DuckDuckWhale/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-12-16 22:46:45 +00:00
Code Issues Releases Wiki Activity
f0812d7848
youtube-dl/youtube_dl/utils.py
Hubert Hirtz f0812d7848
[utils] Handle user:pass in URLs (#28801) 
* Handle user:pass in URLs

Fixes "nonnumeric port" errors when youtube-dl is given URLs with
usernames and passwords such as:

    http://username:password@example.com/myvideo.mp4

Refs:
- https://en.wikipedia.org/wiki/Basic_access_authentication
- https://tools.ietf.org/html/rfc1738#section-3.1
- https://docs.python.org/3.8/library/urllib.parse.html#urllib.parse.urlsplit

Fixes #18276 (point 4)
Fixes #20258
Fixes #26211 (see comment)

* Align code with yt-dlp

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2024-03-04 01:27:55 +00:00

6534 lines
196 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#!/usr/bin/env python
# coding: utf-8
from __future__ import unicode_literals
import base64
import binascii
import calendar
import codecs
import collections
import contextlib
import ctypes
import datetime
import email.utils
import email.header
import errno
import functools
import inspect
import io
import itertools
import json
import locale
import math
import operator
import os
import platform
import random
import re
import socket
import ssl
import subprocess
import sys
import tempfile
import time
import traceback
import unicodedata
import xml.etree.ElementTree
import zlib
from .compat import (
compat_HTMLParseError,
compat_HTMLParser,
compat_basestring,
compat_brotli as brotli,
compat_casefold,
compat_chr,
compat_collections_abc,
compat_cookiejar,
compat_ctypes_WINFUNCTYPE,
compat_datetime_timedelta_total_seconds,
compat_etree_fromstring,
compat_expanduser,
compat_html_entities,
compat_html_entities_html5,
compat_http_client,
compat_integer_types,
compat_kwargs,
compat_ncompress as ncompress,
compat_os_name,
compat_re_Match,
compat_re_Pattern,
compat_shlex_quote,
compat_str,
compat_struct_pack,
compat_struct_unpack,
compat_urllib_error,
compat_urllib_HTTPError,
compat_urllib_parse,
compat_urllib_parse_parse_qs as compat_parse_qs,
compat_urllib_parse_urlencode,
compat_urllib_parse_urlparse,
compat_urllib_parse_unquote_plus,
compat_urllib_request,
compat_xpath,
)
from .socks import (
ProxyType,
sockssocket,
)
def register_socks_protocols():
# "Register" SOCKS protocols
# In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
# URLs with protocols not in urlparse.uses_netloc are not handled correctly
for scheme in ('socks', 'socks4', 'socks4a', 'socks5'):
if scheme not in compat_urllib_parse.uses_netloc:
compat_urllib_parse.uses_netloc.append(scheme)
# Unfavoured alias
compiled_regex_type = compat_re_Pattern
def random_user_agent():
_USER_AGENT_TPL = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
_CHROME_VERSIONS = (
'74.0.3729.129',
'76.0.3780.3',
'76.0.3780.2',
'74.0.3729.128',
'76.0.3780.1',
'76.0.3780.0',
'75.0.3770.15',
'74.0.3729.127',
'74.0.3729.126',
'76.0.3779.1',
'76.0.3779.0',
'75.0.3770.14',
'74.0.3729.125',
'76.0.3778.1',
'76.0.3778.0',
'75.0.3770.13',
'74.0.3729.124',
'74.0.3729.123',
'73.0.3683.121',
'76.0.3777.1',
'76.0.3777.0',
'75.0.3770.12',
'74.0.3729.122',
'76.0.3776.4',
'75.0.3770.11',
'74.0.3729.121',
'76.0.3776.3',
'76.0.3776.2',
'73.0.3683.120',
'74.0.3729.120',
'74.0.3729.119',
'74.0.3729.118',
'76.0.3776.1',
'76.0.3776.0',
'76.0.3775.5',
'75.0.3770.10',
'74.0.3729.117',
'76.0.3775.4',
'76.0.3775.3',
'74.0.3729.116',
'75.0.3770.9',
'76.0.3775.2',
'76.0.3775.1',
'76.0.3775.0',
'75.0.3770.8',
'74.0.3729.115',
'74.0.3729.114',
'76.0.3774.1',
'76.0.3774.0',
'75.0.3770.7',
'74.0.3729.113',
'74.0.3729.112',
'74.0.3729.111',
'76.0.3773.1',
'76.0.3773.0',
'75.0.3770.6',
'74.0.3729.110',
'74.0.3729.109',
'76.0.3772.1',
'76.0.3772.0',
'75.0.3770.5',
'74.0.3729.108',
'74.0.3729.107',
'76.0.3771.1',
'76.0.3771.0',
'75.0.3770.4',
'74.0.3729.106',
'74.0.3729.105',
'75.0.3770.3',
'74.0.3729.104',
'74.0.3729.103',
'74.0.3729.102',
'75.0.3770.2',
'74.0.3729.101',
'75.0.3770.1',
'75.0.3770.0',
'74.0.3729.100',
'75.0.3769.5',
'75.0.3769.4',
'74.0.3729.99',
'75.0.3769.3',
'75.0.3769.2',
'75.0.3768.6',
'74.0.3729.98',
'75.0.3769.1',
'75.0.3769.0',
'74.0.3729.97',
'73.0.3683.119',
'73.0.3683.118',
'74.0.3729.96',
'75.0.3768.5',
'75.0.3768.4',
'75.0.3768.3',
'75.0.3768.2',
'74.0.3729.95',
'74.0.3729.94',
'75.0.3768.1',
'75.0.3768.0',
'74.0.3729.93',
'74.0.3729.92',
'73.0.3683.117',
'74.0.3729.91',
'75.0.3766.3',
'74.0.3729.90',
'75.0.3767.2',
'75.0.3767.1',
'75.0.3767.0',
'74.0.3729.89',
'73.0.3683.116',
'75.0.3766.2',
'74.0.3729.88',
'75.0.3766.1',
'75.0.3766.0',
'74.0.3729.87',
'73.0.3683.115',
'74.0.3729.86',
'75.0.3765.1',
'75.0.3765.0',
'74.0.3729.85',
'73.0.3683.114',
'74.0.3729.84',
'75.0.3764.1',
'75.0.3764.0',
'74.0.3729.83',
'73.0.3683.113',
'75.0.3763.2',
'75.0.3761.4',
'74.0.3729.82',
'75.0.3763.1',
'75.0.3763.0',
'74.0.3729.81',
'73.0.3683.112',
'75.0.3762.1',
'75.0.3762.0',
'74.0.3729.80',
'75.0.3761.3',
'74.0.3729.79',
'73.0.3683.111',
'75.0.3761.2',
'74.0.3729.78',
'74.0.3729.77',
'75.0.3761.1',
'75.0.3761.0',
'73.0.3683.110',
'74.0.3729.76',
'74.0.3729.75',
'75.0.3760.0',
'74.0.3729.74',
'75.0.3759.8',
'75.0.3759.7',
'75.0.3759.6',
'74.0.3729.73',
'75.0.3759.5',
'74.0.3729.72',
'73.0.3683.109',
'75.0.3759.4',
'75.0.3759.3',
'74.0.3729.71',
'75.0.3759.2',
'74.0.3729.70',
'73.0.3683.108',
'74.0.3729.69',
'75.0.3759.1',
'75.0.3759.0',
'74.0.3729.68',
'73.0.3683.107',
'74.0.3729.67',
'75.0.3758.1',
'75.0.3758.0',
'74.0.3729.66',
'73.0.3683.106',
'74.0.3729.65',
'75.0.3757.1',
'75.0.3757.0',
'74.0.3729.64',
'73.0.3683.105',
'74.0.3729.63',
'75.0.3756.1',
'75.0.3756.0',
'74.0.3729.62',
'73.0.3683.104',
'75.0.3755.3',
'75.0.3755.2',
'73.0.3683.103',
'75.0.3755.1',
'75.0.3755.0',
'74.0.3729.61',
'73.0.3683.102',
'74.0.3729.60',
'75.0.3754.2',
'74.0.3729.59',
'75.0.3753.4',
'74.0.3729.58',
'75.0.3754.1',
'75.0.3754.0',
'74.0.3729.57',
'73.0.3683.101',
'75.0.3753.3',
'75.0.3752.2',
'75.0.3753.2',
'74.0.3729.56',
'75.0.3753.1',
'75.0.3753.0',
'74.0.3729.55',
'73.0.3683.100',
'74.0.3729.54',
'75.0.3752.1',
'75.0.3752.0',
'74.0.3729.53',
'73.0.3683.99',
'74.0.3729.52',
'75.0.3751.1',
'75.0.3751.0',
'74.0.3729.51',
'73.0.3683.98',
'74.0.3729.50',
'75.0.3750.0',
'74.0.3729.49',
'74.0.3729.48',
'74.0.3729.47',
'75.0.3749.3',
'74.0.3729.46',
'73.0.3683.97',
'75.0.3749.2',
'74.0.3729.45',
'75.0.3749.1',
'75.0.3749.0',
'74.0.3729.44',
'73.0.3683.96',
'74.0.3729.43',
'74.0.3729.42',
'75.0.3748.1',
'75.0.3748.0',
'74.0.3729.41',
'75.0.3747.1',
'73.0.3683.95',
'75.0.3746.4',
'74.0.3729.40',
'74.0.3729.39',
'75.0.3747.0',
'75.0.3746.3',
'75.0.3746.2',
'74.0.3729.38',
'75.0.3746.1',
'75.0.3746.0',
'74.0.3729.37',
'73.0.3683.94',
'75.0.3745.5',
'75.0.3745.4',
'75.0.3745.3',
'75.0.3745.2',
'74.0.3729.36',
'75.0.3745.1',
'75.0.3745.0',
'75.0.3744.2',
'74.0.3729.35',
'73.0.3683.93',
'74.0.3729.34',
'75.0.3744.1',
'75.0.3744.0',
'74.0.3729.33',
'73.0.3683.92',
'74.0.3729.32',
'74.0.3729.31',
'73.0.3683.91',
'75.0.3741.2',
'75.0.3740.5',
'74.0.3729.30',
'75.0.3741.1',
'75.0.3741.0',
'74.0.3729.29',
'75.0.3740.4',
'73.0.3683.90',
'74.0.3729.28',
'75.0.3740.3',
'73.0.3683.89',
'75.0.3740.2',
'74.0.3729.27',
'75.0.3740.1',
'75.0.3740.0',
'74.0.3729.26',
'73.0.3683.88',
'73.0.3683.87',
'74.0.3729.25',
'75.0.3739.1',
'75.0.3739.0',
'73.0.3683.86',
'74.0.3729.24',
'73.0.3683.85',
'75.0.3738.4',
'75.0.3738.3',
'75.0.3738.2',
'75.0.3738.1',
'75.0.3738.0',
'74.0.3729.23',
'73.0.3683.84',
'74.0.3729.22',
'74.0.3729.21',
'75.0.3737.1',
'75.0.3737.0',
'74.0.3729.20',
'73.0.3683.83',
'74.0.3729.19',
'75.0.3736.1',
'75.0.3736.0',
'74.0.3729.18',
'73.0.3683.82',
'74.0.3729.17',
'75.0.3735.1',
'75.0.3735.0',
'74.0.3729.16',
'73.0.3683.81',
'75.0.3734.1',
'75.0.3734.0',
'74.0.3729.15',
'73.0.3683.80',
'74.0.3729.14',
'75.0.3733.1',
'75.0.3733.0',
'75.0.3732.1',
'74.0.3729.13',
'74.0.3729.12',
'73.0.3683.79',
'74.0.3729.11',
'75.0.3732.0',
'74.0.3729.10',
'73.0.3683.78',
'74.0.3729.9',
'74.0.3729.8',
'74.0.3729.7',
'75.0.3731.3',
'75.0.3731.2',
'75.0.3731.0',
'74.0.3729.6',
'73.0.3683.77',
'73.0.3683.76',
'75.0.3730.5',
'75.0.3730.4',
'73.0.3683.75',
'74.0.3729.5',
'73.0.3683.74',
'75.0.3730.3',
'75.0.3730.2',
'74.0.3729.4',
'73.0.3683.73',
'73.0.3683.72',
'75.0.3730.1',
'75.0.3730.0',
'74.0.3729.3',
'73.0.3683.71',
'74.0.3729.2',
'73.0.3683.70',
'74.0.3729.1',
'74.0.3729.0',
'74.0.3726.4',
'73.0.3683.69',
'74.0.3726.3',
'74.0.3728.0',
'74.0.3726.2',
'73.0.3683.68',
'74.0.3726.1',
'74.0.3726.0',
'74.0.3725.4',
'73.0.3683.67',
'73.0.3683.66',
'74.0.3725.3',
'74.0.3725.2',
'74.0.3725.1',
'74.0.3724.8',
'74.0.3725.0',
'73.0.3683.65',
'74.0.3724.7',
'74.0.3724.6',
'74.0.3724.5',
'74.0.3724.4',
'74.0.3724.3',
'74.0.3724.2',
'74.0.3724.1',
'74.0.3724.0',
'73.0.3683.64',
'74.0.3723.1',
'74.0.3723.0',
'73.0.3683.63',
'74.0.3722.1',
'74.0.3722.0',
'73.0.3683.62',
'74.0.3718.9',
'74.0.3702.3',
'74.0.3721.3',
'74.0.3721.2',
'74.0.3721.1',
'74.0.3721.0',
'74.0.3720.6',
'73.0.3683.61',
'72.0.3626.122',
'73.0.3683.60',
'74.0.3720.5',
'72.0.3626.121',
'74.0.3718.8',
'74.0.3720.4',
'74.0.3720.3',
'74.0.3718.7',
'74.0.3720.2',
'74.0.3720.1',
'74.0.3720.0',
'74.0.3718.6',
'74.0.3719.5',
'73.0.3683.59',
'74.0.3718.5',
'74.0.3718.4',
'74.0.3719.4',
'74.0.3719.3',
'74.0.3719.2',
'74.0.3719.1',
'73.0.3683.58',
'74.0.3719.0',
'73.0.3683.57',
'73.0.3683.56',
'74.0.3718.3',
'73.0.3683.55',
'74.0.3718.2',
'74.0.3718.1',
'74.0.3718.0',
'73.0.3683.54',
'74.0.3717.2',
'73.0.3683.53',
'74.0.3717.1',
'74.0.3717.0',
'73.0.3683.52',
'74.0.3716.1',
'74.0.3716.0',
'73.0.3683.51',
'74.0.3715.1',
'74.0.3715.0',
'73.0.3683.50',
'74.0.3711.2',
'74.0.3714.2',
'74.0.3713.3',
'74.0.3714.1',
'74.0.3714.0',
'73.0.3683.49',
'74.0.3713.1',
'74.0.3713.0',
'72.0.3626.120',
'73.0.3683.48',
'74.0.3712.2',
'74.0.3712.1',
'74.0.3712.0',
'73.0.3683.47',
'72.0.3626.119',
'73.0.3683.46',
'74.0.3710.2',
'72.0.3626.118',
'74.0.3711.1',
'74.0.3711.0',
'73.0.3683.45',
'72.0.3626.117',
'74.0.3710.1',
'74.0.3710.0',
'73.0.3683.44',
'72.0.3626.116',
'74.0.3709.1',
'74.0.3709.0',
'74.0.3704.9',
'73.0.3683.43',
'72.0.3626.115',
'74.0.3704.8',
'74.0.3704.7',
'74.0.3708.0',
'74.0.3706.7',
'74.0.3704.6',
'73.0.3683.42',
'72.0.3626.114',
'74.0.3706.6',
'72.0.3626.113',
'74.0.3704.5',
'74.0.3706.5',
'74.0.3706.4',
'74.0.3706.3',
'74.0.3706.2',
'74.0.3706.1',
'74.0.3706.0',
'73.0.3683.41',
'72.0.3626.112',
'74.0.3705.1',
'74.0.3705.0',
'73.0.3683.40',
'72.0.3626.111',
'73.0.3683.39',
'74.0.3704.4',
'73.0.3683.38',
'74.0.3704.3',
'74.0.3704.2',
'74.0.3704.1',
'74.0.3704.0',
'73.0.3683.37',
'72.0.3626.110',
'72.0.3626.109',
'74.0.3703.3',
'74.0.3703.2',
'73.0.3683.36',
'74.0.3703.1',
'74.0.3703.0',
'73.0.3683.35',
'72.0.3626.108',
'74.0.3702.2',
'74.0.3699.3',
'74.0.3702.1',
'74.0.3702.0',
'73.0.3683.34',
'72.0.3626.107',
'73.0.3683.33',
'74.0.3701.1',
'74.0.3701.0',
'73.0.3683.32',
'73.0.3683.31',
'72.0.3626.105',
'74.0.3700.1',
'74.0.3700.0',
'73.0.3683.29',
'72.0.3626.103',
'74.0.3699.2',
'74.0.3699.1',
'74.0.3699.0',
'73.0.3683.28',
'72.0.3626.102',
'73.0.3683.27',
'73.0.3683.26',
'74.0.3698.0',
'74.0.3696.2',
'72.0.3626.101',
'73.0.3683.25',
'74.0.3696.1',
'74.0.3696.0',
'74.0.3694.8',
'72.0.3626.100',
'74.0.3694.7',
'74.0.3694.6',
'74.0.3694.5',
'74.0.3694.4',
'72.0.3626.99',
'72.0.3626.98',
'74.0.3694.3',
'73.0.3683.24',
'72.0.3626.97',
'72.0.3626.96',
'72.0.3626.95',
'73.0.3683.23',
'72.0.3626.94',
'73.0.3683.22',
'73.0.3683.21',
'72.0.3626.93',
'74.0.3694.2',
'72.0.3626.92',
'74.0.3694.1',
'74.0.3694.0',
'74.0.3693.6',
'73.0.3683.20',
'72.0.3626.91',
'74.0.3693.5',
