import (
format_field,
float_or_none,
format_field,
int_or_none,
parse_iso8601,
remove_start,
@@ -35,7 +33,7 @@ class ArnesIE(InfoExtractor):
'view_count': int,
'tags': ['linearna_algebra'],
'start_time': 10,
}
},
}, {
'url': 'https://video.arnes.si/api/asset/s1YjnV7hadlC/play.mp4',
'only_matching': True,
@@ -93,6 +91,6 @@ class ArnesIE(InfoExtractor):
'duration': float_or_none(video.get('duration'), 1000),
'view_count': int_or_none(video.get('views')),
'tags': video.get('hashtags'),
'start_time': int_or_none(compat_parse_qs(
compat_urllib_parse_urlparse(url).query).get('t', [None])[0]),
'start_time': int_or_none(urllib.parse.parse_qs(
urllib.parse.urlparse(url).query).get('t', [None])[0]),
}
yt_dlp/extractor/art19.py
+303
View File
@@ -0,0 +1,303 @@
import re
from .common import InfoExtractor
from ..utils import float_or_none, int_or_none, parse_iso8601, url_or_none
from ..utils.traversal import traverse_obj
class Art19IE(InfoExtractor):
_UUID_REGEX = r'[\da-f]{8}-?[\da-f]{4}-?[\da-f]{4}-?[\da-f]{4}-?[\da-f]{12}'
_VALID_URL = [
rf'https?://(?:www\.)?art19\.com/shows/[^/#?]+/episodes/(?P<id>{_UUID_REGEX})',
rf'https?://rss\.art19\.com/episodes/(?P<id>{_UUID_REGEX})\.mp3',
]
_EMBED_REGEX = [rf'<iframe[^>]+\bsrc=[\'"](?P<url>{_VALID_URL[0]})']
_TESTS = [{
'url': 'https://rss.art19.com/episodes/5ba1413c-48b8-472b-9cc3-cfd952340bdb.mp3',
'info_dict': {
'id': '5ba1413c-48b8-472b-9cc3-cfd952340bdb',
'ext': 'mp3',
'title': 'Why Did DeSantis Drop Out?',
'series': 'The Daily Briefing',
'release_timestamp': 1705941275,
'description': 'md5:da38961da4a3f7e419471365e3c6b49f',
'episode': 'Episode 582',
'thumbnail': r're:^https?://content\.production\.cdn\.art19\.com.*\.jpeg$',
'series_id': 'ed52a0ab-08b1-4def-8afc-549e4d93296d',
'upload_date': '20240122',
'timestamp': 1705940815,
'episode_number': 582,
'modified_date': '20240122',
'episode_id': '5ba1413c-48b8-472b-9cc3-cfd952340bdb',
'modified_timestamp': 1705941275,
'release_date': '20240122',
'duration': 527.4,
},
}, {
'url': 'https://art19.com/shows/scamfluencers/episodes/8319b776-4153-4d22-8630-631f204a03dd',
'info_dict': {
'id': '8319b776-4153-4d22-8630-631f204a03dd',
'ext': 'mp3',
'title': 'Martha Stewart: The Homemaker Hustler Part 2',
'modified_date': '20240116',
'upload_date': '20240105',
'modified_timestamp': 1705435802,
'episode_id': '8319b776-4153-4d22-8630-631f204a03dd',
'series_id': 'd3c9b8ca-26b3-42f4-9bd8-21d1a9031e75',
'thumbnail': r're:^https?://content\.production\.cdn\.art19\.com.*\.jpeg$',
'description': 'md5:4aa7cfd1358dc57e729835bc208d7893',
'release_timestamp': 1705305660,
'release_date': '20240115',
'timestamp': 1704481536,
'episode_number': 88,
'series': 'Scamfluencers',
'duration': 2588.37501,
'episode': 'Episode 88',
},
}]
_WEBPAGE_TESTS = [{
'url': 'https://www.nu.nl/formule-1/6291456/verstappen-wordt-een-synoniem-voor-formule-1.html',
'info_dict': {
'id': '7d42626a-7301-47db-bb8a-3b6f054d77d7',
'ext': 'mp3',
'title': "'Verstappen wordt een synoniem voor Formule 1'",
'season': 'Seizoen 6',
'description': 'md5:39a7159a31c4cda312b2e893bdd5c071',
'episode_id': '7d42626a-7301-47db-bb8a-3b6f054d77d7',
'duration': 3061.82111,
'series_id': '93f4e113-2a60-4609-a564-755058fa40d8',
'release_date': '20231126',
'modified_timestamp': 1701156004,
'thumbnail': r're:^https?://content\.production\.cdn\.art19\.com.*\.jpeg$',
'season_number': 6,
'episode_number': 52,
'modified_date': '20231128',
'upload_date': '20231126',
'timestamp': 1701025981,
'season_id': '36097c1e-7455-490d-a2fe-e2f10b4d5f26',
'series': 'De Boordradio',
'release_timestamp': 1701026308,
'episode': 'Episode 52',
},
}, {
'url': 'https://www.wishtv.com/podcast-episode/larry-bucshon-announces-retirement-from-congress/',
'info_dict': {
'id': '8da368bd-08d1-46d0-afaa-c134a4af7dc0',
'ext': 'mp3',
'title': 'Larry Bucshon announces retirement from congress',
'upload_date': '20240115',
'episode_number': 148,
'episode': 'Episode 148',
'thumbnail': r're:^https?://content\.production\.cdn\.art19\.com.*\.jpeg$',
'release_date': '20240115',
'timestamp': 1705328205,
'release_timestamp': 1705329275,
'series': 'All INdiana Politics',
'modified_date': '20240117',
'modified_timestamp': 1705458901,
'series_id': 'c4af6c27-b10f-4ff2-9f84-0f407df86ff1',
'episode_id': '8da368bd-08d1-46d0-afaa-c134a4af7dc0',
'description': 'md5:53b5239e4d14973a87125c217c255b2a',
'duration': 1256.18848,
},
}]
@classmethod
def _extract_embed_urls(cls, url, webpage):
yield from super()._extract_embed_urls(url, webpage)
for episode_id in re.findall(
rf'<div[^>]+\bclass=[\'"][^\'"]*art19-web-player[^\'"]*[\'"][^>]+\bdata-episode-id=[\'"]({cls._UUID_REGEX})[\'"]', webpage):
yield f'https://rss.art19.com/episodes/{episode_id}.mp3'
def _real_extract(self, url):
episode_id = self._match_id(url)
player_metadata = self._download_json(
f'https://art19.com/episodes/{episode_id}', episode_id,
note='Downloading player metadata', fatal=False,
headers={'Accept': 'application/vnd.art19.v0+json'})
rss_metadata = self._download_json(
f'https://rss.art19.com/episodes/{episode_id}.json', episode_id, fatal=False,
note='Downloading RSS metadata')
formats = [{
'format_id': 'direct',
'url': f'https://rss.art19.com/episodes/{episode_id}.mp3',
'vcodec': 'none',
'acodec': 'mp3',
}]
for fmt_id, fmt_data in traverse_obj(rss_metadata, ('content', 'media', {dict.items}, ...)):
if fmt_id == 'waveform_bin':
continue
fmt_url = traverse_obj(fmt_data, ('url', {url_or_none}))
if not fmt_url:
continue
formats.append({
'format_id': fmt_id,
'url': fmt_url,
'vcodec': 'none',
'acodec': fmt_id,
'quality': -2 if fmt_id == 'ogg' else -1,
})
return {
'id': episode_id,
'formats': formats,
**traverse_obj(player_metadata, ('episode', {
'title': ('title', {str}),
'description': ('description_plain', {str}),
'episode_id': ('id', {str}),
'episode_number': ('episode_number', {int_or_none}),
'season_id': ('season_id', {str}),
'series_id': ('series_id', {str}),
'timestamp': ('created_at', {parse_iso8601}),
'release_timestamp': ('released_at', {parse_iso8601}),
'modified_timestamp': ('updated_at', {parse_iso8601}),
})),
**traverse_obj(rss_metadata, ('content', {
'title': ('episode_title', {str}),
'description': ('episode_description_plain', {str}),
'episode_id': ('episode_id', {str}),
'episode_number': ('episode_number', {int_or_none}),
'season': ('season_title', {str}),
'season_id': ('season_id', {str}),
'season_number': ('season_number', {int_or_none}),
'series': ('series_title', {str}),
'series_id': ('series_id', {str}),
'thumbnail': ('cover_image', {url_or_none}),
'duration': ('duration', {float_or_none}),
})),
}
class Art19ShowIE(InfoExtractor):
_VALID_URL_BASE = r'https?://(?:www\.)?art19\.com/shows/(?P<id>[\w-]+)(?:/embed)?/?'