'74.0.3693.4',
'74.0.3693.3',
'74.0.3693.2',
'73.0.3683.19',
'74.0.3693.1',
'74.0.3693.0',
'73.0.3683.18',
'72.0.3626.90',
'74.0.3692.1',
'74.0.3692.0',
'73.0.3683.17',
'72.0.3626.89',
'74.0.3687.3',
'74.0.3691.1',
'74.0.3691.0',
'73.0.3683.16',
'72.0.3626.88',
'72.0.3626.87',
'73.0.3683.15',
'74.0.3690.1',
'74.0.3690.0',
'73.0.3683.14',
'72.0.3626.86',
'73.0.3683.13',
'73.0.3683.12',
'74.0.3689.1',
'74.0.3689.0',
'73.0.3683.11',
'72.0.3626.85',
'73.0.3683.10',
'72.0.3626.84',
'73.0.3683.9',
'74.0.3688.1',
'74.0.3688.0',
'73.0.3683.8',
'72.0.3626.83',
'74.0.3687.2',
'74.0.3687.1',
'74.0.3687.0',
'73.0.3683.7',
'72.0.3626.82',
'74.0.3686.4',
'72.0.3626.81',
'74.0.3686.3',
'74.0.3686.2',
'74.0.3686.1',
'74.0.3686.0',
'73.0.3683.6',
'72.0.3626.80',
'74.0.3685.1',
'74.0.3685.0',
'73.0.3683.5',
'72.0.3626.79',
'74.0.3684.1',
'74.0.3684.0',
'73.0.3683.4',
'72.0.3626.78',
'72.0.3626.77',
'73.0.3683.3',
'73.0.3683.2',
'72.0.3626.76',
'73.0.3683.1',
'73.0.3683.0',
'72.0.3626.75',
'71.0.3578.141',
'73.0.3682.1',
'73.0.3682.0',
'72.0.3626.74',
'71.0.3578.140',
'73.0.3681.4',
'73.0.3681.3',
'73.0.3681.2',
'73.0.3681.1',
'73.0.3681.0',
'72.0.3626.73',
'71.0.3578.139',
'72.0.3626.72',
'72.0.3626.71',
'73.0.3680.1',
'73.0.3680.0',
'72.0.3626.70',
'71.0.3578.138',
'73.0.3678.2',
'73.0.3679.1',
'73.0.3679.0',
'72.0.3626.69',
'71.0.3578.137',
'73.0.3678.1',
'73.0.3678.0',
'71.0.3578.136',
'73.0.3677.1',
'73.0.3677.0',
'72.0.3626.68',
'72.0.3626.67',
'71.0.3578.135',
'73.0.3676.1',
'73.0.3676.0',
'73.0.3674.2',
'72.0.3626.66',
'71.0.3578.134',
'73.0.3674.1',
'73.0.3674.0',
'72.0.3626.65',
'71.0.3578.133',
'73.0.3673.2',
'73.0.3673.1',
'73.0.3673.0',
'72.0.3626.64',
'71.0.3578.132',
'72.0.3626.63',
'72.0.3626.62',
'72.0.3626.61',
'72.0.3626.60',
'73.0.3672.1',
'73.0.3672.0',
'72.0.3626.59',
'71.0.3578.131',
'73.0.3671.3',
'73.0.3671.2',
'73.0.3671.1',
'73.0.3671.0',
'72.0.3626.58',
'71.0.3578.130',
'73.0.3670.1',
'73.0.3670.0',
'72.0.3626.57',
'71.0.3578.129',
'73.0.3669.1',
'73.0.3669.0',
'72.0.3626.56',
'71.0.3578.128',
'73.0.3668.2',
'73.0.3668.1',
'73.0.3668.0',
'72.0.3626.55',
'71.0.3578.127',
'73.0.3667.2',
'73.0.3667.1',
'73.0.3667.0',
'72.0.3626.54',
'71.0.3578.126',
'73.0.3666.1',
'73.0.3666.0',
'72.0.3626.53',
'71.0.3578.125',
'73.0.3665.4',
'73.0.3665.3',
'72.0.3626.52',
'73.0.3665.2',
'73.0.3664.4',
'73.0.3665.1',
'73.0.3665.0',
'72.0.3626.51',
'71.0.3578.124',
'72.0.3626.50',
'73.0.3664.3',
'73.0.3664.2',
'73.0.3664.1',
'73.0.3664.0',
'73.0.3663.2',
'72.0.3626.49',
'71.0.3578.123',
'73.0.3663.1',
'73.0.3663.0',
'72.0.3626.48',
'71.0.3578.122',
'73.0.3662.1',
'73.0.3662.0',
'72.0.3626.47',
'71.0.3578.121',
'73.0.3661.1',
'72.0.3626.46',
'73.0.3661.0',
'72.0.3626.45',
'71.0.3578.120',
'73.0.3660.2',
'73.0.3660.1',
'73.0.3660.0',
'72.0.3626.44',
'71.0.3578.119',
'73.0.3659.1',
'73.0.3659.0',
'72.0.3626.43',
'71.0.3578.118',
'73.0.3658.1',
'73.0.3658.0',
'72.0.3626.42',
'71.0.3578.117',
'73.0.3657.1',
'73.0.3657.0',
'72.0.3626.41',
'71.0.3578.116',
'73.0.3656.1',
'73.0.3656.0',
'72.0.3626.40',
'71.0.3578.115',
'73.0.3655.1',
'73.0.3655.0',
'72.0.3626.39',
'71.0.3578.114',
'73.0.3654.1',
'73.0.3654.0',
'72.0.3626.38',
'71.0.3578.113',
'73.0.3653.1',
'73.0.3653.0',
'72.0.3626.37',
'71.0.3578.112',
'73.0.3652.1',
'73.0.3652.0',
'72.0.3626.36',
'71.0.3578.111',
'73.0.3651.1',
'73.0.3651.0',
'72.0.3626.35',
'71.0.3578.110',
'73.0.3650.1',
'73.0.3650.0',
'72.0.3626.34',
'71.0.3578.109',
'73.0.3649.1',
'73.0.3649.0',
'72.0.3626.33',
'71.0.3578.108',
'73.0.3648.2',
'73.0.3648.1',
'73.0.3648.0',
'72.0.3626.32',
'71.0.3578.107',
'73.0.3647.2',
'73.0.3647.1',
'73.0.3647.0',
'72.0.3626.31',
'71.0.3578.106',
'73.0.3635.3',
'73.0.3646.2',
'73.0.3646.1',
'73.0.3646.0',
'72.0.3626.30',
'71.0.3578.105',
'72.0.3626.29',
'73.0.3645.2',
'73.0.3645.1',
'73.0.3645.0',
'72.0.3626.28',
'71.0.3578.104',
'72.0.3626.27',
'72.0.3626.26',
'72.0.3626.25',
'72.0.3626.24',
'73.0.3644.0',
'73.0.3643.2',
'72.0.3626.23',
'71.0.3578.103',
'73.0.3643.1',
'73.0.3643.0',
'72.0.3626.22',
'71.0.3578.102',
'73.0.3642.1',
'73.0.3642.0',
'72.0.3626.21',
'71.0.3578.101',
'73.0.3641.1',
'73.0.3641.0',
'72.0.3626.20',
'71.0.3578.100',
'72.0.3626.19',
'73.0.3640.1',
'73.0.3640.0',
'72.0.3626.18',
'73.0.3639.1',
'71.0.3578.99',
'73.0.3639.0',
'72.0.3626.17',
'73.0.3638.2',
'72.0.3626.16',
'73.0.3638.1',
'73.0.3638.0',
'72.0.3626.15',
'71.0.3578.98',
'73.0.3635.2',
'71.0.3578.97',
'73.0.3637.1',
'73.0.3637.0',
'72.0.3626.14',
'71.0.3578.96',
'71.0.3578.95',
'72.0.3626.13',
'71.0.3578.94',
'73.0.3636.2',
'71.0.3578.93',
'73.0.3636.1',
'73.0.3636.0',
'72.0.3626.12',
'71.0.3578.92',
'73.0.3635.1',
'73.0.3635.0',
'72.0.3626.11',
'71.0.3578.91',
'73.0.3634.2',
'73.0.3634.1',
'73.0.3634.0',
'72.0.3626.10',
'71.0.3578.90',
'71.0.3578.89',
'73.0.3633.2',
'73.0.3633.1',
'73.0.3633.0',
'72.0.3610.4',
'72.0.3626.9',
'71.0.3578.88',
'73.0.3632.5',
'73.0.3632.4',
'73.0.3632.3',
'73.0.3632.2',
'73.0.3632.1',
'73.0.3632.0',
'72.0.3626.8',
'71.0.3578.87',
'73.0.3631.2',
'73.0.3631.1',
'73.0.3631.0',
'72.0.3626.7',
'71.0.3578.86',
'72.0.3626.6',
'73.0.3630.1',
'73.0.3630.0',
'72.0.3626.5',
'71.0.3578.85',
'72.0.3626.4',
'73.0.3628.3',
'73.0.3628.2',
'73.0.3629.1',
'73.0.3629.0',
'72.0.3626.3',
'71.0.3578.84',
'73.0.3628.1',
'73.0.3628.0',
'71.0.3578.83',
'73.0.3627.1',
'73.0.3627.0',
'72.0.3626.2',
'71.0.3578.82',
'71.0.3578.81',
'71.0.3578.80',
'72.0.3626.1',
'72.0.3626.0',
'71.0.3578.79',
'70.0.3538.124',
'71.0.3578.78',
'72.0.3623.4',
'72.0.3625.2',
'72.0.3625.1',
'72.0.3625.0',
'71.0.3578.77',
'70.0.3538.123',
'72.0.3624.4',
'72.0.3624.3',
'72.0.3624.2',
'71.0.3578.76',
'72.0.3624.1',
'72.0.3624.0',
'72.0.3623.3',
'71.0.3578.75',
'70.0.3538.122',
'71.0.3578.74',
'72.0.3623.2',
'72.0.3610.3',
'72.0.3623.1',
'72.0.3623.0',
'72.0.3622.3',
'72.0.3622.2',
'71.0.3578.73',
'70.0.3538.121',
'72.0.3622.1',
'72.0.3622.0',
'71.0.3578.72',
'70.0.3538.120',
'72.0.3621.1',
'72.0.3621.0',
'71.0.3578.71',
'70.0.3538.119',
'72.0.3620.1',
'72.0.3620.0',
'71.0.3578.70',
'70.0.3538.118',
'71.0.3578.69',
'72.0.3619.1',
'72.0.3619.0',
'71.0.3578.68',
'70.0.3538.117',
'71.0.3578.67',
'72.0.3618.1',
'72.0.3618.0',
'71.0.3578.66',
'70.0.3538.116',
'72.0.3617.1',
'72.0.3617.0',
'71.0.3578.65',
'70.0.3538.115',
'72.0.3602.3',
'71.0.3578.64',
'72.0.3616.1',
'72.0.3616.0',
'71.0.3578.63',
'70.0.3538.114',
'71.0.3578.62',
'72.0.3615.1',
'72.0.3615.0',
'71.0.3578.61',
'70.0.3538.113',
'72.0.3614.1',
'72.0.3614.0',
'71.0.3578.60',
'70.0.3538.112',
'72.0.3613.1',
'72.0.3613.0',
'71.0.3578.59',
'70.0.3538.111',
'72.0.3612.2',
'72.0.3612.1',
'72.0.3612.0',
'70.0.3538.110',
'71.0.3578.58',
'70.0.3538.109',
'72.0.3611.2',
'72.0.3611.1',
'72.0.3611.0',
'71.0.3578.57',
'70.0.3538.108',
'72.0.3610.2',
'71.0.3578.56',
'71.0.3578.55',
'72.0.3610.1',
'72.0.3610.0',
'71.0.3578.54',
'70.0.3538.107',
'71.0.3578.53',
'72.0.3609.3',
'71.0.3578.52',
'72.0.3609.2',
'71.0.3578.51',
'72.0.3608.5',
'72.0.3609.1',
'72.0.3609.0',
'71.0.3578.50',
'70.0.3538.106',
'72.0.3608.4',
'72.0.3608.3',
'72.0.3608.2',
'71.0.3578.49',
'72.0.3608.1',
'72.0.3608.0',
'70.0.3538.105',
'71.0.3578.48',
'72.0.3607.1',
'72.0.3607.0',
'71.0.3578.47',
'70.0.3538.104',
'72.0.3606.2',
'72.0.3606.1',
'72.0.3606.0',
'71.0.3578.46',
'70.0.3538.103',
'70.0.3538.102',
'72.0.3605.3',
'72.0.3605.2',
'72.0.3605.1',
'72.0.3605.0',
'71.0.3578.45',
'70.0.3538.101',
'71.0.3578.44',
'71.0.3578.43',
'70.0.3538.100',
'70.0.3538.99',
'71.0.3578.42',
'72.0.3604.1',
'72.0.3604.0',
'71.0.3578.41',
'70.0.3538.98',
'71.0.3578.40',
'72.0.3603.2',
'72.0.3603.1',
'72.0.3603.0',
'71.0.3578.39',
'70.0.3538.97',
'72.0.3602.2',
'71.0.3578.38',
'71.0.3578.37',
'72.0.3602.1',
'72.0.3602.0',
'71.0.3578.36',
'70.0.3538.96',
'72.0.3601.1',
'72.0.3601.0',
'71.0.3578.35',
'70.0.3538.95',
'72.0.3600.1',
'72.0.3600.0',
'71.0.3578.34',
'70.0.3538.94',
'72.0.3599.3',
'72.0.3599.2',
'72.0.3599.1',
'72.0.3599.0',
'71.0.3578.33',
'70.0.3538.93',
'72.0.3598.1',
'72.0.3598.0',
'71.0.3578.32',
'70.0.3538.87',
'72.0.3597.1',
'72.0.3597.0',
'72.0.3596.2',
'71.0.3578.31',
'70.0.3538.86',
'71.0.3578.30',
'71.0.3578.29',
'72.0.3596.1',
'72.0.3596.0',
'71.0.3578.28',
'70.0.3538.85',
'72.0.3595.2',
'72.0.3591.3',
'72.0.3595.1',
'72.0.3595.0',
'71.0.3578.27',
'70.0.3538.84',
'72.0.3594.1',
'72.0.3594.0',
'71.0.3578.26',
'70.0.3538.83',
'72.0.3593.2',
'72.0.3593.1',
'72.0.3593.0',
'71.0.3578.25',
'70.0.3538.82',
'72.0.3589.3',
'72.0.3592.2',
'72.0.3592.1',
'72.0.3592.0',
'71.0.3578.24',
'72.0.3589.2',
'70.0.3538.81',
'70.0.3538.80',
'72.0.3591.2',
'72.0.3591.1',
'72.0.3591.0',
'71.0.3578.23',
'70.0.3538.79',
'71.0.3578.22',
'72.0.3590.1',
'72.0.3590.0',
'71.0.3578.21',
'70.0.3538.78',
'70.0.3538.77',
'72.0.3589.1',
'72.0.3589.0',
'71.0.3578.20',
'70.0.3538.76',
'71.0.3578.19',
'70.0.3538.75',
'72.0.3588.1',
'72.0.3588.0',
'71.0.3578.18',
'70.0.3538.74',
'72.0.3586.2',
'72.0.3587.0',
'71.0.3578.17',
'70.0.3538.73',
'72.0.3586.1',
'72.0.3586.0',
'71.0.3578.16',
'70.0.3538.72',
'72.0.3585.1',
'72.0.3585.0',
'71.0.3578.15',
'70.0.3538.71',
'71.0.3578.14',
'72.0.3584.1',
'72.0.3584.0',
'71.0.3578.13',
'70.0.3538.70',
'72.0.3583.2',
'71.0.3578.12',
'72.0.3583.1',
'72.0.3583.0',
'71.0.3578.11',
'70.0.3538.69',
'71.0.3578.10',
'72.0.3582.0',
'72.0.3581.4',
'71.0.3578.9',
'70.0.3538.67',
'72.0.3581.3',
'72.0.3581.2',
'72.0.3581.1',
'72.0.3581.0',
'71.0.3578.8',
'70.0.3538.66',
'72.0.3580.1',
'72.0.3580.0',
'71.0.3578.7',
'70.0.3538.65',
'71.0.3578.6',
'72.0.3579.1',
'72.0.3579.0',
'71.0.3578.5',
'70.0.3538.64',
'71.0.3578.4',
'71.0.3578.3',
'71.0.3578.2',
'71.0.3578.1',
'71.0.3578.0',
'70.0.3538.63',
'69.0.3497.128',
'70.0.3538.62',
'70.0.3538.61',
'70.0.3538.60',
'70.0.3538.59',
'71.0.3577.1',
'71.0.3577.0',
'70.0.3538.58',
'69.0.3497.127',
'71.0.3576.2',
'71.0.3576.1',
'71.0.3576.0',
'70.0.3538.57',
'70.0.3538.56',
'71.0.3575.2',
'70.0.3538.55',
'69.0.3497.126',
'70.0.3538.54',
'71.0.3575.1',
'71.0.3575.0',
'71.0.3574.1',
'71.0.3574.0',
'70.0.3538.53',
'69.0.3497.125',
'70.0.3538.52',
'71.0.3573.1',
'71.0.3573.0',
'70.0.3538.51',
'69.0.3497.124',
'71.0.3572.1',
'71.0.3572.0',
'70.0.3538.50',
'69.0.3497.123',
'71.0.3571.2',
'70.0.3538.49',
'69.0.3497.122',
'71.0.3571.1',
'71.0.3571.0',
'70.0.3538.48',
'69.0.3497.121',
'71.0.3570.1',
'71.0.3570.0',
'70.0.3538.47',
'69.0.3497.120',
'71.0.3568.2',
'71.0.3569.1',
'71.0.3569.0',
'70.0.3538.46',
'69.0.3497.119',
'70.0.3538.45',
'71.0.3568.1',
'71.0.3568.0',
'70.0.3538.44',
'69.0.3497.118',
'70.0.3538.43',
'70.0.3538.42',
'71.0.3567.1',
'71.0.3567.0',
'70.0.3538.41',
'69.0.3497.117',
'71.0.3566.1',
'71.0.3566.0',
'70.0.3538.40',
'69.0.3497.116',
'71.0.3565.1',
'71.0.3565.0',
'70.0.3538.39',
'69.0.3497.115',
'71.0.3564.1',
'71.0.3564.0',
'70.0.3538.38',
'69.0.3497.114',
'71.0.3563.0',
'71.0.3562.2',
'70.0.3538.37',
'69.0.3497.113',
'70.0.3538.36',
'70.0.3538.35',
'71.0.3562.1',
'71.0.3562.0',
'70.0.3538.34',
'69.0.3497.112',
'70.0.3538.33',
'71.0.3561.1',
'71.0.3561.0',
'70.0.3538.32',
'69.0.3497.111',
'71.0.3559.6',
'71.0.3560.1',
'71.0.3560.0',
'71.0.3559.5',
'71.0.3559.4',
'70.0.3538.31',
'69.0.3497.110',
'71.0.3559.3',
'70.0.3538.30',
'69.0.3497.109',
'71.0.3559.2',
'71.0.3559.1',
'71.0.3559.0',
'70.0.3538.29',
'69.0.3497.108',
'71.0.3558.2',
'71.0.3558.1',
'71.0.3558.0',
'70.0.3538.28',
'69.0.3497.107',
'71.0.3557.2',
'71.0.3557.1',
'71.0.3557.0',
'70.0.3538.27',
'69.0.3497.106',
'71.0.3554.4',
'70.0.3538.26',
'71.0.3556.1',
'71.0.3556.0',
'70.0.3538.25',
'71.0.3554.3',
'69.0.3497.105',
'71.0.3554.2',
'70.0.3538.24',
'69.0.3497.104',
'71.0.3555.2',
'70.0.3538.23',
'71.0.3555.1',