_VALID_URL = [
rf'{_VALID_URL_BASE}(?:$|[#?])',
r'https?://rss\.art19\.com/(?P<id>[\w-]+)/?(?:$|[#?])',
]
_EMBED_REGEX = [rf'<iframe[^>]+\bsrc=[\'"](?P<url>{_VALID_URL_BASE}[^\'"])']
_TESTS = [{
'url': 'https://www.art19.com/shows/5898c087-a14f-48dc-b6fc-a2280a1ff6e0/',
'info_dict': {
'_type': 'playlist',
'id': '5898c087-a14f-48dc-b6fc-a2280a1ff6e0',
'display_id': 'echt-gebeurd',
'title': 'Echt Gebeurd',
'description': 'md5:5fd11dc80b76e51ffd34b6067fd5e560',
'timestamp': 1492642167,
'upload_date': '20170419',
'modified_timestamp': int,
'modified_date': str,
'tags': 'count:7',
},
'playlist_mincount': 425,
}, {
'url': 'https://www.art19.com/shows/echt-gebeurd',
'info_dict': {
'_type': 'playlist',
'id': '5898c087-a14f-48dc-b6fc-a2280a1ff6e0',
'display_id': 'echt-gebeurd',
'title': 'Echt Gebeurd',
'description': 'md5:5fd11dc80b76e51ffd34b6067fd5e560',
'timestamp': 1492642167,
'upload_date': '20170419',
'modified_timestamp': int,
'modified_date': str,
'tags': 'count:7',
},
'playlist_mincount': 425,
}, {
'url': 'https://rss.art19.com/scamfluencers',
'info_dict': {
'_type': 'playlist',
'id': 'd3c9b8ca-26b3-42f4-9bd8-21d1a9031e75',
'display_id': 'scamfluencers',
'title': 'Scamfluencers',
'description': 'md5:7d239d670c0ced6dadbf71c4caf764b7',
'timestamp': 1647368573,
'upload_date': '20220315',
'modified_timestamp': int,
'modified_date': str,
'tags': [],
},
'playlist_mincount': 90,
}, {
'url': 'https://art19.com/shows/enthuellt/embed',
'info_dict': {
'_type': 'playlist',
'id': 'e2cacf57-bb8a-4263-aa81-719bcdd4f80c',
'display_id': 'enthuellt',
'title': 'Enthüllt',
'description': 'md5:17752246643414a2fd51744fc9a1c08e',
'timestamp': 1601645860,
'upload_date': '20201002',
'modified_timestamp': int,
'modified_date': str,
'tags': 'count:10',
},
'playlist_mincount': 10,
}]
_WEBPAGE_TESTS = [{
'url': 'https://deconstructingyourself.com/deconstructing-yourself-podcast',
'info_dict': {
'_type': 'playlist',
'id': 'cfbb9b01-c295-4adb-8726-adde7c03cf21',
'display_id': 'deconstructing-yourself',
'title': 'Deconstructing Yourself',
'description': 'md5:dab5082b28b248a35476abf64768854d',
'timestamp': 1570581181,
'upload_date': '20191009',
'modified_timestamp': int,
'modified_date': str,
'tags': 'count:5',
},
'playlist_mincount': 80,
}, {
'url': 'https://chicagoreader.com/columns-opinion/podcasts/ben-joravsky-show-podcast-episodes/',
'info_dict': {
'_type': 'playlist',
'id': '9dfa2c37-ab87-4c13-8388-4897914313ec',
'display_id': 'the-ben-joravsky-show',
'title': 'The Ben Joravsky Show',
'description': 'md5:c0f3ec0ee0dbea764390e521adc8780a',
'timestamp': 1550875095,
'upload_date': '20190222',
'modified_timestamp': int,
'modified_date': str,
'tags': ['Chicago Politics', 'chicago', 'Ben Joravsky'],
},
'playlist_mincount': 1900,
}]
@classmethod
def _extract_embed_urls(cls, url, webpage):
yield from super()._extract_embed_urls(url, webpage)
for series_id in re.findall(
r'<div[^>]+\bclass=[\'"][^\'"]*art19-web-player[^\'"]*[\'"][^>]+\bdata-series-id=[\'"]([\w-]+)[\'"]', webpage):
yield f'https://art19.com/shows/{series_id}'
def _real_extract(self, url):
series_id = self._match_id(url)
series_metadata = self._download_json(
f'https://art19.com/series/{series_id}', series_id, note='Downloading series metadata',
headers={'Accept': 'application/vnd.art19.v0+json'})
return {
'_type': 'playlist',
'entries': [
self.url_result(f'https://rss.art19.com/episodes/{episode_id}.mp3', Art19IE)
for episode_id in traverse_obj(series_metadata, ('series', 'episode_ids', ..., {str}))
],
**traverse_obj(series_metadata, ('series', {
'id': ('id', {str}),
'display_id': ('slug', {str}),
'title': ('title', {str}),
'description': ('description_plain', {str}),
'timestamp': ('created_at', {parse_iso8601}),
'modified_timestamp': ('updated_at', {parse_iso8601}),
})),
'tags': traverse_obj(series_metadata, ('tags', ..., 'name', {str})),
}
yt_dlp/extractor/arte.py
+62 -51
View File
@@ -5,6 +5,7 @@ from ..utils import (
ExtractorError,
GeoRestrictedError,
int_or_none,
join_nonempty,
parse_iso8601,
parse_qs,
strip_or_none,
@@ -19,46 +20,22 @@ class ArteTVBaseIE(InfoExtractor):
class ArteTVIE(ArteTVBaseIE):
_VALID_URL = r'''(?x)
_VALID_URL = rf'''(?x)
(?:https?://
(?:
(?:www\.)?arte\.tv/(?P<lang>%(langs)s)/videos|
api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>%(langs)s)
(?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos|
api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>{ArteTVBaseIE._ARTE_LANGUAGES})
)
|arte://program)
/(?P<id>\d{6}-\d{3}-[AF]|LIVE)
''' % {'langs': ArteTVBaseIE._ARTE_LANGUAGES}
/(?P<id>\d{{6}}-\d{{3}}-[AF]|LIVE)
'''
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
'only_matching': True,
}, {
'url': 'https://www.arte.tv/pl/videos/100103-000-A/usa-dyskryminacja-na-porodowce/',
'info_dict': {
'id': '100103-000-A',
'title': 'USA: Dyskryminacja na porodówce',
'description': 'md5:242017b7cce59ffae340a54baefcafb1',
'alt_title': 'ARTE Reportage',
'upload_date': '20201103',
'duration': 554,
'thumbnail': r're:https://api-cdn\.arte\.tv/.+940x530',
'timestamp': 1604417980,
'ext': 'mp4',
},
'params': {'skip_download': 'm3u8'}
}, {
'note': 'No alt_title',
'url': 'https://www.arte.tv/fr/videos/110371-000-A/la-chaleur-supplice-des-arbres-de-rue/',
'info_dict': {
'id': '110371-000-A',
'ext': 'mp4',
'upload_date': '20220718',
'duration': 154,
'timestamp': 1658162460,
'description': 'md5:5890f36fe7dccfadb8b7c0891de54786',
'title': 'La chaleur, supplice des arbres de rue',
'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/CPE2sQDtD8GLQgt8DuYHLf/940x530',
},
'params': {'skip_download': 'm3u8'}
'only_matching': True,
}, {
'url': 'https://api.arte.tv/api/player/v2/config/de/100605-013-A',
'only_matching': True,
@@ -67,19 +44,38 @@ class ArteTVIE(ArteTVBaseIE):
'only_matching': True,
}, {
'url': 'https://www.arte.tv/de/videos/110203-006-A/zaz/',
'only_matching': True,
}, {
'url': 'https://www.arte.tv/fr/videos/109067-000-A/la-loi-de-teheran/',
'info_dict': {
'id': '110203-006-A',
'chapters': 'count:16',
'description': 'md5:cf592f1df52fe52007e3f8eac813c084',
'alt_title': 'Zaz',
'title': 'Baloise Session 2022',
'timestamp': 1668445200,
'duration': 4054,
'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/ubQjmVCGyRx3hmBuZEK9QZ/940x530',
'upload_date': '20221114',
'id': '109067-000-A',
'ext': 'mp4',
'description': 'md5:d2ca367b8ecee028dddaa8bd1aebc739',
'timestamp': 1713927600,
'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/3rR6PLzfbigSkkeHtkCZNF/940x530',
'duration': 7599,
'title': 'La loi de Téhéran',
'upload_date': '20240424',
'subtitles': {
'fr': 'mincount:1',
'fr-acc': 'mincount:1',
'fr-forced': 'mincount:1',
},
},
}, {
'note': 'age-restricted',
'url': 'https://www.arte.tv/de/videos/006785-000-A/the-element-of-crime/',
'info_dict': {
'id': '006785-000-A',
'description': 'md5:c2f94fdfefc8a280e4dab68ab96ab0ba',
'title': 'The Element of Crime',
'timestamp': 1696111200,
'duration': 5849,
'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/q82dTTfyuCXupPsGxXsd7B/940x530',
'upload_date': '20230930',
'ext': 'mp4',
},
'expected_warnings': ['geo restricted']
'skip': '404 Not Found',
}]
_GEO_BYPASS = True
@@ -130,13 +126,27 @@ class ArteTVIE(ArteTVBaseIE):
),
}
@staticmethod
def _fix_accessible_subs_locale(subs):
updated_subs = {}
for lang, sub_formats in subs.items():
for fmt in sub_formats:
url = fmt.get('url') or ''
suffix = ('acc' if url.endswith('-MAL.m3u8')
else 'forced' if '_VO' not in url
else None)
updated_subs.setdefault(join_nonempty(lang, suffix), []).append(fmt)
return updated_subs
def _real_extract(self, url):
mobj = self._match_valid_url(url)
video_id = mobj.group('id')
lang = mobj.group('lang') or mobj.group('lang_2')
langauge_code = self._LANG_MAP.get(lang)
language_code = self._LANG_MAP.get(lang)
config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id)
config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id, headers={
'x-validated-age': '18',
})
geoblocking = traverse_obj(config, ('data', 'attributes', 'restriction', 'geoblocking')) or {}
if geoblocking.get('restrictedArea'):
@@ -160,16 +170,16 @@ class ArteTVIE(ArteTVBaseIE):
m = self._VERSION_CODE_RE.match(stream_version_code)
if m:
lang_pref = int(''.join('01'[x] for x in (
m.group('vlang') == langauge_code, # we prefer voice in the requested language
m.group('vlang') == language_code, # we prefer voice in the requested language
not m.group('audio_desc'), # and not the audio description version
bool(m.group('original_voice')), # but if voice is not in the requested language, at least choose the original voice
m.group('sub_lang') == langauge_code, # if subtitles are present, we prefer them in the requested language
m.group('sub_lang') == language_code, # if subtitles are present, we prefer them in the requested language
not m.group('has_sub'), # but we prefer no subtitles otherwise
not m.group('sdh_sub'), # and we prefer not the hard-of-hearing subtitles if there are subtitles
)))
short_label = traverse_obj(stream_version, 'shortLabel', expected_type=str, default='?')
if stream['protocol'].startswith('HLS'):
if 'HLS' in stream['protocol']:
fmts, subs = self._extract_m3u8_formats_and_subtitles(
stream['url'], video_id=video_id, ext='mp4', m3u8_id=stream_version_code, fatal=False)
for fmt in fmts:
@@ -181,6 +191,7 @@ class ArteTVIE(ArteTVBaseIE):
secondary_formats.extend(fmts)
else:
formats.extend(fmts)
subs = self._fix_accessible_subs_locale(subs)
self._merge_subtitles(subs, target=subtitles)
elif stream['protocol'] in ('HTTPS', 'RTMP'):
@@ -236,7 +247,7 @@ class ArteTVEmbedIE(InfoExtractor):
'description': 'md5:be40b667f45189632b78c1425c7c2ce1',
'upload_date': '20201116',
},
'skip': 'No video available'
'skip': 'No video available',
}, {
'url': 'https://www.arte.tv/player/v3/index.php?json_url=https://api.arte.tv/api/player/v2/config/de/100605-013-A',
'only_matching': True,
@@ -251,7 +262,7 @@ class ArteTVEmbedIE(InfoExtractor):
class ArteTVPlaylistIE(ArteTVBaseIE):
_VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>RC-\d{6})' % ArteTVBaseIE._ARTE_LANGUAGES
_VALID_URL = rf'https?://(?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos/(?P<id>RC-\d{{6}})'
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/RC-016954/earn-a-living/',
'only_matching': True,
@@ -287,7 +298,7 @@ class ArteTVPlaylistIE(ArteTVBaseIE):
class ArteTVCategoryIE(ArteTVBaseIE):
_VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>[\w-]+(?:/[\w-]+)*)/?\s*$' % ArteTVBaseIE._ARTE_LANGUAGES
_VALID_URL = rf'https?://(?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos/(?P<id>[\w-]+(?:/[\w-]+)*)/?\s*$'
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/politics-and-society/',
'info_dict': {
@@ -301,7 +312,7 @@ class ArteTVCategoryIE(ArteTVBaseIE):
@classmethod
def suitable(cls, url):
return (
not any(ie.suitable(url) for ie in (ArteTVIE, ArteTVPlaylistIE, ))
not any(ie.suitable(url) for ie in (ArteTVIE, ArteTVPlaylistIE))
and super().suitable(url))
def _real_extract(self, url):
@@ -310,12 +321,12 @@ class ArteTVCategoryIE(ArteTVBaseIE):
items = []
for video in re.finditer(
r'<a\b[^>]*?href\s*=\s*(?P<q>"|\'|\b)(?P<url>https?://www\.arte\.tv/%s/videos/[\w/-]+)(?P=q)' % lang,
rf'<a\b[^>]*?href\s*=\s*(?P<q>"|\'|\b)(?P<url>https?://www\.arte\.tv/{lang}/videos/[\w/-]+)(?P=q)',
webpage):
video = video.group('url')
if video == url:
continue
if any(ie.suitable(video) for ie in (ArteTVIE, ArteTVPlaylistIE, )):
if any(ie.suitable(video) for ie in (ArteTVIE, ArteTVPlaylistIE)):
items.append(video)
title = strip_or_none(self._generic_title('', webpage, default='').rsplit('|', 1)[0]) or None
yt_dlp/extractor/asiancrush.py
-196
View File
@@ -1,196 +0,0 @@
import functools
import re
from .common import InfoExtractor
from .kaltura import KalturaIE
from ..utils import (
extract_attributes,
int_or_none,
OnDemandPagedList,
parse_age_limit,
strip_or_none,
try_get,
)
class AsianCrushBaseIE(InfoExtractor):
_VALID_URL_BASE = r'https?://(?:www\.)?(?P<host>(?:(?:asiancrush|yuyutv|midnightpulp)\.com|(?:cocoro|retrocrush)\.tv))'
_KALTURA_KEYS = [
'video_url', 'progressive_url', 'download_url', 'thumbnail_url',
'widescreen_thumbnail_url', 'screencap_widescreen',
]
_API_SUFFIX = {'retrocrush.tv': '-ott'}
def _call_api(self, host, endpoint, video_id, query, resource):
return self._download_json(
'https://api%s.%s/%s' % (self._API_SUFFIX.get(host, ''), host, endpoint), video_id,
'Downloading %s JSON metadata' % resource, query=query,
headers=self.geo_verification_headers())['objects']
def _download_object_data(self, host, object_id, resource):
return self._call_api(
host, 'search', object_id, {'id': object_id}, resource)[0]
def _get_object_description(self, obj):
return strip_or_none(obj.get('long_description') or obj.get('short_description'))
def _parse_video_data(self, video):
title = video['name']
entry_id, partner_id = [None] * 2
for k in self._KALTURA_KEYS:
k_url = video.get(k)
if k_url:
mobj = re.search(r'/p/(\d+)/.+?/entryId/([^/]+)/', k_url)
if mobj:
partner_id, entry_id = mobj.groups()
break
meta_categories = try_get(video, lambda x: x['meta']['categories'], list) or []
categories = list(filter(None, [c.get('name') for c in meta_categories]))
show_info = video.get('show_info') or {}
return {
'_type': 'url_transparent',
'url': 'kaltura:%s:%s' % (partner_id, entry_id),
'ie_key': KalturaIE.ie_key(),
'id': entry_id,
'title': title,
'description': self._get_object_description(video),
'age_limit': parse_age_limit(video.get('mpaa_rating') or video.get('tv_rating')),
'categories': categories,
'series': show_info.get('show_name'),
'season_number': int_or_none(show_info.get('season_num')),
'season_id': show_info.get('season_id'),
'episode_number': int_or_none(show_info.get('episode_num')),
}
class AsianCrushIE(AsianCrushBaseIE):
_VALID_URL = r'%s/video/(?:[^/]+/)?0+(?P<id>\d+)v\b' % AsianCrushBaseIE._VALID_URL_BASE
_TESTS = [{
'url': 'https://www.asiancrush.com/video/004289v/women-who-flirt',
'md5': 'c3b740e48d0ba002a42c0b72857beae6',
'info_dict': {
'id': '1_y4tmjm5r',
'ext': 'mp4',
'title': 'Women Who Flirt',
'description': 'md5:b65c7e0ae03a85585476a62a186f924c',
'timestamp': 1496936429,
'upload_date': '20170608',
'uploader_id': 'craig@crifkin.com',
'age_limit': 13,
'categories': 'count:5',
'duration': 5812,
},
}, {
'url': 'https://www.asiancrush.com/video/she-was-pretty/011886v-pretty-episode-3/',
'only_matching': True,
}, {
'url': 'https://www.yuyutv.com/video/013886v/the-act-of-killing/',
'only_matching': True,
}, {
'url': 'https://www.yuyutv.com/video/peep-show/013922v-warring-factions/',
'only_matching': True,
}, {
'url': 'https://www.midnightpulp.com/video/010400v/drifters/',
'only_matching': True,
}, {
'url': 'https://www.midnightpulp.com/video/mononoke/016378v-zashikiwarashi-part-1/',
'only_matching': True,
}, {
'url': 'https://www.cocoro.tv/video/the-wonderful-wizard-of-oz/008878v-the-wonderful-wizard-of-oz-ep01/',
'only_matching': True,
}, {
'url': 'https://www.retrocrush.tv/video/true-tears/012328v-i...gave-away-my-tears',
'only_matching': True,
}]
def _real_extract(self, url):
host, video_id = self._match_valid_url(url).groups()
if host == 'cocoro.tv':
webpage = self._download_webpage(url, video_id)
embed_vars = self._parse_json(self._search_regex(
r'iEmbedVars\s*=\s*({.+?})', webpage, 'embed vars',
default='{}'), video_id, fatal=False) or {}
video_id = embed_vars.get('entry_id') or video_id
video = self._download_object_data(host, video_id, 'video')
return self._parse_video_data(video)
class AsianCrushPlaylistIE(AsianCrushBaseIE):
_VALID_URL = r'%s/series/0+(?P<id>\d+)s\b' % AsianCrushBaseIE._VALID_URL_BASE
_TESTS = [{
'url': 'https://www.asiancrush.com/series/006447s/fruity-samurai',
'info_dict': {
'id': '6447',
'title': 'Fruity Samurai',
'description': 'md5:7535174487e4a202d3872a7fc8f2f154',
},
'playlist_count': 13,
}, {
'url': 'https://www.yuyutv.com/series/013920s/peep-show/',
'only_matching': True,
}, {
'url': 'https://www.midnightpulp.com/series/016375s/mononoke/',
'only_matching': True,
}, {
'url': 'https://www.cocoro.tv/series/008549s/the-wonderful-wizard-of-oz/',
'only_matching': True,
}, {
'url': 'https://www.retrocrush.tv/series/012355s/true-tears',
'only_matching': True,
}]
_PAGE_SIZE = 1000000000
def _fetch_page(self, domain, parent_id, page):
videos = self._call_api(
domain, 'getreferencedobjects', parent_id, {
'max': self._PAGE_SIZE,
'object_type': 'video',
'parent_id': parent_id,
'start': page * self._PAGE_SIZE,
}, 'page %d' % (page + 1))
for video in videos:
yield self._parse_video_data(video)
def _real_extract(self, url):
host, playlist_id = self._match_valid_url(url).groups()
if host == 'cocoro.tv':
webpage = self._download_webpage(url, playlist_id)
entries = []
for mobj in re.finditer(
r'<a[^>]+href=(["\'])(?P<url>%s.*?)\1[^>]*>' % AsianCrushIE._VALID_URL,
webpage):
attrs = extract_attributes(mobj.group(0))
if attrs.get('class') == 'clearfix':
entries.append(self.url_result(
mobj.group('url'), ie=AsianCrushIE.ie_key()))
title = self._html_search_regex(
r'(?s)<h1\b[^>]\bid=["\']movieTitle[^>]+>(.+?)</h1>', webpage,
'title', default=None) or self._og_search_title(
webpage, default=None) or self._html_search_meta(
'twitter:title', webpage, 'title',
default=None) or self._html_extract_title(webpage)
if title:
title = re.sub(r'\s*\|\s*.+?$', '', title)
description = self._og_search_description(
webpage, default=None) or self._html_search_meta(
'twitter:description', webpage, 'description', fatal=False)
else:
show = self._download_object_data(host, playlist_id, 'show')
title = show.get('name')
description = self._get_object_description(show)
entries = OnDemandPagedList(
functools.partial(self._fetch_page, host, playlist_id),
self._PAGE_SIZE)
return self.playlist_result(entries, playlist_id, title, description)
yt_dlp/extractor/asobichannel.py
+168
View File
@@ -0,0 +1,168 @@
from .common import InfoExtractor
from ..utils import (
ExtractorError,
clean_html,
merge_dicts,
parse_iso8601,
url_or_none,
)
from ..utils.traversal import traverse_obj
class AsobiChannelBaseIE(InfoExtractor):
_MICROCMS_HEADER = {'X-MICROCMS-API-KEY': 'qRaKehul9AHU8KtL0dnq1OCLKnFec6yrbcz3'}
def _extract_info(self, metadata):
return traverse_obj(metadata, {
'id': ('id', {str}),
'title': ('title', {str}),
'description': ('body', {clean_html}),
'thumbnail': ('contents', 'video_thumb', 'url', {url_or_none}),
'timestamp': ('publishedAt', {parse_iso8601}),
'modified_timestamp': ('updatedAt', {parse_iso8601}),
'channel': ('channel', 'name', {str}),
'channel_id': ('channel', 'id', {str}),
})
class AsobiChannelIE(AsobiChannelBaseIE):
IE_NAME = 'asobichannel'
IE_DESC = 'ASOBI CHANNEL'
_VALID_URL = r'https?://asobichannel\.asobistore\.jp/watch/(?P<id>[\w-]+)'
_TESTS = [{
'url': 'https://asobichannel.asobistore.jp/watch/1ypp48qd32p',
'md5': '39df74e872afe032c4eb27b89144fc92',
'info_dict': {
'id': '1ypp48qd32p',
'ext': 'mp4',