'71.0.3555.0',
'70.0.3538.22',
'69.0.3497.103',
'71.0.3554.1',
'71.0.3554.0',
'70.0.3538.21',
'69.0.3497.102',
'71.0.3553.3',
'70.0.3538.20',
'69.0.3497.101',
'71.0.3553.2',
'69.0.3497.100',
'71.0.3553.1',
'71.0.3553.0',
'70.0.3538.19',
'69.0.3497.99',
'69.0.3497.98',
'69.0.3497.97',
'71.0.3552.6',
'71.0.3552.5',
'71.0.3552.4',
'71.0.3552.3',
'71.0.3552.2',
'71.0.3552.1',
'71.0.3552.0',
'70.0.3538.18',
'69.0.3497.96',
'71.0.3551.3',
'71.0.3551.2',
'71.0.3551.1',
'71.0.3551.0',
'70.0.3538.17',
'69.0.3497.95',
'71.0.3550.3',
'71.0.3550.2',
'71.0.3550.1',
'71.0.3550.0',
'70.0.3538.16',
'69.0.3497.94',
'71.0.3549.1',
'71.0.3549.0',
'70.0.3538.15',
'69.0.3497.93',
'69.0.3497.92',
'71.0.3548.1',
'71.0.3548.0',
'70.0.3538.14',
'69.0.3497.91',
'71.0.3547.1',
'71.0.3547.0',
'70.0.3538.13',
'69.0.3497.90',
'71.0.3546.2',
'69.0.3497.89',
'71.0.3546.1',
'71.0.3546.0',
'70.0.3538.12',
'69.0.3497.88',
'71.0.3545.4',
'71.0.3545.3',
'71.0.3545.2',
'71.0.3545.1',
'71.0.3545.0',
'70.0.3538.11',
'69.0.3497.87',
'71.0.3544.5',
'71.0.3544.4',
'71.0.3544.3',
'71.0.3544.2',
'71.0.3544.1',
'71.0.3544.0',
'69.0.3497.86',
'70.0.3538.10',
'69.0.3497.85',
'70.0.3538.9',
'69.0.3497.84',
'71.0.3543.4',
'70.0.3538.8',
'71.0.3543.3',
'71.0.3543.2',
'71.0.3543.1',
'71.0.3543.0',
'70.0.3538.7',
'69.0.3497.83',
'71.0.3542.2',
'71.0.3542.1',
'71.0.3542.0',
'70.0.3538.6',
'69.0.3497.82',
'69.0.3497.81',
'71.0.3541.1',
'71.0.3541.0',
'70.0.3538.5',
'69.0.3497.80',
'71.0.3540.1',
'71.0.3540.0',
'70.0.3538.4',
'69.0.3497.79',
'70.0.3538.3',
'71.0.3539.1',
'71.0.3539.0',
'69.0.3497.78',
'68.0.3440.134',
'69.0.3497.77',
'70.0.3538.2',
'70.0.3538.1',
'70.0.3538.0',
'69.0.3497.76',
'68.0.3440.133',
'69.0.3497.75',
'70.0.3537.2',
'70.0.3537.1',
'70.0.3537.0',
'69.0.3497.74',
'68.0.3440.132',
'70.0.3536.0',
'70.0.3535.5',
'70.0.3535.4',
'70.0.3535.3',
'69.0.3497.73',
'68.0.3440.131',
'70.0.3532.8',
'70.0.3532.7',
'69.0.3497.72',
'69.0.3497.71',
'70.0.3535.2',
'70.0.3535.1',
'70.0.3535.0',
'69.0.3497.70',
'68.0.3440.130',
'69.0.3497.69',
'68.0.3440.129',
'70.0.3534.4',
'70.0.3534.3',
'70.0.3534.2',
'70.0.3534.1',
'70.0.3534.0',
'69.0.3497.68',
'68.0.3440.128',
'70.0.3533.2',
'70.0.3533.1',
'70.0.3533.0',
'69.0.3497.67',
'68.0.3440.127',
'70.0.3532.6',
'70.0.3532.5',
'70.0.3532.4',
'69.0.3497.66',
'68.0.3440.126',
'70.0.3532.3',
'70.0.3532.2',
'70.0.3532.1',
'69.0.3497.60',
'69.0.3497.65',
'69.0.3497.64',
'70.0.3532.0',
'70.0.3531.0',
'70.0.3530.4',
'70.0.3530.3',
'70.0.3530.2',
'69.0.3497.58',
'68.0.3440.125',
'69.0.3497.57',
'69.0.3497.56',
'69.0.3497.55',
'69.0.3497.54',
'70.0.3530.1',
'70.0.3530.0',
'69.0.3497.53',
'68.0.3440.124',
'69.0.3497.52',
'70.0.3529.3',
'70.0.3529.2',
'70.0.3529.1',
'70.0.3529.0',
'69.0.3497.51',
'70.0.3528.4',
'68.0.3440.123',
'70.0.3528.3',
'70.0.3528.2',
'70.0.3528.1',
'70.0.3528.0',
'69.0.3497.50',
'68.0.3440.122',
'70.0.3527.1',
'70.0.3527.0',
'69.0.3497.49',
'68.0.3440.121',
'70.0.3526.1',
'70.0.3526.0',
'68.0.3440.120',
'69.0.3497.48',
'69.0.3497.47',
'68.0.3440.119',
'68.0.3440.118',
'70.0.3525.5',
'70.0.3525.4',
'70.0.3525.3',
'68.0.3440.117',
'69.0.3497.46',
'70.0.3525.2',
'70.0.3525.1',
'70.0.3525.0',
'69.0.3497.45',
'68.0.3440.116',
'70.0.3524.4',
'70.0.3524.3',
'69.0.3497.44',
'70.0.3524.2',
'70.0.3524.1',
'70.0.3524.0',
'70.0.3523.2',
'69.0.3497.43',
'68.0.3440.115',
'70.0.3505.9',
'69.0.3497.42',
'70.0.3505.8',
'70.0.3523.1',
'70.0.3523.0',
'69.0.3497.41',
'68.0.3440.114',
'70.0.3505.7',
'69.0.3497.40',
'70.0.3522.1',
'70.0.3522.0',
'70.0.3521.2',
'69.0.3497.39',
'68.0.3440.113',
'70.0.3505.6',
'70.0.3521.1',
'70.0.3521.0',
'69.0.3497.38',
'68.0.3440.112',
'70.0.3520.1',
'70.0.3520.0',
'69.0.3497.37',
'68.0.3440.111',
'70.0.3519.3',
'70.0.3519.2',
'70.0.3519.1',
'70.0.3519.0',
'69.0.3497.36',
'68.0.3440.110',
'70.0.3518.1',
'70.0.3518.0',
'69.0.3497.35',
'69.0.3497.34',
'68.0.3440.109',
'70.0.3517.1',
'70.0.3517.0',
'69.0.3497.33',
'68.0.3440.108',
'69.0.3497.32',
'70.0.3516.3',
'70.0.3516.2',
'70.0.3516.1',
'70.0.3516.0',
'69.0.3497.31',
'68.0.3440.107',
'70.0.3515.4',
'68.0.3440.106',
'70.0.3515.3',
'70.0.3515.2',
'70.0.3515.1',
'70.0.3515.0',
'69.0.3497.30',
'68.0.3440.105',
'68.0.3440.104',
'70.0.3514.2',
'70.0.3514.1',
'70.0.3514.0',
'69.0.3497.29',
'68.0.3440.103',
'70.0.3513.1',
'70.0.3513.0',
'69.0.3497.28',
)
return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS)
std_headers = {
'User-Agent': random_user_agent(),
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'en-us,en;q=0.5',
}
USER_AGENTS = {
'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27',
}
NO_DEFAULT = object()
IDENTITY = lambda x: x
ENGLISH_MONTH_NAMES = [
'January', 'February', 'March', 'April', 'May', 'June',
'July', 'August', 'September', 'October', 'November', 'December']
MONTH_NAMES = {
'en': ENGLISH_MONTH_NAMES,
'fr': [
'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
}
# Timezone names for RFC2822 obs-zone
# From https://github.com/python/cpython/blob/3.11/Lib/email/_parseaddr.py#L36-L42
TIMEZONE_NAMES = {
'UT': 0, 'UTC': 0, 'GMT': 0, 'Z': 0,
'AST': -4, 'ADT': -3, # Atlantic (used in Canada)
'EST': -5, 'EDT': -4, # Eastern
'CST': -6, 'CDT': -5, # Central
'MST': -7, 'MDT': -6, # Mountain
'PST': -8, 'PDT': -7 # Pacific
}
KNOWN_EXTENSIONS = (
'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
'flv', 'f4v', 'f4a', 'f4b',
'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
'mkv', 'mka', 'mk3d',
'avi', 'divx',
'mov',
'asf', 'wmv', 'wma',
'3gp', '3g2',
'mp3',
'flac',
'ape',
'wav',
'f4f', 'f4m', 'm3u8', 'smil')
# needed for sanitizing filenames in restricted mode
ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
DATE_FORMATS = (
'%d %B %Y',
'%d %b %Y',
'%B %d %Y',
'%B %dst %Y',
'%B %dnd %Y',
'%B %drd %Y',
'%B %dth %Y',
'%b %d %Y',
'%b %dst %Y',
'%b %dnd %Y',
'%b %drd %Y',
'%b %dth %Y',
'%b %dst %Y %I:%M',
'%b %dnd %Y %I:%M',
'%b %drd %Y %I:%M',
'%b %dth %Y %I:%M',
'%Y %m %d',
'%Y-%m-%d',
'%Y.%m.%d.',
'%Y/%m/%d',
'%Y/%m/%d %H:%M',
'%Y/%m/%d %H:%M:%S',
'%Y%m%d%H%M',
'%Y%m%d%H%M%S',
'%Y%m%d',
'%Y-%m-%d %H:%M',
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
'%Y-%m-%d %H:%M:%S:%f',
'%d.%m.%Y %H:%M',
'%d.%m.%Y %H.%M',
'%Y-%m-%dT%H:%M:%SZ',
'%Y-%m-%dT%H:%M:%S.%fZ',
'%Y-%m-%dT%H:%M:%S.%f0Z',
'%Y-%m-%dT%H:%M:%S',
'%Y-%m-%dT%H:%M:%S.%f',
'%Y-%m-%dT%H:%M',
'%b %d %Y at %H:%M',
'%b %d %Y at %H:%M:%S',
'%B %d %Y at %H:%M',
'%B %d %Y at %H:%M:%S',
'%H:%M %d-%b-%Y',
)
DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
DATE_FORMATS_DAY_FIRST.extend([
'%d-%m-%Y',
'%d.%m.%Y',
'%d.%m.%y',
'%d/%m/%Y',
'%d/%m/%y',
'%d/%m/%Y %H:%M:%S',
'%d-%m-%Y %H:%M',
])
DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS)
DATE_FORMATS_MONTH_FIRST.extend([
'%m-%d-%Y',
'%m.%d.%Y',
'%m/%d/%Y',
'%m/%d/%y',
'%m/%d/%Y %H:%M:%S',
])
PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
def preferredencoding():
"""Get preferred encoding.
Returns the best encoding scheme for the system, based on
locale.getpreferredencoding() and some further tweaks.
"""
try:
pref = locale.getpreferredencoding()
'TEST'.encode(pref)
except Exception:
pref = 'UTF-8'
return pref
def write_json_file(obj, fn):
""" Encode obj as JSON and write it to fn, atomically if possible """
fn = encodeFilename(fn)
if sys.version_info < (3, 0) and sys.platform != 'win32':
encoding = get_filesystem_encoding()
# os.path.basename returns a bytes object, but NamedTemporaryFile
# will fail if the filename contains non-ascii characters unless we
# use a unicode object
path_basename = lambda f: os.path.basename(f).decode(encoding)
# the same for os.path.dirname
path_dirname = lambda f: os.path.dirname(f).decode(encoding)
else:
path_basename = os.path.basename
path_dirname = os.path.dirname
args = {
'suffix': '.tmp',
'prefix': path_basename(fn) + '.',
'dir': path_dirname(fn),
'delete': False,
}
# In Python 2.x, json.dump expects a bytestream.
# In Python 3.x, it writes to a character stream
if sys.version_info < (3, 0):
args['mode'] = 'wb'
else:
args.update({
'mode': 'w',
'encoding': 'utf-8',
})
tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
try:
with tf:
json.dump(obj, tf)
if sys.platform == 'win32':
# Need to remove existing file on Windows, else os.rename raises
# WindowsError or FileExistsError.
try:
os.unlink(fn)
except OSError:
pass
try:
mask = os.umask(0)
os.umask(mask)
os.chmod(tf.name, 0o666 & ~mask)
except OSError:
pass
os.rename(tf.name, fn)
except Exception:
try:
os.remove(tf.name)
except OSError:
pass
raise
if sys.version_info >= (2, 7):
def find_xpath_attr(node, xpath, key, val=None):
""" Find the xpath xpath[@key=val] """
assert re.match(r'^[a-zA-Z_-]+$', key)
expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
return node.find(expr)
else:
def find_xpath_attr(node, xpath, key, val=None):
for f in node.findall(compat_xpath(xpath)):
if key not in f.attrib:
continue
if val is None or f.attrib.get(key) == val:
return f
return None
# On python2.6 the xml.etree.ElementTree.Element methods don't support
# the namespace parameter
def xpath_with_ns(path, ns_map):
components = [c.split(':') for c in path.split('/')]
replaced = []
for c in components:
if len(c) == 1:
replaced.append(c[0])
else:
ns, tag = c
replaced.append('{%s}%s' % (ns_map[ns], tag))
return '/'.join(replaced)
def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
def _find_xpath(xpath):
return node.find(compat_xpath(xpath))
if isinstance(xpath, compat_basestring):
n = _find_xpath(xpath)
else:
for xp in xpath:
n = _find_xpath(xp)
if n is not None:
break
if n is None:
if default is not NO_DEFAULT:
return default
elif fatal:
name = xpath if name is None else name
raise ExtractorError('Could not find XML element %s' % name)
else:
return None
return n
def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
n = xpath_element(node, xpath, name, fatal=fatal, default=default)
if n is None or n == default:
return n
if n.text is None:
if default is not NO_DEFAULT:
return default
elif fatal:
name = xpath if name is None else name
raise ExtractorError('Could not find XML element\'s text %s' % name)
else:
return None
return n.text
def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT):
n = find_xpath_attr(node, xpath, key)
if n is None:
if default is not NO_DEFAULT:
return default
elif fatal:
name = '%s[@%s]' % (xpath, key) if name is None else name
raise ExtractorError('Could not find XML attribute %s' % name)
else:
return None
return n.attrib[key]
def get_element_by_id(id, html):
"""Return the content of the tag with the specified ID in the passed HTML document"""
return get_element_by_attribute('id', id, html)
def get_element_by_class(class_name, html):
"""Return the content of the first tag with the specified class in the passed HTML document"""
retval = get_elements_by_class(class_name, html)
return retval[0] if retval else None
def get_element_by_attribute(attribute, value, html, escape_value=True):
retval = get_elements_by_attribute(attribute, value, html, escape_value)
return retval[0] if retval else None
def get_elements_by_class(class_name, html):
"""Return the content of all tags with the specified class in the passed HTML document as a list"""
return get_elements_by_attribute(
'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
html, escape_value=False)
def get_elements_by_attribute(attribute, value, html, escape_value=True):
"""Return the content of the tag with the specified attribute in the passed HTML document"""
value = re.escape(value) if escape_value else value
retlist = []
for m in re.finditer(r'''(?xs)
<([a-zA-Z0-9:._-]+)
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s+%s=['"]?%s['"]?