'title': 'アイドルマスター ミリオンライブ！ 765プロch 原っぱ通信 #1',
'description': 'md5:b930bd2199c9b2fd75951ce4aaa7efd2',
'thumbnail': 'https://images.microcms-assets.io/assets/d2420de4b9194e11beb164f99edb1f95/a8e6f84119f54eb9ab4ce16729239905/%E3%82%B5%E3%83%A0%E3%83%8D%20(1).png',
'timestamp': 1697098247,
'upload_date': '20231012',
'modified_timestamp': 1698381162,
'modified_date': '20231027',
'channel': 'アイドルマスター',
'channel_id': 'idolmaster',
},
}, {
'url': 'https://asobichannel.asobistore.jp/watch/redigiwnjzqj',
'md5': '229fa8fb5c591c75ce8c37a497f113f6',
'info_dict': {
'id': 'redigiwnjzqj',
'ext': 'mp4',
'title': '【おまけ放送】アイドルマスター ミリオンライブ！ 765プロch 原っぱ通信 #1',
'description': 'md5:7d9cd35fb54425a6967822bd564ea2d9',
'thumbnail': 'https://images.microcms-assets.io/assets/d2420de4b9194e11beb164f99edb1f95/20e5c1d6184242eebc2512a5dec59bf0/P1_%E5%8E%9F%E3%81%A3%E3%81%B1%E3%82%B5%E3%83%A0%E3%83%8D.png',
'modified_timestamp': 1697797125,
'modified_date': '20231020',
'timestamp': 1697261769,
'upload_date': '20231014',
'channel': 'アイドルマスター',
'channel_id': 'idolmaster',
},
}]
_survapi_header = None
def _real_initialize(self):
token = self._download_json(
'https://asobichannel-api.asobistore.jp/api/v1/vspf/token', None,
note='Retrieving API token')
self._survapi_header = {'Authorization': f'Bearer {token}'}
def _process_vod(self, video_id, metadata):
content_id = metadata['contents']['video_id']
vod_data = self._download_json(
f'https://survapi.channel.or.jp/proxy/v1/contents/{content_id}/get_by_cuid', video_id,
headers=self._survapi_header, note='Downloading vod data')
return {
'formats': self._extract_m3u8_formats(vod_data['ex_content']['streaming_url'], video_id),
}
def _process_live(self, video_id, metadata):
content_id = metadata['contents']['video_id']
event_data = self._download_json(
f'https://survapi.channel.or.jp/ex/events/{content_id}?embed=channel', video_id,
headers=self._survapi_header, note='Downloading event data')
player_type = traverse_obj(event_data, ('data', 'Player_type', {str}))
if player_type == 'poster':
self.raise_no_formats('Live event has not yet started', expected=True)
live_status = 'is_upcoming'
formats = []
elif player_type == 'player':
live_status = 'is_live'
formats = self._extract_m3u8_formats(
event_data['data']['Channel']['Custom_live_url'], video_id, live=True)
else:
raise ExtractorError('Unsupported player type {player_type!r}')
return {
'release_timestamp': traverse_obj(metadata, ('period', 'start', {parse_iso8601})),
'live_status': live_status,
'formats': formats,
}
def _real_extract(self, url):
video_id = self._match_id(url)
metadata = self._download_json(
f'https://channel.microcms.io/api/v1/media/{video_id}', video_id,
headers=self._MICROCMS_HEADER)
info = self._extract_info(metadata)
video_type = traverse_obj(metadata, ('contents', 'video_type', 0, {str}))
if video_type == 'VOD':
return merge_dicts(info, self._process_vod(video_id, metadata))
if video_type == 'LIVE':
return merge_dicts(info, self._process_live(video_id, metadata))
raise ExtractorError(f'Unexpected video type {video_type!r}')
class AsobiChannelTagURLIE(AsobiChannelBaseIE):
IE_NAME = 'asobichannel:tag'
IE_DESC = 'ASOBI CHANNEL'
_VALID_URL = r'https?://asobichannel\.asobistore\.jp/tag/(?P<id>[a-z0-9-_]+)'
_TESTS = [{
'url': 'https://asobichannel.asobistore.jp/tag/bjhh-nbcja',
'info_dict': {
'id': 'bjhh-nbcja',
'title': 'アイドルマスター ミリオンライブ！ 765プロch 原っぱ通信',
},
'playlist_mincount': 16,
}, {
'url': 'https://asobichannel.asobistore.jp/tag/hvm5qw3c6od',
'info_dict': {
'id': 'hvm5qw3c6od',
'title': 'アイマスMOIW2023ラジオ',
},
'playlist_mincount': 13,
}]
def _real_extract(self, url):
tag_id = self._match_id(url)
webpage = self._download_webpage(url, tag_id)
title = traverse_obj(self._search_nextjs_data(
webpage, tag_id, fatal=False), ('props', 'pageProps', 'data', 'name', {str}))
media = self._download_json(
f'https://channel.microcms.io/api/v1/media?limit=999&filters=(tag[contains]{tag_id})',
tag_id, headers=self._MICROCMS_HEADER)
def entries():
for metadata in traverse_obj(media, ('contents', lambda _, v: v['id'])):
yield {
'_type': 'url',
'url': f'https://asobichannel.asobistore.jp/watch/{metadata["id"]}',
'ie_key': AsobiChannelIE.ie_key(),
**self._extract_info(metadata),
}
return self.playlist_result(entries(), tag_id, title)
yt_dlp/extractor/asobistage.py
+154
View File
@@ -0,0 +1,154 @@
import functools
from .common import InfoExtractor
from ..utils import str_or_none, url_or_none
from ..utils.traversal import traverse_obj
class AsobiStageIE(InfoExtractor):
IE_DESC = 'ASOBISTAGE (アソビステージ)'
_VALID_URL = r'https?://asobistage\.asobistore\.jp/event/(?P<id>(?P<event>\w+)/(?P<type>archive|player)/(?P<slug>\w+))(?:[?#]|$)'
_TESTS = [{
'url': 'https://asobistage.asobistore.jp/event/315passionhour_2022summer/archive/frame',
'info_dict': {
'id': '315passionhour_2022summer/archive/frame',
'title': '315プロダクションプレゼンツ 315パッションアワー!!!',
'thumbnail': r're:^https?://[\w.-]+/\w+/\w+',
},
'playlist_count': 1,
'playlist': [{
'info_dict': {
'id': 'edff52f2',
'ext': 'mp4',
'title': '315passion_FRAME_only',
'thumbnail': r're:^https?://[\w.-]+/\w+/\w+',
},
}],
}, {
'url': 'https://asobistage.asobistore.jp/event/idolmaster_idolworld2023_goods/archive/live',
'info_dict': {
'id': 'idolmaster_idolworld2023_goods/archive/live',
'title': 'md5:378510b6e830129d505885908bd6c576',
'thumbnail': r're:^https?://[\w.-]+/\w+/\w+',
},
'playlist_count': 1,
'playlist': [{
'info_dict': {
'id': '3aef7110',
'ext': 'mp4',
'title': 'asobistore_station_1020_serverREC',
'thumbnail': r're:^https?://[\w.-]+/\w+/\w+',
},
}],
}, {
'url': 'https://asobistage.asobistore.jp/event/sidem_fclive_bpct/archive/premium_hc',
'playlist_count': 4,
'info_dict': {
'id': 'sidem_fclive_bpct/archive/premium_hc',
'title': '315 Production presents F＠NTASTIC COMBINATION LIVE ～BRAINPOWER!!～/～CONNECTIME!!!!～',
'thumbnail': r're:^https?://[\w.-]+/\w+/\w+',
},
}, {
'url': 'https://asobistage.asobistore.jp/event/ijigenfes_utagassen/player/day1',
'only_matching': True,
}]
_API_HOST = 'https://asobistage-api.asobistore.jp'
_HEADERS = {}
_is_logged_in = False
@functools.cached_property
def _owned_tickets(self):
owned_tickets = set()
if not self._is_logged_in:
return owned_tickets
for path, name in [
('api/v1/purchase_history/list', 'ticket purchase history'),
('api/v1/serialcode/list', 'redemption history'),
]:
response = self._download_json(
f'{self._API_HOST}/{path}', None, f'Downloading {name}',
f'Unable to download {name}', expected_status=400)
if traverse_obj(response, ('payload', 'error_message'), 'error') == 'notlogin':
self._is_logged_in = False
break
owned_tickets.update(
traverse_obj(response, ('payload', 'value', ..., 'digital_product_id', {str_or_none})))
return owned_tickets
def _get_available_channel_id(self, channel):
channel_id = traverse_obj(channel, ('chennel_vspf_id', {str}))
if not channel_id:
return None
# if rights_type_id == 6, then 'No conditions (no login required - non-members are OK)'
if traverse_obj(channel, ('viewrights', lambda _, v: v['rights_type_id'] == 6)):
return channel_id
available_tickets = traverse_obj(channel, (
'viewrights', ..., ('tickets', 'serialcodes'), ..., 'digital_product_id', {str_or_none}))
if not self._owned_tickets.intersection(available_tickets):
self.report_warning(
f'You are not a ticketholder for "{channel.get("channel_name") or channel_id}"')
return None
return channel_id
def _real_initialize(self):
if self._get_cookies(self._API_HOST):
self._is_logged_in = True
token = self._download_json(
f'{self._API_HOST}/api/v1/vspf/token', None, 'Getting token', 'Unable to get token')
self._HEADERS['Authorization'] = f'Bearer {token}'
def _real_extract(self, url):
video_id, event, type_, slug = self._match_valid_url(url).group('id', 'event', 'type', 'slug')
video_type = {'archive': 'archives', 'player': 'broadcasts'}[type_]
webpage = self._download_webpage(url, video_id)
event_data = traverse_obj(
self._search_nextjs_data(webpage, video_id, default={}),
('props', 'pageProps', 'eventCMSData', {
'title': ('event_name', {str}),
'thumbnail': ('event_thumbnail_image', {url_or_none}),
}))
available_channels = traverse_obj(self._download_json(
f'https://asobistage.asobistore.jp/cdn/v101/events/{event}/{video_type}.json',
video_id, 'Getting channel list', 'Unable to get channel list'), (
video_type, lambda _, v: v['broadcast_slug'] == slug,
'channels', lambda _, v: v['chennel_vspf_id'] != '00000'))
entries = []
for channel_id in traverse_obj(available_channels, (..., {self._get_available_channel_id})):
if video_type == 'archives':
channel_json = self._download_json(
f'https://survapi.channel.or.jp/proxy/v1/contents/{channel_id}/get_by_cuid', channel_id,
'Getting archive channel info', 'Unable to get archive channel info', fatal=False,
headers=self._HEADERS)
channel_data = traverse_obj(channel_json, ('ex_content', {
'm3u8_url': 'streaming_url',
'title': 'title',
'thumbnail': ('thumbnail', 'url'),
}))
else: # video_type == 'broadcasts'
channel_json = self._download_json(
f'https://survapi.channel.or.jp/ex/events/{channel_id}', channel_id,
'Getting live channel info', 'Unable to get live channel info', fatal=False,
headers=self._HEADERS, query={'embed': 'channel'})
channel_data = traverse_obj(channel_json, ('data', {
'm3u8_url': ('Channel', 'Custom_live_url'),
'title': 'Name',
'thumbnail': 'Poster_url',
}))
entries.append({
'id': channel_id,
'title': channel_data.get('title'),
'formats': self._extract_m3u8_formats(channel_data.get('m3u8_url'), channel_id, fatal=False),
'is_live': video_type == 'broadcasts',
'thumbnail': url_or_none(channel_data.get('thumbnail')),
})
if not self._is_logged_in and not entries:
self.raise_login_required()
return self.playlist_result(entries, video_id, **event_data)
yt_dlp/extractor/atresplayer.py
+12 -13
View File
@@ -1,5 +1,5 @@
from .common import InfoExtractor
from ..compat import compat_HTTPError
from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
int_or_none,
@@ -20,7 +20,7 @@ class AtresPlayerIE(InfoExtractor):
'description': 'md5:7634cdcb4d50d5381bedf93efb537fbc',
'duration': 3413,
},
'skip': 'This video is only available for registered users'
'skip': 'This video is only available for registered users',
},
{
'url': 'https://www.atresplayer.com/lasexta/programas/el-club-de-la-comedia/temporada-4/capitulo-10-especial-solidario-nochebuena_5ad08edf986b2855ed47adc4/',
@@ -33,14 +33,6 @@ class AtresPlayerIE(InfoExtractor):
]
_API_BASE = 'https://api.atresplayer.com/'
def _handle_error(self, e, code):
if isinstance(e.cause, compat_HTTPError) and e.cause.code == code:
error = self._parse_json(e.cause.read(), None)
if error.get('error') == 'required_registered':
self.raise_login_required()
raise ExtractorError(error['error_description'], expected=True)
raise
def _perform_login(self, username, password):
self._request_webpage(
self._API_BASE + 'login', None, 'Downloading login page')
@@ -49,13 +41,15 @@ class AtresPlayerIE(InfoExtractor):
target_url = self._download_json(
'https://account.atresmedia.com/api/login', None,
'Logging in', headers={
'Content-Type': 'application/x-www-form-urlencoded'
'Content-Type': 'application/x-www-form-urlencoded',
}, data=urlencode_postdata({
'username': username,
'password': password,
}))['targetUrl']
except ExtractorError as e:
self._handle_error(e, 400)
if isinstance(e.cause, HTTPError) and e.cause.status == 400:
raise ExtractorError('Invalid username and/or password', expected=True)
raise
self._request_webpage(target_url, None, 'Following Target URL')
@@ -66,7 +60,12 @@ class AtresPlayerIE(InfoExtractor):
episode = self._download_json(
self._API_BASE + 'client/v1/player/episode/' + video_id, video_id)
except ExtractorError as e:
self._handle_error(e, 403)
if isinstance(e.cause, HTTPError) and e.cause.status == 403:
error = self._parse_json(e.cause.response.read(), None)
if error.get('error') == 'required_registered':
self.raise_login_required()
raise ExtractorError(error['error_description'], expected=True)
raise
title = episode['titulo']
yt_dlp/extractor/atscaleconf.py
+5 -5
View File
@@ -12,7 +12,7 @@ class AtScaleConfEventIE(InfoExtractor):
'info_dict': {
'id': 'data-scale-spring-2022',
'title': 'Data @Scale Spring 2022',
'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55'
'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55',
},
}, {
'url': 'https://atscaleconference.com/events/video-scale-2021/',
@@ -20,15 +20,15 @@ class AtScaleConfEventIE(InfoExtractor):
'info_dict': {
'id': 'video-scale-2021',
'title': 'Video @Scale 2021',
'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55'
'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55',
},
}]
def _real_extract(self, url):
id = self._match_id(url)
webpage = self._download_webpage(url, id)
playlist_id = self._match_id(url)
webpage = self._download_webpage(url, playlist_id)
return self.playlist_from_matches(
re.findall(r'data-url\s*=\s*"(https?://(?:www\.)?atscaleconference\.com/videos/[^"]+)"', webpage),
ie='Generic', playlist_id=id,
ie='Generic', playlist_id=playlist_id,
title=self._og_search_title(webpage), description=self._og_search_description(webpage))
yt_dlp/extractor/atttechchannel.py
-53
View File
@@ -1,53 +0,0 @@
from .common import InfoExtractor
from ..utils import unified_strdate
class ATTTechChannelIE(InfoExtractor):
_VALID_URL = r'https?://techchannel\.att\.com/play-video\.cfm/([^/]+/)*(?P<id>.+)'
_TEST = {
'url': 'http://techchannel.att.com/play-video.cfm/2014/1/27/ATT-Archives-The-UNIX-System-Making-Computers-Easier-to-Use',
'info_dict': {
'id': '11316',
'display_id': 'ATT-Archives-The-UNIX-System-Making-Computers-Easier-to-Use',
'ext': 'flv',
'title': 'AT&T Archives : The UNIX System: Making Computers Easier to Use',
'description': 'A 1982 film about UNIX is the foundation for software in use around Bell Labs and AT&T.',
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20140127',
},
'params': {
# rtmp download
'skip_download': True,
},
}
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
video_url = self._search_regex(
r"url\s*:\s*'(rtmp://[^']+)'",
webpage, 'video URL')
video_id = self._search_regex(
r'mediaid\s*=\s*(\d+)',
webpage, 'video id', fatal=False)
title = self._og_search_title(webpage)
description = self._og_search_description(webpage)
thumbnail = self._og_search_thumbnail(webpage)
upload_date = unified_strdate(self._search_regex(
r'[Rr]elease\s+date:\s*(\d{1,2}/\d{1,2}/\d{4})',
webpage, 'upload date', fatal=False), False)
return {
'id': video_id,
'display_id': display_id,
'url': video_url,
'ext': 'flv',
'title': title,
'description': description,
'thumbnail': thumbnail,
'upload_date': upload_date,
}
yt_dlp/extractor/atvat.py
+13 -13
View File
@@ -1,11 +1,11 @@
import datetime
import datetime as dt
from .common import InfoExtractor
from ..utils import (
ExtractorError,
float_or_none,
jwt_encode_hs256,
try_get,
ExtractorError,
)
@@ -19,7 +19,7 @@ class ATVAtIE(InfoExtractor):
'id': 'v-ce9cgn1e70n5-1',
'ext': 'mp4',
'title': 'Bauer sucht Frau - Staffel 18 Folge 3 - Die Hofwochen',
}
},
}, {
'url': 'https://www.atv.at/tv/bauer-sucht-frau/staffel-18/episode-01/bauer-sucht-frau-staffel-18-vorstellungsfolge-1',
'only_matching': True,
@@ -66,14 +66,14 @@ class ATVAtIE(InfoExtractor):
video_id=video_id)
video_title = json_data['views']['default']['page']['title']
contentResource = json_data['views']['default']['page']['contentResource']
content_id = contentResource[0]['id']
content_ids = [{'id': id, 'subclip_start': content['start'], 'subclip_end': content['end']}
for id, content in enumerate(contentResource)]
content_resource = json_data['views']['default']['page']['contentResource']
content_id = content_resource[0]['id']
content_ids = [{'id': id_, 'subclip_start': content['start'], 'subclip_end': content['end']}
for id_, content in enumerate(content_resource)]
time_of_request = datetime.datetime.now()
not_before = time_of_request - datetime.timedelta(minutes=5)
expire = time_of_request + datetime.timedelta(minutes=5)
time_of_request = dt.datetime.now()
not_before = time_of_request - dt.timedelta(minutes=5)
expire = time_of_request + dt.timedelta(minutes=5)
payload = {
'content_ids': {
content_id: content_ids,
@@ -87,17 +87,17 @@ class ATVAtIE(InfoExtractor):
videos = self._download_json(
'https://vas-v4.p7s1video.net/4.0/getsources',
content_id, 'Downloading videos JSON', query={
'token': jwt_token.decode('utf-8')
'token': jwt_token.decode('utf-8'),
})
video_id, videos_data = list(videos['data'].items())[0]
video_id, videos_data = next(iter(videos['data'].items()))
error_msg = try_get(videos_data, lambda x: x['error']['title'])
if error_msg == 'Geo check failed':
self.raise_geo_restricted(error_msg)
elif error_msg:
raise ExtractorError(error_msg)
entries = [
self._extract_video_info(url, contentResource[video['id']], video)
self._extract_video_info(url, content_resource[video['id']], video)
for video in videos_data]
return {
yt_dlp/extractor/audimedia.py
+2 -2
View File
@@ -19,7 +19,7 @@ class AudiMediaIE(InfoExtractor):
'timestamp': 1448354940,
'duration': 74022,
'view_count': int,
}
},
}, {
'url': 'https://www.audi-mediacenter.com/en/audimediatv/video/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test-2991',
'only_matching': True,
@@ -73,7 +73,7 @@ class AudiMediaIE(InfoExtractor):
bitrate = self._search_regex(r'(\d+)k', video_version_url, 'bitrate', default=None)
if bitrate:
f.update({
'format_id': 'http-%s' % bitrate,
'format_id': f'http-{bitrate}',
})
formats.append(f)
yt_dlp/extractor/audioboom.py
+2 -2
View File
@@ -15,7 +15,7 @@ class AudioBoomIE(InfoExtractor):
'duration': 4000.99,
'uploader': 'Sue Perkins: An hour or so with...',
'uploader_url': r're:https?://(?:www\.)?audioboom\.com/channel/perkins',
}
},
}, { # Direct mp3-file link
'url': 'https://audioboom.com/posts/8128496.mp3',
'md5': 'e329edf304d450def95c7f86a9165ee1',
@@ -27,7 +27,7 @@ class AudioBoomIE(InfoExtractor):
'duration': 1689.7,
'uploader': 'Lost Dot Podcast: The Trans Pyrenees and Transcontinental Race',
'uploader_url': r're:https?://(?:www\.)?audioboom\.com/channels/5003904',
}
},
}, {
'url': 'https://audioboom.com/posts/4279833-3-09-2016-czaban-hour-3?t=0',
'only_matching': True,
yt_dlp/extractor/audiodraft.py
+7 -6
View File
@@ -9,7 +9,7 @@ class AudiodraftBaseIE(InfoExtractor):
headers={
'Content-type': 'application/x-www-form-urlencoded; charset=UTF-8',
'X-Requested-With': 'XMLHttpRequest',
}, data=f'id={player_entry_id}'.encode('utf-8'))
}, data=f'id={player_entry_id}'.encode())
return {
'id': str(data_json['entry_id']),
@@ -65,9 +65,10 @@ class AudiodraftCustomIE(AudiodraftBaseIE):
}]
def _real_extract(self, url):
id = self._match_id(url)
webpage = self._download_webpage(url, id)
player_entry_id = self._search_regex(r'playAudio\(\'(player_entry_\d+)\'\);', webpage, id, 'play entry id')
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
player_entry_id = self._search_regex(
r'playAudio\(\'(player_entry_\d+)\'\);', webpage, video_id, 'play entry id')
return self._audiodraft_extract_from_id(player_entry_id)
@@ -89,5 +90,5 @@ class AudiodraftGenericIE(AudiodraftBaseIE):
}]
def _real_extract(self, url):
id = self._match_id(url)
return self._audiodraft_extract_from_id(f'player_entry_{id}')
video_id = self._match_id(url)
return self._audiodraft_extract_from_id(f'player_entry_{video_id}')
yt_dlp/extractor/audiomack.py
+13 -14
View File
@@ -3,7 +3,6 @@ import time
from .common import InfoExtractor
from .soundcloud import SoundcloudIE
from ..compat import compat_str
from ..utils import (
ExtractorError,
url_basename,
@@ -22,8 +21,8 @@ class AudiomackIE(InfoExtractor):
'id': '310086',
'ext': 'mp3',
'uploader': 'Roosh Williams',
'title': 'Extraordinary'