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s*>
(?P<content>.*?)
</\1>
''' % (re.escape(attribute), value), html):
res = m.group('content')
if res.startswith('"') or res.startswith("'"):
res = res[1:-1]
retlist.append(unescapeHTML(res))
return retlist
class HTMLAttributeParser(compat_HTMLParser):
"""Trivial HTML parser to gather the attributes for a single element"""
def __init__(self):
self.attrs = {}
compat_HTMLParser.__init__(self)
def handle_starttag(self, tag, attrs):
self.attrs = dict(attrs)
def extract_attributes(html_element):
"""Given a string for an HTML element such as
<el
a="foo" B="bar" c="&98;az" d=boz
empty= noval entity="&amp;"
sq='"' dq="'"
>
Decode and return a dictionary of attributes.
{
'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz',
'empty': '', 'noval': None, 'entity': '&',
'sq': '"', 'dq': '\''
}.
NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
"""
parser = HTMLAttributeParser()
try:
parser.feed(html_element)
parser.close()
# Older Python may throw HTMLParseError in case of malformed HTML
except compat_HTMLParseError:
pass
return parser.attrs
def clean_html(html):
"""Clean an HTML snippet into a readable string"""
if html is None: # Convenience for sanitizing descriptions etc.
return html
# Newline vs <br />
html = html.replace('\n', ' ')
html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
# Strip html tags
html = re.sub('<.*?>', '', html)
# Replace html entities
html = unescapeHTML(html)
return html.strip()
def sanitize_open(filename, open_mode):
"""Try to open the given filename, and slightly tweak it if this fails.
Attempts to open the given filename. If this fails, it tries to change
the filename slightly, step by step, until it's either able to open it
or it fails and raises a final exception, like the standard open()
function.
It returns the tuple (stream, definitive_file_name).
"""
try:
if filename == '-':
if sys.platform == 'win32':
import msvcrt
msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
stream = open(encodeFilename(filename), open_mode)
return (stream, filename)
except (IOError, OSError) as err:
if err.errno in (errno.EACCES,):
raise
# In case of error, try to remove win32 forbidden chars
alt_filename = sanitize_path(filename)
if alt_filename == filename:
raise
else:
# An exception here should be caught in the caller
stream = open(encodeFilename(alt_filename), open_mode)
return (stream, alt_filename)
def timeconvert(timestr):
"""Convert RFC 2822 defined time string into system timestamp"""
timestamp = None
timetuple = email.utils.parsedate_tz(timestr)
if timetuple is not None:
timestamp = email.utils.mktime_tz(timetuple)
return timestamp
def sanitize_filename(s, restricted=False, is_id=False):
"""Sanitizes a string so it could be used as part of a filename.
If restricted is set, use a stricter subset of allowed characters.
Set is_id if this is not an arbitrary string, but an ID that should be kept
if possible.
"""
def replace_insane(char):
if restricted and char in ACCENT_CHARS:
return ACCENT_CHARS[char]
if char == '?' or ord(char) < 32 or ord(char) == 127:
return ''
elif char == '"':
return '' if restricted else '\''
elif char == ':':
return '_-' if restricted else ' -'
elif char in '\\/|*<>':
return '_'
if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()):
return '_'
if restricted and ord(char) > 127:
return '' if unicodedata.category(char)[0] in 'CM' else '_'
return char
# Replace look-alike Unicode glyphs
if restricted and not is_id:
s = unicodedata.normalize('NFKC', s)
# Handle timestamps
s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
result = ''.join(map(replace_insane, s))
if not is_id:
while '__' in result:
result = result.replace('__', '_')
result = result.strip('_')
# Common case of "Foreign band name - English song title"
if restricted and result.startswith('-_'):
result = result[2:]
if result.startswith('-'):
result = '_' + result[len('-'):]
result = result.lstrip('.')
if not result:
result = '_'
return result
def sanitize_path(s):
"""Sanitizes and normalizes path on Windows"""
if sys.platform != 'win32':
return s
drive_or_unc, _ = os.path.splitdrive(s)
if sys.version_info < (2, 7) and not drive_or_unc:
drive_or_unc, _ = os.path.splitunc(s)
norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep)
if drive_or_unc:
norm_path.pop(0)
sanitized_path = [
path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
for path_part in norm_path]
if drive_or_unc:
sanitized_path.insert(0, drive_or_unc + os.path.sep)
return os.path.join(*sanitized_path)
def sanitize_url(url):
# Prepend protocol-less URLs with `http:` scheme in order to mitigate
# the number of unwanted failures due to missing protocol
if url.startswith('//'):
return 'http:%s' % url
# Fix some common typos seen so far
COMMON_TYPOS = (
# https://github.com/ytdl-org/youtube-dl/issues/15649
(r'^httpss://', r'https://'),
# https://bx1.be/lives/direct-tv/
(r'^rmtp([es]?)://', r'rtmp\1://'),
)
for mistake, fixup in COMMON_TYPOS:
if re.match(mistake, url):
return re.sub(mistake, fixup, url)
return url
def extract_basic_auth(url):
parts = compat_urllib_parse.urlsplit(url)
if parts.username is None:
return url, None
url = compat_urllib_parse.urlunsplit(parts._replace(netloc=(
parts.hostname if parts.port is None
else '%s:%d' % (parts.hostname, parts.port))))
auth_payload = base64.b64encode(
('%s:%s' % (parts.username, parts.password or '')).encode('utf-8'))
return url, 'Basic {0}'.format(auth_payload.decode('ascii'))
def sanitized_Request(url, *args, **kwargs):
url, auth_header = extract_basic_auth(escape_url(sanitize_url(url)))
if auth_header is not None:
headers = args[1] if len(args) > 1 else kwargs.get('headers')
headers = headers or {}
headers['Authorization'] = auth_header
if len(args) <= 1 and kwargs.get('headers') is None:
kwargs['headers'] = headers
kwargs = compat_kwargs(kwargs)
return compat_urllib_request.Request(url, *args, **kwargs)
def expand_path(s):
"""Expand shell variables and ~"""
return os.path.expandvars(compat_expanduser(s))
def orderedSet(iterable):
""" Remove all duplicates from the input iterable """
res = []
for el in iterable:
if el not in res:
res.append(el)
return res
def _htmlentity_transform(entity_with_semicolon):
"""Transforms an HTML entity to a character."""
entity = entity_with_semicolon[:-1]
# Known non-numeric HTML entity
if entity in compat_html_entities.name2codepoint:
return compat_chr(compat_html_entities.name2codepoint[entity])
# TODO: HTML5 allows entities without a semicolon. For example,
# '&Eacuteric' should be decoded as 'Éric'.
if entity_with_semicolon in compat_html_entities_html5:
return compat_html_entities_html5[entity_with_semicolon]
mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity)
if mobj is not None:
numstr = mobj.group(1)
if numstr.startswith('x'):
base = 16
numstr = '0%s' % numstr
else:
base = 10
# See https://github.com/ytdl-org/youtube-dl/issues/7518
try:
return compat_chr(int(numstr, base))
except ValueError:
pass
# Unknown entity in name, return its literal representation
return '&%s;' % entity
def unescapeHTML(s):
if s is None:
return None
assert isinstance(s, compat_str)
return re.sub(
r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
def process_communicate_or_kill(p, *args, **kwargs):
try:
return p.communicate(*args, **kwargs)
except BaseException: # Including KeyboardInterrupt
p.kill()
p.wait()
raise
def get_subprocess_encoding():
if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
# For subprocess calls, encode with locale encoding
# Refer to http://stackoverflow.com/a/9951851/35070
encoding = preferredencoding()
else:
encoding = sys.getfilesystemencoding()
if encoding is None:
encoding = 'utf-8'
return encoding
# Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
if sys.version_info < (3, 0) and not sys.platform.startswith('java'):
def encodeFilename(s, for_subprocess=False):
"""
@param s The name of the file
"""
# Pass '' directly to use Unicode APIs on Windows 2000 and up
# (Detecting Windows NT 4 is tricky because 'major >= 4' would
# match Windows 9x series as well. Besides, NT 4 is obsolete.)
if (not for_subprocess
and sys.platform == 'win32'
and sys.getwindowsversion()[0] >= 5
and isinstance(s, compat_str)):
return s
return _encode_compat_str(s, get_subprocess_encoding(), 'ignore')
def decodeFilename(b, for_subprocess=False):
return _decode_compat_str(b, get_subprocess_encoding(), 'ignore')
else:
# Python 3 has a Unicode API
encodeFilename = decodeFilename = lambda *s, **k: s[0]
def encodeArgument(s):
if not isinstance(s, compat_str):
# Legacy code that uses byte strings
# Uncomment the following line after fixing all post processors
# assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
s = s.decode('ascii')
return encodeFilename(s, True)
def decodeArgument(b):
return decodeFilename(b, True)
def decodeOption(optval):
if optval is None:
return optval
return _decode_compat_str(optval)
def formatSeconds(secs):
if secs > 3600:
return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60)
elif secs > 60:
return '%d:%02d' % (secs // 60, secs % 60)
else:
return '%d' % secs
def make_HTTPS_handler(params, **kwargs):
# https://www.rfc-editor.org/info/rfc7301
ALPN_PROTOCOLS = ['http/1.1']
def set_alpn_protocols(ctx):
# From https://github.com/yt-dlp/yt-dlp/commit/2c6dcb65fb612fc5bc5c61937bf438d3c473d8d0
# Thanks @coletdjnz
# Some servers may (wrongly) reject requests if ALPN extension is not sent. See:
# https://github.com/python/cpython/issues/85140
# https://github.com/yt-dlp/yt-dlp/issues/3878
try:
ctx.set_alpn_protocols(ALPN_PROTOCOLS)
except (AttributeError, NotImplementedError):
# Python < 2.7.10, not ssl.HAS_ALPN
pass
opts_no_check_certificate = params.get('nocheckcertificate', False)
if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9
context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH)
set_alpn_protocols(context)
if opts_no_check_certificate:
context.check_hostname = False
context.verify_mode = ssl.CERT_NONE
try:
return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
except TypeError:
# Python 2.7.8
# (create_default_context present but HTTPSHandler has no context=)
pass
if sys.version_info < (3, 2):
return YoutubeDLHTTPSHandler(params, **kwargs)
else: # Python3 < 3.4
context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
context.verify_mode = (ssl.CERT_NONE
if opts_no_check_certificate
else ssl.CERT_REQUIRED)
context.set_default_verify_paths()
set_alpn_protocols(context)
return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
def bug_reports_message():
if ytdl_is_updateable():
update_cmd = 'type youtube-dl -U to update'
else:
update_cmd = 'see https://yt-dl.org/update on how to update'
msg = '; please report this issue on https://yt-dl.org/bug .'
msg += ' Make sure you are using the latest version; %s.' % update_cmd
msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.'
return msg
class YoutubeDLError(Exception):
"""Base exception for YoutubeDL errors."""
pass
class ExtractorError(YoutubeDLError):
"""Error during info extraction."""
def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
""" tb, if given, is the original traceback (so that it can be printed out).
If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
"""
if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
expected = True
if video_id is not None:
msg = video_id + ': ' + msg
if cause:
msg += ' (caused by %r)' % cause
if not expected:
msg += bug_reports_message()
super(ExtractorError, self).__init__(msg)
self.traceback = tb
self.exc_info = sys.exc_info() # preserve original exception
self.cause = cause
self.video_id = video_id
def format_traceback(self):
if self.traceback is None:
return None
return ''.join(traceback.format_tb(self.traceback))
class UnsupportedError(ExtractorError):
def __init__(self, url):
super(UnsupportedError, self).__init__(
'Unsupported URL: %s' % url, expected=True)
self.url = url
class RegexNotFoundError(ExtractorError):
"""Error when a regex didn't match"""
pass
class GeoRestrictedError(ExtractorError):
"""Geographic restriction Error exception.
This exception may be thrown when a video is not available from your
geographic location due to geographic restrictions imposed by a website.
"""
def __init__(self, msg, countries=None):
super(GeoRestrictedError, self).__init__(msg, expected=True)
self.msg = msg
self.countries = countries
class DownloadError(YoutubeDLError):
"""Download Error exception.
This exception may be thrown by FileDownloader objects if they are not
configured to continue on errors. They will contain the appropriate
error message.
"""
def __init__(self, msg, exc_info=None):
""" exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
super(DownloadError, self).__init__(msg)
self.exc_info = exc_info
class SameFileError(YoutubeDLError):
"""Same File exception.
This exception will be thrown by FileDownloader objects if they detect
multiple files would have to be downloaded to the same file on disk.
"""
pass
class PostProcessingError(YoutubeDLError):
"""Post Processing exception.
This exception may be raised by PostProcessor's .run() method to
indicate an error in the postprocessing task.
"""
def __init__(self, msg):
super(PostProcessingError, self).__init__(msg)
self.msg = msg
class MaxDownloadsReached(YoutubeDLError):
""" --max-downloads limit has been reached. """
pass
class UnavailableVideoError(YoutubeDLError):
"""Unavailable Format exception.
This exception will be thrown when a video is requested
in a format that is not available for that video.
"""
pass
class ContentTooShortError(YoutubeDLError):
"""Content Too Short exception.
This exception may be raised by FileDownloader objects when a file they
download is too small for what the server announced first, indicating
the connection was probably interrupted.
"""
def __init__(self, downloaded, expected):
super(ContentTooShortError, self).__init__(
'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
)
# Both in bytes
self.downloaded = downloaded
self.expected = expected
class XAttrMetadataError(YoutubeDLError):
def __init__(self, code=None, msg='Unknown error'):
super(XAttrMetadataError, self).__init__(msg)
self.code = code
self.msg = msg
# Parsing code and msg
if (self.code in (errno.ENOSPC, errno.EDQUOT)
or 'No space left' in self.msg or 'Disk quota exceeded' in self.msg):
self.reason = 'NO_SPACE'
elif self.code == errno.E2BIG or 'Argument list too long' in self.msg:
self.reason = 'VALUE_TOO_LONG'
else:
self.reason = 'NOT_SUPPORTED'
class XAttrUnavailableError(YoutubeDLError):
pass
def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
# Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
# expected HTTP responses to meet HTTP/1.0 or later (see also
# https://github.com/ytdl-org/youtube-dl/issues/6727)
if sys.version_info < (3, 0):
kwargs['strict'] = True
hc = http_class(*args, **compat_kwargs(kwargs))
source_address = ydl_handler._params.get('source_address')
if source_address is not None:
# This is to workaround _create_connection() from socket where it will try all
# address data from getaddrinfo() including IPv6. This filters the result from
# getaddrinfo() based on the source_address value.
# This is based on the cpython socket.create_connection() function.
# https://github.com/python/cpython/blob/master/Lib/socket.py#L691
def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
host, port = address
err = None
addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6
ip_addrs = [addr for addr in addrs if addr[0] == af]
if addrs and not ip_addrs:
ip_version = 'v4' if af == socket.AF_INET else 'v6'
raise socket.error(
"No remote IP%s addresses available for connect, can't use '%s' as source address"
% (ip_version, source_address[0]))
for res in ip_addrs:
af, socktype, proto, canonname, sa = res
sock = None
try:
sock = socket.socket(af, socktype, proto)
if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
sock.settimeout(timeout)
sock.bind(source_address)
sock.connect(sa)
err = None # Explicitly break reference cycle
return sock
except socket.error as _:
err = _
if sock is not None:
sock.close()
if err is not None:
raise err
else:
raise socket.error('getaddrinfo returns an empty list')
if hasattr(hc, '_create_connection'):
hc._create_connection = _create_connection
sa = (source_address, 0)
if hasattr(hc, 'source_address'): # Python 2.7+
hc.source_address = sa
else: # Python 2.6
def _hc_connect(self, *args, **kwargs):
sock = _create_connection(
(self.host, self.port), self.timeout, sa)
if is_https:
self.sock = ssl.wrap_socket(
sock, self.key_file, self.cert_file,
ssl_version=ssl.PROTOCOL_TLSv1)
else:
self.sock = sock
hc.connect = functools.partial(_hc_connect, hc)
return hc
def handle_youtubedl_headers(headers):
filtered_headers = headers
if 'Youtubedl-no-compression' in filtered_headers:
filtered_headers = filter_dict(filtered_headers, cndn=lambda k, _: k.lower() != 'accept-encoding')
del filtered_headers['Youtubedl-no-compression']
return filtered_headers
class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
"""Handler for HTTP requests and responses.