}
'title': 'Extraordinary',
},
},
# audiomack wrapper around soundcloud song
# Needs new test URL.
@@ -56,7 +55,7 @@ class AudiomackIE(InfoExtractor):
# API is inconsistent with errors
if 'url' not in api_response or not api_response['url'] or 'error' in api_response:
raise ExtractorError('Invalid url %s' % url)
raise ExtractorError(f'Invalid url {url}')
# Audiomack wraps a lot of soundcloud tracks in their branded wrapper
# if so, pass the work off to the soundcloud extractor
@@ -64,7 +63,7 @@ class AudiomackIE(InfoExtractor):
return self.url_result(api_response['url'], SoundcloudIE.ie_key())
return {
'id': compat_str(api_response.get('id', album_url_tag)),
'id': str(api_response.get('id', album_url_tag)),
'uploader': api_response.get('artist'),
'title': api_response.get('title'),
'url': api_response['url'],
@@ -82,8 +81,8 @@ class AudiomackAlbumIE(InfoExtractor):
'info_dict':
{
'id': '812251',
'title': 'Tha Tour: Part 2 (Official Mixtape)'
}
'title': 'Tha Tour: Part 2 (Official Mixtape)',
},
},
# Album playlist ripped from fakeshoredrive with no metadata
{
@@ -98,16 +97,16 @@ class AudiomackAlbumIE(InfoExtractor):
'id': '837576',
'ext': 'mp3',
'uploader': 'Lil Herb a.k.a. G Herbo',
}
},
}, {
'info_dict': {
'title': 'PPP (Pistol P Project) - 10. 4 Minutes Of Hell Part 4 (prod by DY OF 808 MAFIA)',
'id': '837580',
'ext': 'mp3',
'uploader': 'Lil Herb a.k.a. G Herbo',
}
},
}],
}
},
]
def _real_extract(self, url):
@@ -123,12 +122,12 @@ class AudiomackAlbumIE(InfoExtractor):
api_response = self._download_json(
'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
% (album_url_tag, track_no, time.time()), album_url_tag,
note='Querying song information (%d)' % (track_no + 1))
note=f'Querying song information ({track_no + 1})')
# Total failure, only occurs when url is totally wrong
# Won't happen in middle of valid playlist (next case)
if 'url' not in api_response or 'error' in api_response:
raise ExtractorError('Invalid url for track %d of album url %s' % (track_no, url))
raise ExtractorError(f'Invalid url for track {track_no} of album url {url}')
# URL is good but song id doesn't exist - usually means end of playlist
elif not api_response['url']:
break
@@ -136,10 +135,10 @@ class AudiomackAlbumIE(InfoExtractor):
# Pull out the album metadata and add to result (if it exists)
for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
if apikey in api_response and resultkey not in result:
result[resultkey] = compat_str(api_response[apikey])
result[resultkey] = str(api_response[apikey])
song_id = url_basename(api_response['url']).rpartition('.')[0]
result['entries'].append({
'id': compat_str(api_response.get('id', song_id)),
'id': str(api_response.get('id', song_id)),
'uploader': api_response.get('artist'),
'title': api_response.get('title', song_id),
'url': api_response['url'],
yt_dlp/extractor/audius.py
+23 -23
View File
@@ -1,7 +1,7 @@
import random
import urllib.parse
from .common import InfoExtractor
from ..compat import compat_str, compat_urllib_parse_unquote
from ..utils import ExtractorError, str_or_none, try_get
@@ -15,13 +15,13 @@ class AudiusBaseIE(InfoExtractor):
if response_data is not None:
return response_data
if len(response) == 1 and 'message' in response:
raise ExtractorError('API error: %s' % response['message'],
raise ExtractorError('API error: {}'.format(response['message']),
expected=True)
raise ExtractorError('Unexpected API response')
def _select_api_base(self):
"""Selecting one of the currently available API hosts"""
response = super(AudiusBaseIE, self)._download_json(
response = super()._download_json(
'https://api.audius.co/', None,
note='Requesting available API hosts',
errnote='Unable to request available API hosts')
@@ -41,8 +41,8 @@ class AudiusBaseIE(InfoExtractor):
anything from this link, since the Audius API won't be able to resolve
this url
"""
url = compat_urllib_parse_unquote(url)
title = compat_urllib_parse_unquote(title)
url = urllib.parse.unquote(url)
title = urllib.parse.unquote(title)
if '/' in title or '%2F' in title:
fixed_title = title.replace('/', '%5C').replace('%2F', '%5C')
return url.replace(title, fixed_title)
@@ -54,19 +54,19 @@ class AudiusBaseIE(InfoExtractor):
if self._API_BASE is None:
self._select_api_base()
try:
response = super(AudiusBaseIE, self)._download_json(
'%s%s%s' % (self._API_BASE, self._API_V, path), item_id, note=note,
response = super()._download_json(
f'{self._API_BASE}{self._API_V}{path}', item_id, note=note,
errnote=errnote, expected_status=expected_status)
except ExtractorError as exc:
# some of Audius API hosts may not work as expected and return HTML
if 'Failed to parse JSON' in compat_str(exc):
if 'Failed to parse JSON' in str(exc):
raise ExtractorError('An error occurred while receiving data. Try again',
expected=True)
raise exc
return self._get_response_data(response)
def _resolve_url(self, url, item_id):
return self._api_request('/resolve?url=%s' % url, item_id,
return self._api_request(f'/resolve?url={url}', item_id,
expected_status=404)
@@ -91,7 +91,7 @@ class AudiusIE(AudiusBaseIE):
'view_count': int,
'like_count': int,
'repost_count': int,
}
},
},
{
# Regular track
@@ -109,14 +109,14 @@ class AudiusIE(AudiusBaseIE):
'view_count': int,
'like_count': int,
'repost_count': int,
}
},
},
]
_ARTWORK_MAP = {
"150x150": 150,
"480x480": 480,
"1000x1000": 1000
'150x150': 150,
'480x480': 480,
'1000x1000': 1000,
}
def _real_extract(self, url):
@@ -130,7 +130,7 @@ class AudiusIE(AudiusBaseIE):
else: # API link
title = None
# uploader = None
track_data = self._api_request('/tracks/%s' % track_id, track_id)
track_data = self._api_request(f'/tracks/{track_id}', track_id)
if not isinstance(track_data, dict):
raise ExtractorError('Unexpected API response')
@@ -144,7 +144,7 @@ class AudiusIE(AudiusBaseIE):
if isinstance(artworks_data, dict):
for quality_key, thumbnail_url in artworks_data.items():
thumbnail = {
"url": thumbnail_url
'url': thumbnail_url,
}
quality_code = self._ARTWORK_MAP.get(quality_key)
if quality_code is not None:
@@ -154,12 +154,12 @@ class AudiusIE(AudiusBaseIE):
return {
'id': track_id,
'title': track_data.get('title', title),
'url': '%s/v1/tracks/%s/stream' % (self._API_BASE, track_id),
'url': f'{self._API_BASE}/v1/tracks/{track_id}/stream',
'ext': 'mp3',
'description': track_data.get('description'),
'duration': track_data.get('duration'),
'track': track_data.get('title'),
'artist': try_get(track_data, lambda x: x['user']['name'], compat_str),
'artist': try_get(track_data, lambda x: x['user']['name'], str),
'genre': track_data.get('genre'),
'thumbnails': thumbnails,
'view_count': track_data.get('play_count'),
@@ -175,11 +175,11 @@ class AudiusTrackIE(AudiusIE): # XXX: Do not subclass from concrete IE
_TESTS = [
{
'url': 'audius:9RWlo',
'only_matching': True
'only_matching': True,
},
{
'url': 'audius:http://discoveryprovider.audius.prod-us-west-2.staked.cloud/v1/tracks/9RWlo',
'only_matching': True
'only_matching': True,
},
]
@@ -207,7 +207,7 @@ class AudiusPlaylistIE(AudiusBaseIE):
if not track_id:
raise ExtractorError('Unable to get track ID from playlist')
entries.append(self.url_result(
'audius:%s' % track_id,
f'audius:{track_id}',
ie=AudiusTrackIE.ie_key(), video_id=track_id))
return entries
@@ -231,7 +231,7 @@ class AudiusPlaylistIE(AudiusBaseIE):
raise ExtractorError('Unable to get playlist ID')
playlist_tracks = self._api_request(
'/playlists/%s/tracks' % playlist_id,
f'/playlists/{playlist_id}/tracks',
title, note='Downloading playlist tracks metadata',
errnote='Unable to download playlist tracks metadata')
if not isinstance(playlist_tracks, list):
@@ -267,5 +267,5 @@ class AudiusProfileIE(AudiusPlaylistIE): # XXX: Do not subclass from concrete I
profile_audius_id = _profile_data[0]['id']
profile_bio = _profile_data[0].get('bio')
api_call = self._api_request('/full/users/handle/%s/tracks' % profile_id, profile_id)
api_call = self._api_request(f'/full/users/handle/{profile_id}/tracks', profile_id)
return self.playlist_result(self._build_playlist(api_call), profile_audius_id, profile_id, profile_bio)
yt_dlp/extractor/awaan.py
+12 -15
View File
@@ -1,10 +1,7 @@
import base64
import urllib.parse
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlencode,
compat_str,
)
from ..utils import (
format_field,
int_or_none,
@@ -22,14 +19,14 @@ class AWAANIE(InfoExtractor):
show_id, video_id, season_id = self._match_valid_url(url).groups()
if video_id and int(video_id) > 0:
return self.url_result(
'http://awaan.ae/media/%s' % video_id, 'AWAANVideo')
f'http://awaan.ae/media/{video_id}', 'AWAANVideo')
elif season_id and int(season_id) > 0:
return self.url_result(smuggle_url(
'http://awaan.ae/program/season/%s' % season_id,
f'http://awaan.ae/program/season/{season_id}',
{'show_id': show_id}), 'AWAANSeason')
else:
return self.url_result(
'http://awaan.ae/program/%s' % show_id, 'AWAANSeason')
f'http://awaan.ae/program/{show_id}', 'AWAANSeason')
class AWAANBaseIE(InfoExtractor):
@@ -75,11 +72,11 @@ class AWAANVideoIE(AWAANBaseIE):
video_id = self._match_id(url)
video_data = self._download_json(
'http://admin.mangomolo.com/analytics/index.php/plus/video?id=%s' % video_id,
f'http://admin.mangomolo.com/analytics/index.php/plus/video?id={video_id}',
video_id, headers={'Origin': 'http://awaan.ae'})
info = self._parse_video_data(video_data, video_id, False)
embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' + compat_urllib_parse_urlencode({
embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' + urllib.parse.urlencode({
'id': video_data['id'],
'user_id': video_data['user_id'],