This class, when installed with an OpenerDirector, automatically adds
the standard headers to every HTTP request and handles gzipped and
deflated responses from web servers. If compression is to be avoided in
a particular request, the original request in the program code only has
to include the HTTP header "Youtubedl-no-compression", which will be
removed before making the real request.
Part of this code was copied from:
http://techknack.net/python-urllib2-handlers/, archived at
https://web.archive.org/web/20130527205558/http://techknack.net/python-urllib2-handlers/
Andrew Rowls, the author of that code, agreed to release it to the
public domain.
"""
def __init__(self, params, *args, **kwargs):
compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs)
self._params = params
def http_open(self, req):
conn_class = compat_http_client.HTTPConnection
socks_proxy = req.headers.get('Ytdl-socks-proxy')
if socks_proxy:
conn_class = make_socks_conn_class(conn_class, socks_proxy)
del req.headers['Ytdl-socks-proxy']
return self.do_open(functools.partial(
_create_http_connection, self, conn_class, False),
req)
@staticmethod
def deflate_gz(data):
try:
# format:zlib,gzip + windowsize:32768
return data and zlib.decompress(data, 32 + zlib.MAX_WBITS)
except zlib.error:
# raw zlib * windowsize:32768 (RFC 9110: "non-conformant")
return zlib.decompress(data, -zlib.MAX_WBITS)
@staticmethod
def gzip(data):
from gzip import GzipFile
def _gzip(data):
with io.BytesIO(data) as data_buf:
gz = GzipFile(fileobj=data_buf, mode='rb')
return gz.read()
try:
return _gzip(data)
except IOError as original_ioerror:
# There may be junk at the end of the file
# See http://stackoverflow.com/q/4928560/35070 for details
for i in range(1, 1024):
try:
return _gzip(data[:-i])
except IOError:
continue
else:
raise original_ioerror
@staticmethod
def brotli(data):
return data and brotli.decompress(data)
@staticmethod
def compress(data):
return data and ncompress.decompress(data)
@staticmethod
def _fix_path(url):
# an embedded /../ or /./ sequence is not automatically handled by urllib2
# see https://github.com/yt-dlp/yt-dlp/issues/3355
parsed_url = compat_urllib_parse.urlsplit(url)
path = parsed_url.path
if not path.endswith('/'):
path += '/'
parts = path.partition('/./')
if not parts[1]:
parts = path.partition('/../')
if parts[1]:
path = compat_urllib_parse.urljoin(
parts[0] + parts[1][:1],
parts[1][1:] + (parts[2] if parsed_url.path.endswith('/') else parts[2][:-1]))
url = parsed_url._replace(path=path).geturl()
if '/.' in url:
# worse, URL path may have initial /../ against RFCs: work-around
# by stripping such prefixes, like eg Firefox
path = parsed_url.path + '/'
while path.startswith('/.'):
if path.startswith('/../'):
path = path[3:]
elif path.startswith('/./'):
path = path[2:]
else:
break
path = path[:-1]
if not path.startswith('/') and parsed_url.path.startswith('/'):
path = '/' + path
url = parsed_url._replace(path=path).geturl()
return url
def http_request(self, req):
url = req.get_full_url()
# resolve embedded . and ..
url_fixed = self._fix_path(url)
# According to RFC 3986, URLs can not contain non-ASCII characters; however this is not
# always respected by websites: some tend to give out URLs with non percent-encoded
# non-ASCII characters (see telemb.py, ard.py [#3412])
# urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
# To work around aforementioned issue we will replace request's original URL with
# percent-encoded one
# Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
# the code of this workaround has been moved here from YoutubeDL.urlopen()
url_escaped = escape_url(url_fixed)
# Substitute URL if any change after escaping
if url != url_escaped:
req = update_Request(req, url=url_escaped)
for h, v in std_headers.items():
# Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
# The dict keys are capitalized because of this bug by urllib
if h.capitalize() not in req.headers:
req.add_header(h, v)
# Similarly, 'Accept-encoding'
if 'Accept-encoding' not in req.headers:
req.add_header(
'Accept-Encoding', join_nonempty(
'gzip', 'deflate', brotli and 'br', ncompress and 'compress',
delim=', '))
req.headers = handle_youtubedl_headers(req.headers)
if sys.version_info < (2, 7):
# avoid possible race where __r_type may be unset
req.get_type()
if '#' in req.get_full_url():
# Python 2.6 is brain-dead when it comes to fragments
req._Request__original = req._Request__original.partition('#')[0]
req._Request__r_type = req._Request__r_type.partition('#')[0]
# Use the totally undocumented AbstractHTTPHandler per
# https://github.com/yt-dlp/yt-dlp/pull/4158
return compat_urllib_request.AbstractHTTPHandler.do_request_(self, req)
def http_response(self, req, resp):
old_resp = resp
# Content-Encoding header lists the encodings in order that they were applied [1].
# To decompress, we simply do the reverse.
# [1]: https://datatracker.ietf.org/doc/html/rfc9110#name-content-encoding
decoded_response = None
decoders = {
'gzip': self.deflate_gz,
'deflate': self.deflate_gz,
}
if brotli:
decoders['br'] = self.brotli
if ncompress:
decoders['compress'] = self.compress
if sys.platform.startswith('java'):
# Jython zlib implementation misses gzip
decoders['gzip'] = self.gzip
def encodings(hdrs):
# A header field that allows multiple values can have multiple instances [2].
# [2]: https://datatracker.ietf.org/doc/html/rfc9110#name-fields
for e in reversed(','.join(hdrs).split(',')):
if e:
yield e.strip()
encodings_left = []
try:
resp.headers.get_all
hdrs = resp.headers
except AttributeError:
# Py2 has no get_all() method: headers are rfc822.Message
from email.message import Message
hdrs = Message()
for k, v in resp.headers.items():
hdrs[k] = v
decoder, decoded_response = True, None
for encoding in encodings(hdrs.get_all('Content-Encoding', [])):
# "SHOULD consider" x-compress, x-gzip as compress, gzip
decoder = decoder and decoders.get(remove_start(encoding, 'x-'))
if not decoder:
encodings_left.insert(0, encoding)
continue
decoded_response = decoder(decoded_response or resp.read())
if decoded_response is not None:
resp = compat_urllib_request.addinfourl(
io.BytesIO(decoded_response), old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
del resp.headers['Content-Length']
resp.headers['Content-Length'] = '%d' % len(decoded_response)
del resp.headers['Content-Encoding']
if encodings_left:
resp.headers['Content-Encoding'] = ', '.join(encodings_left)
# Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
# https://github.com/ytdl-org/youtube-dl/issues/6457).
if 300 <= resp.code < 400:
location = resp.headers.get('Location')
if location:
# As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
if sys.version_info >= (3, 0):
location = location.encode('iso-8859-1')
location = location.decode('utf-8')
# resolve embedded . and ..
location_fixed = self._fix_path(location)
location_escaped = escape_url(location_fixed)
if location != location_escaped:
del resp.headers['Location']
if not isinstance(location_escaped, str): # Py 2 case
location_escaped = location_escaped.encode('utf-8')
resp.headers['Location'] = location_escaped
return resp
https_request = http_request
https_response = http_response
def make_socks_conn_class(base_class, socks_proxy):
assert issubclass(base_class, (
compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection))
url_components = compat_urllib_parse.urlparse(socks_proxy)
if url_components.scheme.lower() == 'socks5':
socks_type = ProxyType.SOCKS5
elif url_components.scheme.lower() in ('socks', 'socks4'):
socks_type = ProxyType.SOCKS4
elif url_components.scheme.lower() == 'socks4a':
socks_type = ProxyType.SOCKS4A
def unquote_if_non_empty(s):
if not s:
return s
return compat_urllib_parse_unquote_plus(s)
proxy_args = (
socks_type,
url_components.hostname, url_components.port or 1080,
True, # Remote DNS
unquote_if_non_empty(url_components.username),
unquote_if_non_empty(url_components.password),
)
class SocksConnection(base_class):
def connect(self):
self.sock = sockssocket()
self.sock.setproxy(*proxy_args)
if type(self.timeout) in (int, float):
self.sock.settimeout(self.timeout)
self.sock.connect((self.host, self.port))
if isinstance(self, compat_http_client.HTTPSConnection):
if hasattr(self, '_context'): # Python > 2.6
self.sock = self._context.wrap_socket(
self.sock, server_hostname=self.host)
else:
self.sock = ssl.wrap_socket(self.sock)
return SocksConnection
class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
def __init__(self, params, https_conn_class=None, *args, **kwargs):
compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs)
self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection
self._params = params
def https_open(self, req):
kwargs = {}
conn_class = self._https_conn_class
if hasattr(self, '_context'): # python > 2.6
kwargs['context'] = self._context
if hasattr(self, '_check_hostname'): # python 3.x
kwargs['check_hostname'] = self._check_hostname
socks_proxy = req.headers.get('Ytdl-socks-proxy')
if socks_proxy:
conn_class = make_socks_conn_class(conn_class, socks_proxy)
del req.headers['Ytdl-socks-proxy']
return self.do_open(functools.partial(
_create_http_connection, self, conn_class, True),
req, **kwargs)
class YoutubeDLCookieJar(compat_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 youtube-dl. Do not edit.
'''
_CookieFileEntry = collections.namedtuple(
'CookieFileEntry',
('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
def save(self, filename=None, ignore_discard=False, ignore_expires=False):
"""
Save cookies to a file.
Most of the code is taken from CPython 3.8 and slightly adapted
to support cookie files with UTF-8 in both python 2 and 3.
"""
if filename is None:
if self.filename is not None:
filename = self.filename
else:
raise ValueError(compat_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 io.open(filename, 'w', encoding='utf-8') as f:
f.write(self._HEADER)
now = time.time()
for cookie in self:
if not ignore_discard and cookie.discard:
continue
if not ignore_expires and cookie.is_expired(now):
continue
if cookie.secure:
secure = 'TRUE'
else:
secure = 'FALSE'
if cookie.domain.startswith('.'):
initial_dot = 'TRUE'
else:
initial_dot = 'FALSE'
if cookie.expires is not None:
expires = compat_str(cookie.expires)
else:
expires = ''
if cookie.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 = cookie.name
else:
name = cookie.name
value = cookie.value
f.write(
'\t'.join([cookie.domain, initial_dot, cookie.path,
secure, expires, name, value]) + '\n')
def load(self, filename=None, ignore_discard=False, ignore_expires=False):
"""Load cookies from a file."""
if filename is None:
if self.filename is not None:
filename = self.filename
else:
raise ValueError(compat_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 compat_cookiejar.LoadError('invalid length %d' % len(cookie_list))
cookie = self._CookieFileEntry(*cookie_list)
if cookie.expires_at and not cookie.expires_at.isdigit():
raise compat_cookiejar.LoadError('invalid expires at %s' % cookie.expires_at)
return line
cf = io.StringIO()
with io.open(filename, encoding='utf-8') as f:
for line in f:
try:
cf.write(prepare_line(line))
except compat_cookiejar.LoadError as e:
write_string(
'WARNING: skipping cookie file entry due to %s: %r\n'
% (e, line), sys.stderr)
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 = sanitized_Request(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(sanitized_Request(url))
class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
def __init__(self, cookiejar=None):
compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
def http_response(self, request, response):
# Python 2 will choke on next HTTP request in row if there are non-ASCII
# characters in Set-Cookie HTTP header of last response (see
# https://github.com/ytdl-org/youtube-dl/issues/6769).
# In order to at least prevent crashing we will percent encode Set-Cookie
# header before HTTPCookieProcessor starts processing it.
# if sys.version_info < (3, 0) and response.headers:
# for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
# set_cookie = response.headers.get(set_cookie_header)
# if set_cookie:
# set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
# if set_cookie != set_cookie_escaped:
# del response.headers[set_cookie_header]
# response.headers[set_cookie_header] = set_cookie_escaped
return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response)
https_request = compat_urllib_request.HTTPCookieProcessor.http_request
https_response = http_response
class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
"""YoutubeDL redirect handler
The code is based on HTTPRedirectHandler implementation from CPython [1].
This redirect handler fixes and improves the logic to better align with RFC7261
and what browsers tend to do [2][3]
1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py
2. https://datatracker.ietf.org/doc/html/rfc7231
3. https://github.com/python/cpython/issues/91306
"""
# Supply possibly missing alias
http_error_308 = compat_urllib_request.HTTPRedirectHandler.http_error_302
def redirect_request(self, req, fp, code, msg, headers, newurl):
"""Return a Request or None in response to a redirect.
This is called by the http_error_30x methods when a
redirection response is received. If a redirection should
take place, return a new Request to allow http_error_30x to
perform the redirect. Otherwise, raise HTTPError if no-one
else should try to handle this url. Return None if you can't
but another Handler might.
"""
if code not in (301, 302, 303, 307, 308):
raise compat_urllib_HTTPError(req.full_url, code, msg, headers, fp)
new_method = req.get_method()
new_data = req.data
# On python 2 urlh.geturl() may sometimes return redirect URL
# as a byte string instead of unicode. This workaround forces
# it to return unicode.
newurl = _decode_compat_str(newurl)
# Be conciliant with URIs containing a space. This is mainly
# redundant with the more complete encoding done in http_error_302(),
# but it is kept for compatibility with other callers.
newurl = newurl.replace(' ', '%20')
# Technically the Cookie header should be in unredirected_hdrs;
# however in practice some may set it in normal headers anyway.
# We will remove it here to prevent any leaks.
remove_headers = ['Cookie']
# A 303 must either use GET or HEAD for subsequent request
# https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.4
if code == 303 and req.get_method() != 'HEAD':
new_method = 'GET'
# 301 and 302 redirects are commonly turned into a GET from a POST
# for subsequent requests by browsers, so we'll do the same.
# https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.2
# https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.3
elif code in (301, 302) and req.get_method() == 'POST':
new_method = 'GET'
# only remove payload if method changed (e.g. POST to GET)
if new_method != req.get_method():
new_data = None
remove_headers.extend(['Content-Length', 'Content-Type'])
new_headers = filter_dict(req.headers, cndn=lambda k, _: k.title() not in remove_headers)
return compat_urllib_request.Request(
newurl, headers=new_headers, origin_req_host=req.origin_req_host,
unverifiable=True, method=new_method, data=new_data)
def extract_timezone(date_str):
m = re.search(
r'''(?x)
^.{8,}? # >=8 char non-TZ prefix, if present
(?P<tz>Z| # just the UTC Z, or
(?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or
(?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d)) # not preceded by 3 alpha word or >= 4 alpha or 2 digits
[ ]? # optional space
(?P<sign>\+|-) # +/-
(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2}) # hh[:]mm
$)
''', date_str)
if not m:
m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip())
if timezone is not None:
date_str = date_str[:-len(m.group('tz'))]
timezone = datetime.timedelta(hours=timezone or 0)
else:
date_str = date_str[:-len(m.group('tz'))]
if not m.group('sign'):
timezone = datetime.timedelta()
else:
sign = 1 if m.group('sign') == '+' else -1
timezone = datetime.timedelta(
hours=sign * int(m.group('hours')),
minutes=sign * int(m.group('minutes')))
return timezone, date_str
def parse_iso8601(date_str, delimiter='T', timezone=None):
""" Return a UNIX timestamp from the given date """
if date_str is None:
return None
date_str = re.sub(r'\.[0-9]+', '', date_str)
if timezone is None:
timezone, date_str = extract_timezone(date_str)
try:
date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
dt = datetime.datetime.strptime(date_str, date_format) - timezone
return calendar.timegm(dt.timetuple())
except ValueError:
pass
def date_formats(day_first=True):
return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST
def unified_strdate(date_str, day_first=True):
"""Return a string with the date in the format YYYYMMDD"""
if date_str is None:
return None
upload_date = None
# Replace commas
date_str = date_str.replace(',', ' ')
# Remove AM/PM + timezone
date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
_, date_str = extract_timezone(date_str)
for expression in date_formats(day_first):
try:
upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
except ValueError:
pass
if upload_date is None:
timetuple = email.utils.parsedate_tz(date_str)
if timetuple:
try:
upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
except ValueError:
pass
if upload_date is not None:
return compat_str(upload_date)
def unified_timestamp(date_str, day_first=True):
if date_str is None:
return None
date_str = re.sub(r'\s+', ' ', re.sub(
r'(?i)[,|]|(mon|tues?|wed(nes)?|thu(rs)?|fri|sat(ur)?)(day)?', '', date_str))
pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
timezone, date_str = extract_timezone(date_str)
# Remove AM/PM + timezone
date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
# Remove unrecognized timezones from ISO 8601 alike timestamps
m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
if m:
date_str = date_str[:-len(m.group('tz'))]
# Python only supports microseconds, so remove nanoseconds
m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str)
if m:
date_str = m.group(1)
for expression in date_formats(day_first):
try:
dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
return calendar.timegm(dt.timetuple())
except ValueError:
pass
timetuple = email.utils.parsedate_tz(date_str)
if timetuple:
return calendar.timegm(timetuple) + pm_delta * 3600 - compat_datetime_timedelta_total_seconds(timezone)
def determine_ext(url, default_ext='unknown_video'):
if url is None or '.' not in url:
return default_ext
guess = url.partition('?')[0].rpartition('.')[2]
if re.match(r'^[A-Za-z0-9]+$', guess):
return guess
# Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
elif guess.rstrip('/') in KNOWN_EXTENSIONS:
return guess.rstrip('/')
else:
return default_ext
def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None):
return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext)
def date_from_str(date_str):
"""
Return a datetime object from a string in the format YYYYMMDD or
(now|today)[+-][0-9](day|week|month|year)(s)?"""
today = datetime.date.today()
if date_str in ('now', 'today'):
return today
if date_str == 'yesterday':
return today - datetime.timedelta(days=1)
match = re.match(r'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
if match is not None:
sign = match.group('sign')
time = int(match.group('time'))
if sign == '-':
time = -time
unit = match.group('unit')
# A bad approximation?
if unit == 'month':
unit = 'day'
time *= 30
elif unit == 'year':
unit = 'day'
time *= 365
unit += 's'
delta = datetime.timedelta(**{unit: time})
return today + delta
return datetime.datetime.strptime(date_str, '%Y%m%d').date()
def hyphenate_date(date_str):
"""
Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str)
if match is not None:
return '-'.join(match.groups())
else:
return date_str
class DateRange(object):
"""Represents a time interval between two dates"""
def __init__(self, start=None, end=None):
"""start and end must be strings in the format accepted by date"""
if start is not None:
self.start = date_from_str(start)
else:
self.start = datetime.datetime.min.date()
if end is not None:
self.end = date_from_str(end)
else:
self.end = datetime.datetime.max.date()
if self.start > self.end:
raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
@classmethod
def day(cls, day):
"""Returns a range that only contains the given day"""
return cls(day, day)
def __contains__(self, date):
"""Check if the date is in the range"""
if not isinstance(date, datetime.date):
date = date_from_str(date)
return self.start <= date <= self.end
def __str__(self):
return '%s - %s' % (self.start.isoformat(), self.end.isoformat())
def __eq__(self, other):
return (isinstance(other, DateRange)
and self.start == other.start and self.end == other.end)
def platform_name():
""" Returns the platform name as a compat_str """
res = platform.platform()
return _decode_compat_str(res)
def _windows_write_string(s, out):
""" Returns True if the string was written using special methods,
False if it has yet to be written out."""
# Adapted from http://stackoverflow.com/a/3259271/35070
import ctypes
import ctypes.wintypes
WIN_OUTPUT_IDS = {
1: -11,
2: -12,
}
try:
fileno = out.fileno()
except AttributeError:
# If the output stream doesn't have a fileno, it's virtual
return False
except io.UnsupportedOperation:
# Some strange Windows pseudo files?
return False
if fileno not in WIN_OUTPUT_IDS:
return False
GetStdHandle = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)(
('GetStdHandle', ctypes.windll.kernel32))
h = GetStdHandle(WIN_OUTPUT_IDS[fileno])
WriteConsoleW = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR,
ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD),
ctypes.wintypes.LPVOID)(('WriteConsoleW', ctypes.windll.kernel32))
written = ctypes.wintypes.DWORD(0)
GetFileType = compat_ctypes_WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(('GetFileType', ctypes.windll.kernel32))
FILE_TYPE_CHAR = 0x0002
FILE_TYPE_REMOTE = 0x8000
GetConsoleMode = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE,
ctypes.POINTER(ctypes.wintypes.DWORD))(
('GetConsoleMode', ctypes.windll.kernel32))
INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value
def not_a_console(handle):
if handle == INVALID_HANDLE_VALUE or handle is None:
return True
return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR
or GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0)
if not_a_console(h):
return False
def next_nonbmp_pos(s):
try:
return next(i for i, c in enumerate(s) if ord(c) > 0xffff)
except StopIteration:
return len(s)
while s:
count = min(next_nonbmp_pos(s), 1024)
ret = WriteConsoleW(
h, s, count if count else 2, ctypes.byref(written), None)
if ret == 0:
raise OSError('Failed to write string')
if not count: # We just wrote a non-BMP character
assert written.value == 2
s = s[1:]
else:
assert written.value > 0
s = s[written.value:]
return True
def write_string(s, out=None, encoding=None):
if out is None:
out = sys.stderr
assert isinstance(s, compat_str)
if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'):
if _windows_write_string(s, out):
return
if ('b' in getattr(out, 'mode', '')
or sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr
byt = s.encode(encoding or preferredencoding(), 'ignore')
out.write(byt)
elif hasattr(out, 'buffer'):
enc = encoding or getattr(out, 'encoding', None) or preferredencoding()
byt = s.encode(enc, 'ignore')
out.buffer.write(byt)
else:
out.write(s)
out.flush()
def bytes_to_intlist(bs):
if not bs:
return []
if isinstance(bs[0], int): # Python 3
return list(bs)
else:
return [ord(c) for c in bs]
def intlist_to_bytes(xs):
if not xs:
return b''
return compat_struct_pack('%dB' % len(xs), *xs)
# Cross-platform file locking
if sys.platform == 'win32':
import ctypes.wintypes
import msvcrt
class OVERLAPPED(ctypes.Structure):
_fields_ = [
('Internal', ctypes.wintypes.LPVOID),
('InternalHigh', ctypes.wintypes.LPVOID),
('Offset', ctypes.wintypes.DWORD),
('OffsetHigh', ctypes.wintypes.DWORD),
('hEvent', ctypes.wintypes.HANDLE),
]
kernel32 = ctypes.windll.kernel32
LockFileEx = kernel32.LockFileEx
LockFileEx.argtypes = [
ctypes.wintypes.HANDLE, # hFile
ctypes.wintypes.DWORD, # dwFlags
ctypes.wintypes.DWORD, # dwReserved
ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
ctypes.POINTER(OVERLAPPED) # Overlapped
]
LockFileEx.restype = ctypes.wintypes.BOOL
UnlockFileEx = kernel32.UnlockFileEx
UnlockFileEx.argtypes = [
ctypes.wintypes.HANDLE, # hFile
ctypes.wintypes.DWORD, # dwReserved
ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
ctypes.POINTER(OVERLAPPED) # Overlapped
]
UnlockFileEx.restype = ctypes.wintypes.BOOL
whole_low = 0xffffffff
whole_high = 0x7fffffff
def _lock_file(f, exclusive):
overlapped = OVERLAPPED()
overlapped.Offset = 0
overlapped.OffsetHigh = 0
overlapped.hEvent = 0
f._lock_file_overlapped_p = ctypes.pointer(overlapped)
handle = msvcrt.get_osfhandle(f.fileno())
if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
whole_low, whole_high, f._lock_file_overlapped_p):
raise OSError('Locking file failed: %r' % ctypes.FormatError())
def _unlock_file(f):
assert f._lock_file_overlapped_p
handle = msvcrt.get_osfhandle(f.fileno())
if not UnlockFileEx(handle, 0,
whole_low, whole_high, f._lock_file_overlapped_p):
raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
else:
# Some platforms, such as Jython, is missing fcntl
try:
import fcntl
def _lock_file(f, exclusive):
fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
def _unlock_file(f):
fcntl.flock(f, fcntl.LOCK_UN)
except ImportError:
UNSUPPORTED_MSG = 'file locking is not supported on this platform'
def _lock_file(f, exclusive):
raise IOError(UNSUPPORTED_MSG)
def _unlock_file(f):
raise IOError(UNSUPPORTED_MSG)
class locked_file(object):
def __init__(self, filename, mode, encoding=None):
assert mode in ['r', 'a', 'w']
self.f = io.open(filename, mode, encoding=encoding)
self.mode = mode
def __enter__(self):
exclusive = self.mode != 'r'
try:
_lock_file(self.f, exclusive)
except IOError:
self.f.close()
raise
return self
def __exit__(self, etype, value, traceback):
try:
_unlock_file(self.f)
finally:
self.f.close()
def __iter__(self):
return iter(self.f)
def write(self, *args):
return self.f.write(*args)
def read(self, *args):
return self.f.read(*args)
def get_filesystem_encoding():
encoding = sys.getfilesystemencoding()
return encoding if encoding is not None else 'utf-8'
def shell_quote(args):
quoted_args = []
encoding = get_filesystem_encoding()
for a in args:
# We may get a filename encoded with 'encodeFilename'
a = _decode_compat_str(a, encoding)
quoted_args.append(compat_shlex_quote(a))
return ' '.join(quoted_args)
def smuggle_url(url, data):
""" Pass additional data in a URL for internal use. """
url, idata = unsmuggle_url(url, {})
data.update(idata)
sdata = compat_urllib_parse_urlencode(
{'__youtubedl_smuggle': json.dumps(data)})
return url + '#' + sdata
def unsmuggle_url(smug_url, default=None):
if '#__youtubedl_smuggle' not in smug_url:
return smug_url, default
url, _, sdata = smug_url.rpartition('#')
jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
data = json.loads(jsond)
return url, data
def format_bytes(bytes):
if bytes is None:
return 'N/A'
if type(bytes) is str:
bytes = float(bytes)
if bytes == 0.0:
exponent = 0
else:
exponent = int(math.log(bytes, 1024.0))
suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
converted = float(bytes) / float(1024 ** exponent)
return '%.2f%s' % (converted, suffix)
def lookup_unit_table(unit_table, s):
units_re = '|'.join(re.escape(u) for u in unit_table)
m = re.match(
r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s)
if not m:
return None
num_str = m.group('num').replace(',', '.')
mult = unit_table[m.group('unit')]
return int(float(num_str) * mult)
def parse_filesize(s):
if s is None:
return None
# The lower-case forms are of course incorrect and unofficial,
# but we support those too
_UNIT_TABLE = {
'B': 1,
'b': 1,
'bytes': 1,
'KiB': 1024,
'KB': 1000,
'kB': 1024,
'Kb': 1000,
'kb': 1000,
'kilobytes': 1000,
'kibibytes': 1024,
'MiB': 1024 ** 2,
'MB': 1000 ** 2,
'mB': 1024 ** 2,
'Mb': 1000 ** 2,
'mb': 1000 ** 2,
'megabytes': 1000 ** 2,
'mebibytes': 1024 ** 2,
'GiB': 1024 ** 3,
'GB': 1000 ** 3,
'gB': 1024 ** 3,
'Gb': 1000 ** 3,
'gb': 1000 ** 3,
'gigabytes': 1000 ** 3,
'gibibytes': 1024 ** 3,
'TiB': 1024 ** 4,
'TB': 1000 ** 4,
'tB': 1024 ** 4,
'Tb': 1000 ** 4,
'tb': 1000 ** 4,
'terabytes': 1000 ** 4,
'tebibytes': 1024 ** 4,
'PiB': 1024 ** 5,
'PB': 1000 ** 5,
'pB': 1024 ** 5,
'Pb': 1000 ** 5,
'pb': 1000 ** 5,
'petabytes': 1000 ** 5,
'pebibytes': 1024 ** 5,
'EiB': 1024 ** 6,
'EB': 1000 ** 6,
'eB': 1024 ** 6,
'Eb': 1000 ** 6,
'eb': 1000 ** 6,
'exabytes': 1000 ** 6,
'exbibytes': 1024 ** 6,
'ZiB': 1024 ** 7,
'ZB': 1000 ** 7,
'zB': 1024 ** 7,
'Zb': 1000 ** 7,
'zb': 1000 ** 7,
'zettabytes': 1000 ** 7,
'zebibytes': 1024 ** 7,
'YiB': 1024 ** 8,
'YB': 1000 ** 8,
'yB': 1024 ** 8,
'Yb': 1000 ** 8,
'yb': 1000 ** 8,
'yottabytes': 1000 ** 8,
'yobibytes': 1024 ** 8,
}
return lookup_unit_table(_UNIT_TABLE, s)
def parse_count(s):
if s is None:
return None
s = s.strip()
if re.match(r'^[\d,.]+$', s):
return str_to_int(s)
_UNIT_TABLE = {
'k': 1000,
'K': 1000,
'm': 1000 ** 2,
'M': 1000 ** 2,
'kk': 1000 ** 2,
'KK': 1000 ** 2,
}
return lookup_unit_table(_UNIT_TABLE, s)
def parse_resolution(s):
if s is None:
return {}
mobj = re.search(r'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s)
if mobj:
return {
'width': int(mobj.group('w')),
'height': int(mobj.group('h')),
}
mobj = re.search(r'\b(\d+)[pPiI]\b', s)
if mobj:
return {'height': int(mobj.group(1))}
mobj = re.search(r'\b([48])[kK]\b', s)
if mobj:
return {'height': int(mobj.group(1)) * 540}
return {}
def parse_bitrate(s):
s = txt_or_none(s)
if not s:
return None
mobj = re.search(r'\b(\d+)\s*kbps', s)
if mobj:
return int(mobj.group(1))
def month_by_name(name, lang='en'):
""" Return the number of a month by (locale-independently) English name """
month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en'])
try:
return month_names.index(name) + 1
except ValueError:
return None
def month_by_abbreviation(abbrev):
""" Return the number of a month by (locale-independently) English
abbreviations """
try:
return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1
except ValueError:
return None
def fix_xml_ampersands(xml_str):
"""Replace all the '&' by '&amp;' in XML"""
return re.sub(
r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
'&amp;',
xml_str)
def setproctitle(title):
assert isinstance(title, compat_str)
# ctypes in Jython is not complete
# http://bugs.jython.org/issue2148
if sys.platform.startswith('java'):
return
try:
libc = ctypes.cdll.LoadLibrary('libc.so.6')
except OSError:
return
except TypeError:
# LoadLibrary in Windows Python 2.7.13 only expects
# a bytestring, but since unicode_literals turns
# every string into a unicode string, it fails.
return
title_bytes = title.encode('utf-8')
buf = ctypes.create_string_buffer(len(title_bytes))
buf.value = title_bytes
try:
libc.prctl(15, buf, 0, 0, 0)
except AttributeError:
return # Strange libc, just skip this
def remove_start(s, start):
return s[len(start):] if s is not None and s.startswith(start) else s
def remove_end(s, end):
return s[:-len(end)] if s is not None and s.endswith(end) else s
def remove_quotes(s):
if s is None or len(s) < 2:
return s
for quote in ('"', "'", ):
if s[0] == quote and s[-1] == quote:
return s[1:-1]
return s
def url_basename(url):
path = compat_urllib_parse.urlparse(url).path
return path.strip('/').split('/')[-1]
def base_url(url):
return re.match(r'https?://[^?#&]+/', url).group()
def urljoin(base, path):
path = _decode_compat_str(path, encoding='utf-8', or_none=True)
if not path:
return None
if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path):
return path
base = _decode_compat_str(base, encoding='utf-8', or_none=True)
if not base:
return None
return (
re.match(r'^(?:https?:)?//', base)
and compat_urllib_parse.urljoin(base, path))
class HEADRequest(compat_urllib_request.Request):
def get_method(self):
return 'HEAD'
class PUTRequest(compat_urllib_request.Request):
def get_method(self):
return 'PUT'
def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
if get_attr:
if v is not None:
v = getattr(v, get_attr, None)
if v in (None, ''):
return default
try:
return int(v) * invscale // scale
except (ValueError, TypeError, OverflowError):
return default
def str_or_none(v, default=None):
return default if v is None else compat_str(v)
def str_to_int(int_str):
""" A more relaxed version of int_or_none """
if isinstance(int_str, compat_integer_types):
return int_str
elif isinstance(int_str, compat_str):
int_str = re.sub(r'[,\.\+]', '', int_str)
return int_or_none(int_str)
def float_or_none(v, scale=1, invscale=1, default=None):
if v is None:
return default
try:
return float(v) * invscale / scale
except (ValueError, TypeError):
return default
def bool_or_none(v, default=None):
return v if isinstance(v, bool) else default
def strip_or_none(v, default=None):
return v.strip() if isinstance(v, compat_str) else default
def txt_or_none(v, default=None):
""" Combine str/strip_or_none, disallow blank value (for traverse_obj) """
return default if v is None else (compat_str(v).strip() or default)
def url_or_none(url):
if not url or not isinstance(url, compat_str):
return None
url = url.strip()
return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None
def parse_duration(s):
if not isinstance(s, compat_basestring):
return None
s = s.strip()
days, hours, mins, secs, ms = [None] * 5
m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s)
if m:
days, hours, mins, secs, ms = m.groups()
else:
m = re.match(
r'''(?ix)(?:P?
(?:
[0-9]+\s*y(?:ears?)?\s*
)?
(?:
[0-9]+\s*m(?:onths?)?\s*
)?
(?:
[0-9]+\s*w(?:eeks?)?\s*
)?
(?:
(?P<days>[0-9]+)\s*d(?:ays?)?\s*
)?
T)?
(?:
(?P<hours>[0-9]+)\s*h(?:ours?)?\s*
)?
(?:
(?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
)?