'signature': video_data['signature'],
@@ -117,11 +114,11 @@ class AWAANLiveIE(AWAANBaseIE):
channel_id = self._match_id(url)
channel_data = self._download_json(
'http://admin.mangomolo.com/analytics/index.php/plus/getchanneldetails?channel_id=%s' % channel_id,
f'http://admin.mangomolo.com/analytics/index.php/plus/getchanneldetails?channel_id={channel_id}',
channel_id, headers={'Origin': 'http://awaan.ae'})
info = self._parse_video_data(channel_data, channel_id, True)
embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/index?' + compat_urllib_parse_urlencode({
embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/index?' + urllib.parse.urlencode({
'id': base64.b64encode(channel_data['user_id'].encode()).decode(),
'channelid': base64.b64encode(channel_data['id'].encode()).decode(),
'signature': channel_data['signature'],
@@ -159,7 +156,7 @@ class AWAANSeasonIE(InfoExtractor):
show_id = smuggled_data.get('show_id')
if show_id is None:
season = self._download_json(
'http://admin.mangomolo.com/analytics/index.php/plus/season_info?id=%s' % season_id,
f'http://admin.mangomolo.com/analytics/index.php/plus/season_info?id={season_id}',
season_id, headers={'Origin': 'http://awaan.ae'})
show_id = season['id']
data['show_id'] = show_id
@@ -167,7 +164,7 @@ class AWAANSeasonIE(InfoExtractor):
'http://admin.mangomolo.com/analytics/index.php/plus/show',
show_id, data=urlencode_postdata(data), headers={
'Origin': 'http://awaan.ae',
'Content-Type': 'application/x-www-form-urlencoded'
'Content-Type': 'application/x-www-form-urlencoded',
})
if not season_id:
season_id = show['default_season']
@@ -177,8 +174,8 @@ class AWAANSeasonIE(InfoExtractor):
entries = []
for video in show['videos']:
video_id = compat_str(video['id'])
video_id = str(video['id'])
entries.append(self.url_result(
'http://awaan.ae/media/%s' % video_id, 'AWAANVideo', video_id))
f'http://awaan.ae/media/{video_id}', 'AWAANVideo', video_id))
return self.playlist_result(entries, season_id, title)
yt_dlp/extractor/aws.py
+14 -14
View File
@@ -1,9 +1,9 @@
import datetime
import datetime as dt
import hashlib
import hmac
import urllib.parse
from .common import InfoExtractor
from ..compat import compat_urllib_parse_urlencode
class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
@@ -12,26 +12,26 @@ class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with
def _aws_execute_api(self, aws_dict, video_id, query=None):
query = query or {}
amz_date = datetime.datetime.utcnow().strftime('%Y%m%dT%H%M%SZ')
amz_date = dt.datetime.now(dt.timezone.utc).strftime('%Y%m%dT%H%M%SZ')
date = amz_date[:8]
headers = {
'Accept': 'application/json',
'Host': self._AWS_PROXY_HOST,
'X-Amz-Date': amz_date,
'X-Api-Key': self._AWS_API_KEY
'X-Api-Key': self._AWS_API_KEY,
}
session_token = aws_dict.get('session_token')
if session_token:
headers['X-Amz-Security-Token'] = session_token
def aws_hash(s):
return hashlib.sha256(s.encode('utf-8')).hexdigest()
return hashlib.sha256(s.encode()).hexdigest()
# Task 1: http://docs.aws.amazon.com/general/latest/gr/sigv4-create-canonical-request.html
canonical_querystring = compat_urllib_parse_urlencode(query)
canonical_querystring = urllib.parse.urlencode(query)
canonical_headers = ''
for header_name, header_value in sorted(headers.items()):
canonical_headers += '%s:%s\n' % (header_name.lower(), header_value)
canonical_headers += f'{header_name.lower()}:{header_value}\n'
signed_headers = ';'.join([header.lower() for header in sorted(headers.keys())])
canonical_request = '\n'.join([
'GET',
@@ -39,7 +39,7 @@ class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with
canonical_querystring,
canonical_headers,
signed_headers,
aws_hash('')
aws_hash(''),
])
# Task 2: http://docs.aws.amazon.com/general/latest/gr/sigv4-create-string-to-sign.html
@@ -49,7 +49,7 @@ class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with
# Task 3: http://docs.aws.amazon.com/general/latest/gr/sigv4-calculate-signature.html
def aws_hmac(key, msg):
return hmac.new(key, msg.encode('utf-8'), hashlib.sha256)
return hmac.new(key, msg.encode(), hashlib.sha256)
def aws_hmac_digest(key, msg):
return aws_hmac(key, msg).digest()
@@ -57,7 +57,7 @@ class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with
def aws_hmac_hexdigest(key, msg):
return aws_hmac(key, msg).hexdigest()
k_signing = ('AWS4' + aws_dict['secret_key']).encode('utf-8')
k_signing = ('AWS4' + aws_dict['secret_key']).encode()
for value in credential_scope_list:
k_signing = aws_hmac_digest(k_signing, value)
@@ -65,11 +65,11 @@ class AWSIE(InfoExtractor): # XXX: Conventionally, base classes should end with
# Task 4: http://docs.aws.amazon.com/general/latest/gr/sigv4-add-signature-to-request.html
headers['Authorization'] = ', '.join([
'%s Credential=%s/%s' % (self._AWS_ALGORITHM, aws_dict['access_key'], credential_scope),
'SignedHeaders=%s' % signed_headers,
'Signature=%s' % signature,
'{} Credential={}/{}'.format(self._AWS_ALGORITHM, aws_dict['access_key'], credential_scope),
f'SignedHeaders={signed_headers}',
f'Signature={signature}',
])
return self._download_json(
'https://%s%s%s' % (self._AWS_PROXY_HOST, aws_dict['uri'], '?' + canonical_querystring if canonical_querystring else ''),
'https://{}{}{}'.format(self._AWS_PROXY_HOST, aws_dict['uri'], '?' + canonical_querystring if canonical_querystring else ''),
video_id, headers=headers)
yt_dlp/extractor/axs.py
+89
View File
@@ -0,0 +1,89 @@
from .common import InfoExtractor
from ..utils import (
float_or_none,
js_to_json,
parse_iso8601,
traverse_obj,
url_or_none,
)
class AxsIE(InfoExtractor):
IE_NAME = 'axs.tv'
_VALID_URL = r'https?://(?:www\.)?axs\.tv/(?:channel/(?:[^/?#]+/)+)?video/(?P<id>[^/?#]+)'
_TESTS = [{
'url': 'https://www.axs.tv/video/5f4dc776b70e4f1c194f22ef/',
'md5': '8d97736ae8e50c64df528e5e676778cf',
'info_dict': {
'id': '5f4dc776b70e4f1c194f22ef',
'title': 'Small Town',
'ext': 'mp4',
'description': 'md5:e314d28bfaa227a4d7ec965fae19997f',
'upload_date': '20230602',
'timestamp': 1685729564,
'duration': 1284.216,
'series': 'Rock & Roll Road Trip with Sammy Hagar',
'season': 'Season 2',
'season_number': 2,
'episode': '3',
'thumbnail': 'https://images.dotstudiopro.com/5f4e9d330a0c3b295a7e8394',
},
}, {
'url': 'https://www.axs.tv/channel/rock-star-interview/video/daryl-hall',
'md5': '300ae795cd8f9984652c0949734ffbdc',
'info_dict': {
'id': '5f488148b70e4f392572977c',
'display_id': 'daryl-hall',
'title': 'Daryl Hall',
'ext': 'mp4',
'description': 'md5:e54ecaa0f4b5683fc9259e9e4b196628',
'upload_date': '20230214',
'timestamp': 1676403615,
'duration': 2570.668,
'series': 'The Big Interview with Dan Rather',
'season': 'Season 3',
'season_number': 3,
'episode': '5',
'thumbnail': 'https://images.dotstudiopro.com/5f4d1901f340b50d937cec32',
},
}]
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
webpage_json_data = self._search_json(
r'mountObj\s*=', webpage, 'video ID data', display_id,
transform_source=js_to_json)
video_id = webpage_json_data['video_id']
company_id = webpage_json_data['company_id']
meta = self._download_json(
f'https://api.myspotlight.tv/dotplayer/video/{company_id}/{video_id}',
video_id, query={'device_type': 'desktop_web'})['video']
formats = self._extract_m3u8_formats(
meta['video_m3u8'], video_id, 'mp4', m3u8_id='hls')
subtitles = {}
for cc in traverse_obj(meta, ('closeCaption', lambda _, v: url_or_none(v['srtPath']))):
subtitles.setdefault(cc.get('srtShortLang') or 'en', []).append(
{'ext': cc.get('srtExt'), 'url': cc['srtPath']})
return {
'id': video_id,
'display_id': display_id,
'formats': formats,
**traverse_obj(meta, {
'title': ('title', {str}),
'description': ('description', {str}),
'series': ('seriestitle', {str}),
'season_number': ('season', {int}),
'episode': ('episode', {str}),
'duration': ('duration', {float_or_none}),
'timestamp': ('updated_at', {parse_iso8601}),
'thumbnail': ('thumb', {url_or_none}),
}),
'subtitles': subtitles,
}
yt_dlp/extractor/azmedien.py
+3 -3
View File
@@ -38,14 +38,14 @@ class AZMedienIE(InfoExtractor):
'timestamp': 1538328802,
'view_count': int,
'thumbnail': 'http://cfvod.kaltura.com/p/1719221/sp/171922100/thumbnail/entry_id/1_anruz3wy/version/100031',
'duration': 1930
'duration': 1930,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.telebaern.tv/telebaern-news/montag-1-oktober-2018-ganze-sendung-133531189#video=0_7xjo9lf1',
'only_matching': True
'only_matching': True,
}]
_API_TEMPL = 'https://www.%s/api/pub/gql/%s/NewsArticleTeaser/a4016f65fe62b81dc6664dd9f4910e4ab40383be'
_PARTNER_ID = '1719221'
@@ -62,5 +62,5 @@ class AZMedienIE(InfoExtractor):
})['data']['context']['mainAsset']['video']['kaltura']['kalturaId']
return self.url_result(
'kaltura:%s:%s' % (self._PARTNER_ID, entry_id),
f'kaltura:{self._PARTNER_ID}:{entry_id}',
ie=KalturaIE.ie_key(), video_id=entry_id)
yt_dlp/extractor/baidu.py
+4 -3
View File
@@ -24,8 +24,9 @@ class BaiduVideoIE(InfoExtractor):
}]
def _call_api(self, path, category, playlist_id, note):
return self._download_json('http://app.video.baidu.com/%s/?worktype=adnative%s&id=%s' % (
path, category, playlist_id), playlist_id, note)
return self._download_json(
f'http://app.video.baidu.com/{path}/?worktype=adnative{category}&id={playlist_id}',
playlist_id, note)
def _real_extract(self, url):
category, playlist_id = self._match_valid_url(url).groups()
@@ -44,7 +45,7 @@ class BaiduVideoIE(InfoExtractor):
'xqsingle', category, playlist_id, 'Download episodes JSON metadata')
entries = [self.url_result(
episode['url'], video_title=episode['title']
episode['url'], video_title=episode['title'],
) for episode in episodes_detail['videos']]