(?:
(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
)?Z?$''', s)
if m:
days, hours, mins, secs, ms = m.groups()
else:
m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s)
if m:
hours, mins = m.groups()
else:
return None
duration = 0
if secs:
duration += float(secs)
if mins:
duration += float(mins) * 60
if hours:
duration += float(hours) * 60 * 60
if days:
duration += float(days) * 24 * 60 * 60
if ms:
duration += float(ms)
return duration
def prepend_extension(filename, ext, expected_real_ext=None):
name, real_ext = os.path.splitext(filename)
return (
'{0}.{1}{2}'.format(name, ext, real_ext)
if not expected_real_ext or real_ext[1:] == expected_real_ext
else '{0}.{1}'.format(filename, ext))
def replace_extension(filename, ext, expected_real_ext=None):
name, real_ext = os.path.splitext(filename)
return '{0}.{1}'.format(
name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename,
ext)
def check_executable(exe, args=[]):
""" Checks if the given binary is installed somewhere in PATH, and returns its name.
args can be a list of arguments for a short output (like -version) """
try:
process_communicate_or_kill(subprocess.Popen(
[exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE))
except OSError:
return False
return exe
def get_exe_version(exe, args=['--version'],
version_re=None, unrecognized='present'):
""" Returns the version of the specified executable,
or False if the executable is not present """
try:
# STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
# SIGTTOU if youtube-dl is run in the background.
# See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
out, _ = process_communicate_or_kill(subprocess.Popen(
[encodeArgument(exe)] + args,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE, stderr=subprocess.STDOUT))
except OSError:
return False
out = _decode_compat_str(out, 'ascii', 'ignore')
return detect_exe_version(out, version_re, unrecognized)
def detect_exe_version(output, version_re=None, unrecognized='present'):
assert isinstance(output, compat_str)
if version_re is None:
version_re = r'version\s+([-0-9._a-zA-Z]+)'
m = re.search(version_re, output)
if m:
return m.group(1)
else:
return unrecognized
class LazyList(compat_collections_abc.Iterable):
"""Lazy immutable list from an iterable
Note that slices of a LazyList are lists and not LazyList"""
class IndexError(IndexError):
def __init__(self, cause=None):
if cause:
# reproduce `raise from`
self.__cause__ = cause
super(IndexError, self).__init__()
def __init__(self, iterable, **kwargs):
# kwarg-only
reverse = kwargs.get('reverse', False)
_cache = kwargs.get('_cache')
self._iterable = iter(iterable)
self._cache = [] if _cache is None else _cache
self._reversed = reverse
def __iter__(self):
if self._reversed:
# We need to consume the entire iterable to iterate in reverse
for item in self.exhaust():
yield item
return
for item in self._cache:
yield item
for item in self._iterable:
self._cache.append(item)
yield item
def _exhaust(self):
self._cache.extend(self._iterable)
self._iterable = [] # Discard the emptied iterable to make it pickle-able
return self._cache
def exhaust(self):
"""Evaluate the entire iterable"""
return self._exhaust()[::-1 if self._reversed else 1]
@staticmethod
def _reverse_index(x):
return None if x is None else ~x
def __getitem__(self, idx):
if isinstance(idx, slice):
if self._reversed:
idx = slice(self._reverse_index(idx.start), self._reverse_index(idx.stop), -(idx.step or 1))
start, stop, step = idx.start, idx.stop, idx.step or 1
elif isinstance(idx, int):
if self._reversed:
idx = self._reverse_index(idx)
start, stop, step = idx, idx, 0
else:
raise TypeError('indices must be integers or slices')
if ((start or 0) < 0 or (stop or 0) < 0
or (start is None and step < 0)
or (stop is None and step > 0)):
# We need to consume the entire iterable to be able to slice from the end
# Obviously, never use this with infinite iterables
self._exhaust()
try:
return self._cache[idx]
except IndexError as e:
raise self.IndexError(e)
n = max(start or 0, stop or 0) - len(self._cache) + 1
if n > 0:
self._cache.extend(itertools.islice(self._iterable, n))
try:
return self._cache[idx]
except IndexError as e:
raise self.IndexError(e)
def __bool__(self):
try:
self[-1] if self._reversed else self[0]
except self.IndexError:
return False
return True
def __len__(self):
self._exhaust()
return len(self._cache)
def __reversed__(self):
return type(self)(self._iterable, reverse=not self._reversed, _cache=self._cache)
def __copy__(self):
return type(self)(self._iterable, reverse=self._reversed, _cache=self._cache)
def __repr__(self):
# repr and str should mimic a list. So we exhaust the iterable
return repr(self.exhaust())
def __str__(self):
return repr(self.exhaust())
class PagedList(object):
def __len__(self):
# This is only useful for tests
return len(self.getslice())
class OnDemandPagedList(PagedList):
def __init__(self, pagefunc, pagesize, use_cache=True):
self._pagefunc = pagefunc
self._pagesize = pagesize
self._use_cache = use_cache
if use_cache:
self._cache = {}
def getslice(self, start=0, end=None):
res = []
for pagenum in itertools.count(start // self._pagesize):
firstid = pagenum * self._pagesize
nextfirstid = pagenum * self._pagesize + self._pagesize
if start >= nextfirstid:
continue
page_results = None
if self._use_cache:
page_results = self._cache.get(pagenum)
if page_results is None:
page_results = list(self._pagefunc(pagenum))
if self._use_cache:
self._cache[pagenum] = page_results
startv = (
start % self._pagesize
if firstid <= start < nextfirstid
else 0)
endv = (
((end - 1) % self._pagesize) + 1
if (end is not None and firstid <= end <= nextfirstid)
else None)
if startv != 0 or endv is not None:
page_results = page_results[startv:endv]
res.extend(page_results)
# A little optimization - if current page is not "full", ie. does
# not contain page_size videos then we can assume that this page
# is the last one - there are no more ids on further pages -
# i.e. no need to query again.
if len(page_results) + startv < self._pagesize:
break
# If we got the whole page, but the next page is not interesting,
# break out early as well
if end == nextfirstid:
break
return res
class InAdvancePagedList(PagedList):
def __init__(self, pagefunc, pagecount, pagesize):
self._pagefunc = pagefunc
self._pagecount = pagecount
self._pagesize = pagesize
def getslice(self, start=0, end=None):
res = []
start_page = start // self._pagesize
end_page = (
self._pagecount if end is None else (end // self._pagesize + 1))
skip_elems = start - start_page * self._pagesize
only_more = None if end is None else end - start
for pagenum in range(start_page, end_page):
page = list(self._pagefunc(pagenum))
if skip_elems:
page = page[skip_elems:]
skip_elems = None
if only_more is not None:
if len(page) < only_more:
only_more -= len(page)
else:
page = page[:only_more]
res.extend(page)
break
res.extend(page)
return res
def uppercase_escape(s):
unicode_escape = codecs.getdecoder('unicode_escape')
return re.sub(
r'\\U[0-9a-fA-F]{8}',
lambda m: unicode_escape(m.group(0))[0],
s)
def lowercase_escape(s):
unicode_escape = codecs.getdecoder('unicode_escape')
return re.sub(
r'\\u[0-9a-fA-F]{4}',
lambda m: unicode_escape(m.group(0))[0],
s)
def escape_rfc3986(s):
"""Escape non-ASCII characters as suggested by RFC 3986"""
if sys.version_info < (3, 0):
s = _encode_compat_str(s, 'utf-8')
# ensure unicode: after quoting, it can always be converted
return compat_str(compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]"))
def escape_url(url):
"""Escape URL as suggested by RFC 3986"""
url_parsed = compat_urllib_parse_urlparse(url)
return url_parsed._replace(
netloc=url_parsed.netloc.encode('idna').decode('ascii'),
path=escape_rfc3986(url_parsed.path),
params=escape_rfc3986(url_parsed.params),
query=escape_rfc3986(url_parsed.query),
fragment=escape_rfc3986(url_parsed.fragment)
).geturl()
def parse_qs(url, **kwargs):
return compat_parse_qs(compat_urllib_parse.urlparse(url).query, **kwargs)
def read_batch_urls(batch_fd):
def fixup(url):
url = _decode_compat_str(url, 'utf-8', 'replace')
BOM_UTF8 = '\xef\xbb\xbf'
if url.startswith(BOM_UTF8):
url = url[len(BOM_UTF8):]
url = url.strip()
if url.startswith(('#', ';', ']')):
return False
return url
with contextlib.closing(batch_fd) as fd:
return [url for url in map(fixup, fd) if url]
def urlencode_postdata(*args, **kargs):
return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii')
def update_url(url, **kwargs):
"""Replace URL components specified by kwargs
url: compat_str or parsed URL tuple
if query_update is in kwargs, update query with
its value instead of replacing (overrides any `query`)
NB: query_update expects parse_qs() format: [key: value_list, ...]
returns: compat_str
"""
if not kwargs:
return compat_urllib_parse.urlunparse(url) if isinstance(url, tuple) else url
if not isinstance(url, tuple):
url = compat_urllib_parse.urlparse(url)
query = kwargs.pop('query_update', None)
if query:
qs = compat_parse_qs(url.query)
qs.update(query)
kwargs['query'] = compat_urllib_parse_urlencode(qs, True)
kwargs = compat_kwargs(kwargs)
return compat_urllib_parse.urlunparse(url._replace(**kwargs))
def update_url_query(url, query):
return update_url(url, query_update=query)
def update_Request(req, url=None, data=None, headers={}, query={}):
req_headers = req.headers.copy()
req_headers.update(headers)
req_data = data if data is not None else req.data
req_url = update_url_query(url or req.get_full_url(), query)
req_type = {'HEAD': HEADRequest, 'PUT': PUTRequest}.get(
req.get_method(), compat_urllib_request.Request)
new_req = req_type(
req_url, data=req_data, headers=req_headers,
origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
if hasattr(req, 'timeout'):
new_req.timeout = req.timeout
return new_req
def _multipart_encode_impl(data, boundary):
content_type = 'multipart/form-data; boundary=%s' % boundary
out = b''
for k, v in data.items():
out += b'--' + boundary.encode('ascii') + b'\r\n'
k = _encode_compat_str(k, 'utf-8')
v = _encode_compat_str(v, 'utf-8')
# RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
# suggests sending UTF-8 directly. Firefox sends UTF-8, too
content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n'
if boundary.encode('ascii') in content:
raise ValueError('Boundary overlaps with data')
out += content
out += b'--' + boundary.encode('ascii') + b'--\r\n'
return out, content_type
def multipart_encode(data, boundary=None):
'''
Encode a dict to RFC 7578-compliant form-data
data:
A dict where keys and values can be either Unicode or bytes-like
objects.
boundary:
If specified a Unicode object, it's used as the boundary. Otherwise
a random boundary is generated.
Reference: https://tools.ietf.org/html/rfc7578
'''
has_specified_boundary = boundary is not None
while True:
if boundary is None:
boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff))
try:
out, content_type = _multipart_encode_impl(data, boundary)
break
except ValueError:
if has_specified_boundary:
raise
boundary = None
return out, content_type
def is_iterable_like(x, allowed_types=compat_collections_abc.Iterable, blocked_types=NO_DEFAULT):
if blocked_types is NO_DEFAULT:
blocked_types = (compat_str, bytes, compat_collections_abc.Mapping)
return isinstance(x, allowed_types) and not isinstance(x, blocked_types)
def variadic(x, allowed_types=NO_DEFAULT):
if isinstance(allowed_types, compat_collections_abc.Iterable):
allowed_types = tuple(allowed_types)
return x if is_iterable_like(x, blocked_types=allowed_types) else (x,)
def dict_get(d, key_or_keys, default=None, skip_false_values=True):
exp = (lambda x: x or None) if skip_false_values else IDENTITY
return traverse_obj(d, *variadic(key_or_keys), expected_type=exp,
default=default, get_all=False)
def try_call(*funcs, **kwargs):
# parameter defaults
expected_type = kwargs.get('expected_type')
fargs = kwargs.get('args', [])
fkwargs = kwargs.get('kwargs', {})
for f in funcs:
try:
val = f(*fargs, **fkwargs)
except (AttributeError, KeyError, TypeError, IndexError, ZeroDivisionError):
pass
else:
if expected_type is None or isinstance(val, expected_type):
return val
def try_get(src, getter, expected_type=None):
if not isinstance(getter, (list, tuple)):
getter = [getter]
for get in getter:
try:
v = get(src)
except (AttributeError, KeyError, TypeError, IndexError):
pass
else:
if expected_type is None or isinstance(v, expected_type):
return v
def filter_dict(dct, cndn=lambda _, v: v is not None):
# NB: don't use dict comprehension for python 2.6 compatibility
return dict((k, v) for k, v in dct.items() if cndn(k, v))
def merge_dicts(*dicts, **kwargs):
"""
Merge the `dict`s in `dicts` using the first valid value for each key.
Normally valid: not None and not an empty string
Keyword-only args:
unblank: allow empty string if False (default True)
rev: merge dicts in reverse order (default False)
merge_dicts(dct1, dct2, ..., unblank=False, rev=True)
matches {**dct1, **dct2, ...}
However, merge_dicts(dct1, dct2, ..., rev=True) may often be better.
"""
unblank = kwargs.get('unblank', True)
rev = kwargs.get('rev', False)
if unblank:
def can_merge_str(k, v, to_dict):
return (isinstance(v, compat_str) and v
and isinstance(to_dict[k], compat_str)
and not to_dict[k])
else:
can_merge_str = lambda k, v, to_dict: False
merged = {}
for a_dict in reversed(dicts) if rev else dicts:
for k, v in a_dict.items():
if v is None:
continue
if (k not in merged) or can_merge_str(k, v, merged):
merged[k] = v
return merged
# very poor choice of name, as if Python string encodings weren't confusing enough
def encode_compat_str(s, encoding=preferredencoding(), errors='strict'):
assert isinstance(s, compat_basestring)
return s if isinstance(s, compat_str) else compat_str(s, encoding, errors)
# what it could have been
def _decode_compat_str(s, encoding=preferredencoding(), errors='strict', or_none=False):
if not or_none:
assert isinstance(s, compat_basestring)
return (
s if isinstance(s, compat_str)
else compat_str(s, encoding, errors) if isinstance(s, compat_basestring)
else None)
# the real encode_compat_str, but only for internal use
def _encode_compat_str(s, encoding=preferredencoding(), errors='strict'):
assert isinstance(s, compat_basestring)
return s.encode(encoding, errors) if isinstance(s, compat_str) else s
US_RATINGS = {
'G': 0,
'PG': 10,
'PG-13': 13,
'R': 16,
'NC': 18,
}
TV_PARENTAL_GUIDELINES = {
'TV-Y': 0,
'TV-Y7': 7,
'TV-G': 0,
'TV-PG': 0,
'TV-14': 14,
'TV-MA': 17,
}
def parse_age_limit(s):
if not isinstance(s, bool):
age = int_or_none(s)
if age is not None:
return age if 0 <= age <= 21 else None
if not isinstance(s, compat_basestring):
return None
m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
if m:
return int(m.group('age'))
if s in US_RATINGS:
return US_RATINGS[s]
m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s)
if m:
return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)]
return None
def strip_jsonp(code):
return re.sub(
r'''(?sx)^
(?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
(?:\s*&&\s*(?P=func_name))?
\s*\(\s*(?P<callback_data>.*)\);?