return self.playlist_result(
yt_dlp/extractor/banbye.py
+94 -16
View File
@@ -1,15 +1,16 @@
import math
import urllib.parse
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlparse,
compat_parse_qs,
)
from ..utils import (
format_field,
InAdvancePagedList,
determine_ext,
format_field,
int_or_none,
join_nonempty,
traverse_obj,
unified_timestamp,
url_or_none,
)
@@ -20,8 +21,8 @@ class BanByeBaseIE(InfoExtractor):
@staticmethod
def _extract_playlist_id(url, param='playlist'):
return compat_parse_qs(
compat_urllib_parse_urlparse(url).query).get(param, [None])[0]
return urllib.parse.parse_qs(
urllib.parse.urlparse(url).query).get(param, [None])[0]
def _extract_playlist(self, playlist_id):
data = self._download_json(f'{self._API_BASE}/playlists/{playlist_id}', playlist_id)
@@ -31,8 +32,9 @@ class BanByeBaseIE(InfoExtractor):
class BanByeIE(BanByeBaseIE):
_VALID_URL = r'https?://(?:www\.)?banbye.com/(?:en/)?watch/(?P<id>\w+)'
_VALID_URL = r'https?://(?:www\.)?banbye\.com/(?:en/)?watch/(?P<id>[\w-]+)'
_TESTS = [{
# ['src']['mp4']['levels'] direct mp4 urls only
'url': 'https://banbye.com/watch/v_ytfmvkVYLE8T',
'md5': '2f4ea15c5ca259a73d909b2cfd558eb5',
'info_dict': {
@@ -59,7 +61,70 @@ class BanByeIE(BanByeBaseIE):
'title': 'Krzysztof Karoń',
'id': 'p_Ld82N6gBw_OJ',
},
'playlist_count': 9,
'playlist_mincount': 9,
}, {
# ['src']['mp4']['levels'] direct mp4 urls only
'url': 'https://banbye.com/watch/v_kb6_o1Kyq-CD',
'info_dict': {
'id': 'v_kb6_o1Kyq-CD',
'ext': 'mp4',
'title': 'Co tak naprawdę dzieje się we Francji?! Czy Warszawa a potem cała Polska będzie drugim Paryżem?!🤔🇵🇱',
'description': 'md5:82be4c0e13eae8ea1ca8b9f2e07226a8',
'uploader': 'Marcin Rola - MOIM ZDANIEM!🇵🇱',
'channel_id': 'ch_QgWnHvDG2fo5',
'channel_url': 'https://banbye.com/channel/ch_QgWnHvDG2fo5',
'duration': 597,
'timestamp': 1688642656,
'upload_date': '20230706',
'thumbnail': 'https://cdn.banbye.com/video/v_kb6_o1Kyq-CD/96.webp',
'tags': ['Paryż', 'Francja', 'Polska', 'Imigranci', 'Morawiecki', 'Tusk'],
'like_count': int,
'dislike_count': int,
'view_count': int,
'comment_count': int,
},
}, {
# ['src']['hls']['levels'] variant m3u8 urls only; master m3u8 is 404
'url': 'https://banbye.com/watch/v_a_gPFuC9LoW5',
'info_dict': {
'id': 'v_a_gPFuC9LoW5',
'ext': 'mp4',
'title': 'md5:183524056bebdfa245fd6d214f63c0fe',
'description': 'md5:943ac87287ca98d28d8b8797719827c6',
'uploader': 'wRealu24',
'channel_id': 'ch_wrealu24',
'channel_url': 'https://banbye.com/channel/ch_wrealu24',
'upload_date': '20231113',
'timestamp': 1699874062,
'view_count': int,
'like_count': int,
'dislike_count': int,
'comment_count': int,
'thumbnail': 'https://cdn.banbye.com/video/v_a_gPFuC9LoW5/96.webp',
'tags': ['jaszczur', 'sejm', 'lewica', 'polska', 'ukrainizacja', 'pierwszeposiedzeniesejmu'],
},
'expected_warnings': ['Failed to download m3u8'],
}, {
# ['src']['hls']['masterPlaylist'] m3u8 only
'url': 'https://banbye.com/watch/v_B0rsKWsr-aaa',
'info_dict': {
'id': 'v_B0rsKWsr-aaa',
'ext': 'mp4',
'title': 'md5:00b254164b82101b3f9e5326037447ed',
'description': 'md5:3fd8b48aa81954ba024bc60f5de6e167',
'uploader': 'PSTV Piotr Szlachtowicz ',
'channel_id': 'ch_KV9EVObkB9wB',
'channel_url': 'https://banbye.com/channel/ch_KV9EVObkB9wB',
'upload_date': '20240629',
'timestamp': 1719646816,
'duration': 2377,
'view_count': int,
'like_count': int,
'dislike_count': int,
'comment_count': int,
'thumbnail': 'https://cdn.banbye.com/video/v_B0rsKWsr-aaa/96.webp',
'tags': ['Biden', 'Trump', 'Wybory', 'USA'],
},
}]
def _real_extract(self, url):
@@ -74,11 +139,24 @@ class BanByeIE(BanByeBaseIE):
'id': f'{quality}p',
'url': f'{self._CDN_BASE}/video/{video_id}/{quality}.webp',
} for quality in [48, 96, 144, 240, 512, 1080]]
formats = [{
'format_id': f'http-{quality}p',
'quality': quality,
'url': f'{self._CDN_BASE}/video/{video_id}/{quality}.mp4',
} for quality in data['quality']]
formats = []
url_data = self._download_json(f'{self._API_BASE}/videos/{video_id}/url', video_id, data=b'')
if master_url := traverse_obj(url_data, ('src', 'hls', 'masterPlaylist', {url_or_none})):
formats = self._extract_m3u8_formats(master_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
for format_id, format_url in traverse_obj(url_data, (
'src', ('mp4', 'hls'), 'levels', {dict.items}, lambda _, v: url_or_none(v[1]))):
ext = determine_ext(format_url)
is_hls = ext == 'm3u8'
formats.append({
'url': format_url,
'ext': 'mp4' if is_hls else ext,
'format_id': join_nonempty(is_hls and 'hls', format_id),
'protocol': 'm3u8_native' if is_hls else 'https',
'height': int_or_none(format_id),
})
self._remove_duplicate_formats(formats)
return {
'id': video_id,
@@ -100,7 +178,7 @@ class BanByeIE(BanByeBaseIE):
class BanByeChannelIE(BanByeBaseIE):
_VALID_URL = r'https?://(?:www\.)?banbye.com/(?:en/)?channel/(?P<id>\w+)'
_VALID_URL = r'https?://(?:www\.)?banbye\.com/(?:en/)?channel/(?P<id>\w+)'
_TESTS = [{
'url': 'https://banbye.com/channel/ch_wrealu24',
'info_dict': {
@@ -132,7 +210,7 @@ class BanByeChannelIE(BanByeBaseIE):
'sort': 'new',
'limit': self._PAGE_SIZE,
'offset': page_num * self._PAGE_SIZE,
}, note=f'Downloading page {page_num+1}')
}, note=f'Downloading page {page_num + 1}')
return [
self.url_result(f"{self._VIDEO_BASE}/{video['_id']}", BanByeIE)
for video in data['items']
yt_dlp/extractor/bandcamp.py
+15 -16
View File
@@ -3,7 +3,6 @@ import re
import time
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
KNOWN_EXTENSIONS,
ExtractorError,
@@ -42,7 +41,7 @@ class BandcampIE(InfoExtractor):
'uploader_id': 'youtube-dl',
'thumbnail': 'https://f4.bcbits.com/img/a3216802731_5.jpg',
},
'_skip': 'There is a limit of 200 free downloads / month for the test song'
'skip': 'There is a limit of 200 free downloads / month for the test song',
}, {
# free download
'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
@@ -119,7 +118,7 @@ class BandcampIE(InfoExtractor):
def _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
return self._parse_json(self._html_search_regex(
r'data-%s=(["\'])({.+?})\1' % attr, webpage,
rf'data-{attr}=(["\'])({{.+?}})\1', webpage,
attr + ' data', group=2), video_id, fatal=fatal)
def _real_extract(self, url):
@@ -167,7 +166,7 @@ class BandcampIE(InfoExtractor):
download_link = tralbum.get('freeDownloadPage')
if download_link:
track_id = compat_str(tralbum['id'])
track_id = str(tralbum['id'])
download_webpage = self._download_webpage(
download_link, track_id, 'Downloading free downloads page')
@@ -192,7 +191,7 @@ class BandcampIE(InfoExtractor):
if isinstance(download_formats_list, list):
for f in blob['download_formats']:
name, ext = f.get('name'), f.get('file_extension')
if all(isinstance(x, compat_str) for x in (name, ext)):
if all(isinstance(x, str) for x in (name, ext)):
download_formats[name] = ext.strip('.')
for format_id, f in downloads.items():
@@ -207,7 +206,7 @@ class BandcampIE(InfoExtractor):
})
format_id = f.get('encoding_name') or format_id
stat = self._download_json(
stat_url, track_id, 'Downloading %s JSON' % format_id,
stat_url, track_id, f'Downloading {format_id} JSON',
transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
fatal=False)
if not stat:
@@ -225,7 +224,7 @@ class BandcampIE(InfoExtractor):
'acodec': format_id.split('-')[0],
})
title = '%s - %s' % (artist, track) if artist else track
title = f'{artist} - {track}' if artist else track
if not duration:
duration = float_or_none(self._html_search_meta(
@@ -267,7 +266,7 @@ class BandcampAlbumIE(BandcampIE): # XXX: Do not subclass from concrete IE
'timestamp': 1311756226,
'upload_date': '20110727',
'uploader': 'Blazo',
}
},
},
{
'md5': '1a2c32e2691474643e912cc6cd4bffaa',
@@ -278,7 +277,7 @@ class BandcampAlbumIE(BandcampIE): # XXX: Do not subclass from concrete IE
'timestamp': 1311757238,
'upload_date': '20110727',
'uploader': 'Blazo',
}
},
},
],
'info_dict': {
@@ -287,9 +286,9 @@ class BandcampAlbumIE(BandcampIE): # XXX: Do not subclass from concrete IE
'uploader_id': 'blazo',
},
'params': {
'playlistend': 2
'playlistend': 2,
},
'skip': 'Bandcamp imposes download limits.'
'skip': 'Bandcamp imposes download limits.',
}, {
'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
'info_dict': {
@@ -324,7 +323,7 @@ class BandcampAlbumIE(BandcampIE): # XXX: Do not subclass from concrete IE
def suitable(cls, url):
return (False
if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
else super(BandcampAlbumIE, cls).suitable(url))
else super().suitable(url))
def _real_extract(self, url):
uploader_id, album_id = self._match_valid_url(url).groups()
@@ -376,7 +375,7 @@ class BandcampWeeklyIE(BandcampIE): # XXX: Do not subclass from concrete IE
},
}, {
'url': 'https://bandcamp.com/?blah/blah@&show=228',
'only_matching': True
'only_matching': True,
}]
def _real_extract(self, url):
@@ -407,7 +406,7 @@ class BandcampWeeklyIE(BandcampIE): # XXX: Do not subclass from concrete IE
title = show.get('audio_title') or 'Bandcamp Weekly'
subtitle = show.get('subtitle')
if subtitle:
title += ' - %s' % subtitle
title += f' - {subtitle}'
return {
'id': show_id,
@@ -419,7 +418,7 @@ class BandcampWeeklyIE(BandcampIE): # XXX: Do not subclass from concrete IE
'series': 'Bandcamp Weekly',
'episode': show.get('subtitle'),
'episode_id': show_id,
'formats': formats
'formats': formats,
}
@@ -440,7 +439,7 @@ class BandcampUserIE(InfoExtractor):
'url': 'http://dotscale.bandcamp.com',
'info_dict': {
'id': 'dotscale',
'title': 'Discography of dotscale'
'title': 'Discography of dotscale',
},
'playlist_count': 1,
}, {

Some files were not shown because too many files have changed in this diff Show More