\s*?(?://[^\n]*)*$''',
r'\g<callback_data>', code)
def js_to_json(code, *args, **kwargs):
# vars is a dict of (var, val) pairs to substitute
vars = args[0] if len(args) > 0 else kwargs.get('vars', {})
strict = kwargs.get('strict', False)
STRING_QUOTES = '\'"`'
STRING_RE = '|'.join(r'{0}(?:\\.|[^\\{0}])*{0}'.format(q) for q in STRING_QUOTES)
COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n'
SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
INTEGER_TABLE = (
(r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
(r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
(r'(?s)^(\d+){skip}:?$'.format(skip=SKIP_RE), 10),
)
# compat candidate
JSONDecodeError = json.JSONDecodeError if 'JSONDecodeError' in dir(json) else ValueError
def process_escape(match):
JSON_PASSTHROUGH_ESCAPES = r'"\bfnrtu'
escape = match.group(1) or match.group(2)
return ('\\' + escape if escape in JSON_PASSTHROUGH_ESCAPES
else '\\u00' if escape == 'x'
else '' if escape == '\n'
else escape)
def template_substitute(match):
evaluated = js_to_json(match.group(1), vars, strict=strict)
if evaluated[0] == '"':
return json.loads(evaluated)
return evaluated
def fix_kv(m):
v = m.group(0)
if v in ('true', 'false', 'null'):
return v
elif v in ('undefined', 'void 0'):
return 'null'
elif v.startswith('/*') or v.startswith('//') or v == ',':
return ''
if v[0] in STRING_QUOTES:
v = re.sub(r'(?s)\${([^}]+)}', template_substitute, v[1:-1]) if v[0] == '`' else v[1:-1]
escaped = re.sub(r'(?s)(")|\\(.)', process_escape, v)
return '"{0}"'.format(escaped)
inv = IDENTITY
im = re.split(r'^!+', v)
if len(im) > 1 and not im[-1].endswith(':'):
if (len(v) - len(im[1])) % 2 == 1:
inv = lambda x: 'true' if x == 0 else 'false'
else:
inv = lambda x: 'false' if x == 0 else 'true'
if not any(x for x in im):
return
v = im[-1]
for regex, base in INTEGER_TABLE:
im = re.match(regex, v)
if im:
i = int(im.group(1), base)
return ('"%s":' if v.endswith(':') else '%s') % inv(i)
if v in vars:
try:
if not strict:
json.loads(vars[v])
except JSONDecodeError:
return inv(json.dumps(vars[v]))
else:
return inv(vars[v])
if not strict:
v = try_call(inv, args=(v,), default=v)
if v in ('true', 'false'):
return v
return '"{0}"'.format(v)
raise ValueError('Unknown value: ' + v)
def create_map(mobj):
return json.dumps(dict(json.loads(js_to_json(mobj.group(1) or '[]', vars=vars))))
code = re.sub(r'new Map\((\[.*?\])?\)', create_map, code)
if not strict:
code = re.sub(r'new Date\((".+")\)', r'\g<1>', code)
code = re.sub(r'new \w+\((.*?)\)', lambda m: json.dumps(m.group(0)), code)
code = re.sub(r'parseInt\([^\d]+(\d+)[^\d]+\)', r'\1', code)
code = re.sub(r'\(function\([^)]*\)\s*\{[^}]*\}\s*\)\s*\(\s*(["\'][^)]*["\'])\s*\)', r'\1', code)
return re.sub(r'''(?sx)
{str_}|
{comment}|
,(?={skip}[\]}}])|
void\s0|
!*(?:(?<!\d)[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*|
(?:\b|!+)0(?:[xX][\da-fA-F]+|[0-7]+)(?:{skip}:)?|
!+\d+(?:\.\d*)?(?:{skip}:)?|
[0-9]+(?:{skip}:)|
!+
'''.format(comment=COMMENT_RE, skip=SKIP_RE, str_=STRING_RE), fix_kv, code)
def qualities(quality_ids):
""" Get a numeric quality value out of a list of possible values """
def q(qid):
try:
return quality_ids.index(qid)
except ValueError:
return -1
return q
DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s'
def limit_length(s, length):
""" Add ellipses to overly long strings """
if s is None:
return None
ELLIPSES = '...'
if len(s) > length:
return s[:length - len(ELLIPSES)] + ELLIPSES
return s
def version_tuple(v):
return tuple(int(e) for e in re.split(r'[-.]', v))
def is_outdated_version(version, limit, assume_new=True):
if not version:
return not assume_new
try:
return version_tuple(version) < version_tuple(limit)
except ValueError:
return not assume_new
def ytdl_is_updateable():
""" Returns if youtube-dl can be updated with -U """
from zipimport import zipimporter
return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen')
def args_to_str(args):
# Get a short string representation for a subprocess command
return ' '.join(compat_shlex_quote(a) for a in args)
def error_to_compat_str(err):
return _decode_compat_str(str(err))
def mimetype2ext(mt):
if mt is None:
return None
ext = {
'audio/mp4': 'm4a',
# Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
# it's the most popular one
'audio/mpeg': 'mp3',
}.get(mt)
if ext is not None:
return ext
_, _, res = mt.rpartition('/')
res = res.split(';')[0].strip().lower()
return {
'3gpp': '3gp',
'smptett+xml': 'tt',
'ttaf+xml': 'dfxp',
'ttml+xml': 'ttml',
'x-flv': 'flv',
'x-mp4-fragmented': 'mp4',
'x-ms-sami': 'sami',
'x-ms-wmv': 'wmv',
'mpegurl': 'm3u8',
'x-mpegurl': 'm3u8',
'vnd.apple.mpegurl': 'm3u8',
'dash+xml': 'mpd',
'f4m+xml': 'f4m',
'hds+xml': 'f4m',
'vnd.ms-sstr+xml': 'ism',
'quicktime': 'mov',
'mp2t': 'ts',
'x-wav': 'wav',
}.get(res, res)
def parse_codecs(codecs_str):
# http://tools.ietf.org/html/rfc6381
if not codecs_str:
return {}
split_codecs = list(filter(None, map(
lambda str: str.strip(), codecs_str.strip().strip(',').split(','))))
vcodec, acodec = None, None
for full_codec in split_codecs:
codec = full_codec.split('.')[0]
if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
if not vcodec:
vcodec = full_codec
elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
if not acodec:
acodec = full_codec
else:
write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr)
if not vcodec and not acodec:
if len(split_codecs) == 2:
return {
'vcodec': split_codecs[0],
'acodec': split_codecs[1],
}
else:
return {
'vcodec': vcodec or 'none',
'acodec': acodec or 'none',
}
return {}
def urlhandle_detect_ext(url_handle):
getheader = url_handle.headers.get
cd = getheader('Content-Disposition')
if cd:
m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd)
if m:
e = determine_ext(m.group('filename'), default_ext=None)
if e:
return e
return mimetype2ext(getheader('Content-Type'))
def encode_data_uri(data, mime_type):
return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii'))
def age_restricted(content_limit, age_limit):
""" Returns True iff the content should be blocked """
if age_limit is None: # No limit set
return False
if content_limit is None:
return False # Content available for everyone
return age_limit < content_limit
def is_html(first_bytes):
""" Detect whether a file contains HTML by examining its first bytes. """
BOMS = [
(b'\xef\xbb\xbf', 'utf-8'),
(b'\x00\x00\xfe\xff', 'utf-32-be'),
(b'\xff\xfe\x00\x00', 'utf-32-le'),
(b'\xff\xfe', 'utf-16-le'),
(b'\xfe\xff', 'utf-16-be'),
]
for bom, enc in BOMS:
if first_bytes.startswith(bom):
s = first_bytes[len(bom):].decode(enc, 'replace')
break
else:
s = first_bytes.decode('utf-8', 'replace')
return re.match(r'^\s*<', s)
def determine_protocol(info_dict):
protocol = info_dict.get('protocol')
if protocol is not None:
return protocol
url = info_dict['url']
if url.startswith('rtmp'):
return 'rtmp'
elif url.startswith('mms'):
return 'mms'
elif url.startswith('rtsp'):
return 'rtsp'
ext = determine_ext(url)
if ext == 'm3u8':
return 'm3u8'
elif ext == 'f4m':
return 'f4m'
return compat_urllib_parse_urlparse(url).scheme
def render_table(header_row, data):
""" Render a list of rows, each as a list of values """
table = [header_row] + data
max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)]
format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s'
return '\n'.join(format_str % tuple(row) for row in table)
def _match_one(filter_part, dct):
COMPARISON_OPERATORS = {
'<': operator.lt,
'<=': operator.le,
'>': operator.gt,
'>=': operator.ge,
'=': operator.eq,
'!=': operator.ne,
}
operator_rex = re.compile(r'''(?x)\s*
(?P<key>[a-z_]+)
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?:
(?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
(?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
(?P<strval>(?![0-9.])[a-z0-9A-Z]*)
)
\s*$
''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
m = operator_rex.search(filter_part)
if m:
op = COMPARISON_OPERATORS[m.group('op')]
actual_value = dct.get(m.group('key'))
if (m.group('quotedstrval') is not None
or m.group('strval') is not None
# If the original field is a string and matching comparisonvalue is
# a number we should respect the origin of the original field
# and process comparison value as a string (see
# https://github.com/ytdl-org/youtube-dl/issues/11082).
or actual_value is not None and m.group('intval') is not None
and isinstance(actual_value, compat_str)):
if m.group('op') not in ('=', '!='):
raise ValueError(
'Operator %s does not support string values!' % m.group('op'))
comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
quote = m.group('quote')
if quote is not None:
comparison_value = comparison_value.replace(r'\%s' % quote, quote)
else:
try:
comparison_value = int(m.group('intval'))
except ValueError:
comparison_value = parse_filesize(m.group('intval'))
if comparison_value is None:
comparison_value = parse_filesize(m.group('intval') + 'B')
if comparison_value is None:
raise ValueError(
'Invalid integer value %r in filter part %r' % (
m.group('intval'), filter_part))
if actual_value is None:
return m.group('none_inclusive')
return op(actual_value, comparison_value)
UNARY_OPERATORS = {
'': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
'!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
}
operator_rex = re.compile(r'''(?x)\s*
(?P<op>%s)\s*(?P<key>[a-z_]+)
\s*$
''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
m = operator_rex.search(filter_part)
if m:
op = UNARY_OPERATORS[m.group('op')]
actual_value = dct.get(m.group('key'))
return op(actual_value)
raise ValueError('Invalid filter part %r' % filter_part)
def match_str(filter_str, dct):
""" Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
return all(
_match_one(filter_part, dct) for filter_part in filter_str.split('&'))
def match_filter_func(filter_str):
def _match_func(info_dict):
if match_str(filter_str, info_dict):
return None
else:
video_title = info_dict.get('title', info_dict.get('id', 'video'))
return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
return _match_func
def parse_dfxp_time_expr(time_expr):
if not time_expr:
return
mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
if mobj:
return float(mobj.group('time_offset'))
mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
if mobj:
return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
def srt_subtitles_timecode(seconds):
return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
def dfxp2srt(dfxp_data):
'''
@param dfxp_data A bytes-like object containing DFXP data
@returns A unicode object containing converted SRT data
'''
LEGACY_NAMESPACES = (
(b'http://www.w3.org/ns/ttml', [
b'http://www.w3.org/2004/11/ttaf1',
b'http://www.w3.org/2006/04/ttaf1',
b'http://www.w3.org/2006/10/ttaf1',
]),
(b'http://www.w3.org/ns/ttml#styling', [
b'http://www.w3.org/ns/ttml#style',
]),
)
SUPPORTED_STYLING = [
'color',
'fontFamily',
'fontSize',
'fontStyle',
'fontWeight',
'textDecoration'
]
_x = functools.partial(xpath_with_ns, ns_map={
'xml': 'http://www.w3.org/XML/1998/namespace',
'ttml': 'http://www.w3.org/ns/ttml',
'tts': 'http://www.w3.org/ns/ttml#styling',
})
styles = {}
default_style = {}
class TTMLPElementParser(object):
_out = ''
_unclosed_elements = []
_applied_styles = []
def start(self, tag, attrib):
if tag in (_x('ttml:br'), 'br'):
self._out += '\n'
else:
unclosed_elements = []
style = {}
element_style_id = attrib.get('style')
if default_style:
style.update(default_style)
if element_style_id:
style.update(styles.get(element_style_id, {}))
for prop in SUPPORTED_STYLING:
prop_val = attrib.get(_x('tts:' + prop))
if prop_val:
style[prop] = prop_val
if style:
font = ''
for k, v in sorted(style.items()):
if self._applied_styles and self._applied_styles[-1].get(k) == v:
continue
if k == 'color':
font += ' color="%s"' % v
elif k == 'fontSize':
font += ' size="%s"' % v
elif k == 'fontFamily':
font += ' face="%s"' % v
elif k == 'fontWeight' and v == 'bold':
self._out += '<b>'
unclosed_elements.append('b')
elif k == 'fontStyle' and v == 'italic':
self._out += '<i>'
unclosed_elements.append('i')
elif k == 'textDecoration' and v == 'underline':
self._out += '<u>'
unclosed_elements.append('u')
if font:
self._out += '<font' + font + '>'
unclosed_elements.append('font')
applied_style = {}
if self._applied_styles:
applied_style.update(self._applied_styles[-1])
applied_style.update(style)
self._applied_styles.append(applied_style)
self._unclosed_elements.append(unclosed_elements)
def end(self, tag):
if tag not in (_x('ttml:br'), 'br'):
unclosed_elements = self._unclosed_elements.pop()
for element in reversed(unclosed_elements):
self._out += '</%s>' % element
if unclosed_elements and self._applied_styles:
self._applied_styles.pop()
def data(self, data):
self._out += data
def close(self):
return self._out.strip()
def parse_node(node):
target = TTMLPElementParser()
parser = xml.etree.ElementTree.XMLParser(target=target)
parser.feed(xml.etree.ElementTree.tostring(node))
return parser.close()
for k, v in LEGACY_NAMESPACES:
for ns in v:
dfxp_data = dfxp_data.replace(ns, k)
dfxp = compat_etree_fromstring(dfxp_data)
out = []
paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
if not paras:
raise ValueError('Invalid dfxp/TTML subtitle')
repeat = False
while True:
for style in dfxp.findall(_x('.//ttml:style')):
style_id = style.get('id') or style.get(_x('xml:id'))
if not style_id:
continue
parent_style_id = style.get('style')
if parent_style_id:
if parent_style_id not in styles:
repeat = True
continue
styles[style_id] = styles[parent_style_id].copy()
for prop in SUPPORTED_STYLING:
prop_val = style.get(_x('tts:' + prop))
if prop_val:
styles.setdefault(style_id, {})[prop] = prop_val
if repeat:
repeat = False
else:
break
for p in ('body', 'div'):
ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
if ele is None:
continue
style = styles.get(ele.get('style'))
if not style:
continue
default_style.update(style)
for para, index in zip(paras, itertools.count(1)):
begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
end_time = parse_dfxp_time_expr(para.attrib.get('end'))
dur = parse_dfxp_time_expr(para.attrib.get('dur'))
if begin_time is None:
continue
if not end_time:
if not dur:
continue
end_time = begin_time + dur
out.append('%d\n%s --> %s\n%s\n\n' % (
index,
srt_subtitles_timecode(begin_time),
srt_subtitles_timecode(end_time),
parse_node(para)))
return ''.join(out)
def cli_option(params, command_option, param):
param = params.get(param)
if param:
param = compat_str(param)
return [command_option, param] if param is not None else []
def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
param = params.get(param)
if param is None:
return []
assert isinstance(param, bool)
if separator:
return [command_option + separator + (true_value if param else false_value)]
return [command_option, true_value if param else false_value]
def cli_valueless_option(params, command_option, param, expected_value=True):
param = params.get(param)
return [command_option] if param == expected_value else []
def cli_configuration_args(params, param, default=[]):
ex_args = params.get(param)
if ex_args is None:
return default
assert isinstance(ex_args, list)
return ex_args
class ISO639Utils(object):
# See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
_lang_map = {
'aa': 'aar',
'ab': 'abk',
'ae': 'ave',
'af': 'afr',
'ak': 'aka',
'am': 'amh',
'an': 'arg',
'ar': 'ara',
'as': 'asm',
'av': 'ava',
'ay': 'aym',
'az': 'aze',
'ba': 'bak',
'be': 'bel',
'bg': 'bul',
'bh': 'bih',
'bi': 'bis',
'bm': 'bam',
'bn': 'ben',
'bo': 'bod',
'br': 'bre',
'bs': 'bos',
'ca': 'cat',
'ce': 'che',
'ch': 'cha',
'co': 'cos',
'cr': 'cre',
'cs': 'ces',
'cu': 'chu',
'cv': 'chv',
'cy': 'cym',
'da': 'dan',
'de': 'deu',
'dv': 'div',
'dz': 'dzo',
'ee': 'ewe',
'el': 'ell',
'en': 'eng',
'eo': 'epo',
'es': 'spa',
'et': 'est',
'eu': 'eus',
'fa': 'fas',
'ff': 'ful',
'fi': 'fin',
'fj': 'fij',
'fo': 'fao',
'fr': 'fra',
'fy': 'fry',
'ga': 'gle',
'gd': 'gla',
'gl': 'glg',
'gn': 'grn',
'gu': 'guj',
'gv': 'glv',
'ha': 'hau',
'he': 'heb',
'iw': 'heb', # Replaced by he in 1989 revision
'hi': 'hin',
'ho': 'hmo',
'hr': 'hrv',
'ht': 'hat',
'hu': 'hun',
'hy': 'hye',
'hz': 'her',
'ia': 'ina',
'id': 'ind',
'in': 'ind', # Replaced by id in 1989 revision
'ie': 'ile',
'ig': 'ibo',
'ii': 'iii',
'ik': 'ipk',
'io': 'ido',
'is': 'isl',
'it': 'ita',
'iu': 'iku',
'ja': 'jpn',
'jv': 'jav',
'ka': 'kat',
'kg': 'kon',
'ki': 'kik',
'kj': 'kua',
'kk': 'kaz',
'kl': 'kal',
'km': 'khm',
'kn': 'kan',
'ko': 'kor',
'kr': 'kau',
'ks': 'kas',
'ku': 'kur',
'kv': 'kom',
'kw': 'cor',
'ky': 'kir',
'la': 'lat',
'lb': 'ltz',
'lg': 'lug',
'li': 'lim',
'ln': 'lin',
'lo': 'lao',
'lt': 'lit',
'lu': 'lub',
'lv': 'lav',
'mg': 'mlg',
'mh': 'mah',
'mi': 'mri',
'mk': 'mkd',
'ml': 'mal',
'mn': 'mon',
'mr': 'mar',
'ms': 'msa',
'mt': 'mlt',
'my': 'mya',
'na': 'nau',
'nb': 'nob',
'nd': 'nde',
'ne': 'nep',
'ng': 'ndo',
'nl': 'nld',
'nn': 'nno',
'no': 'nor',
'nr': 'nbl',
'nv': 'nav',
'ny': 'nya',
'oc': 'oci',
'oj': 'oji',
'om': 'orm',
'or': 'ori',
'os': 'oss',
'pa': 'pan',
'pi': 'pli',
'pl': 'pol',
'ps': 'pus',
'pt': 'por',
'qu': 'que',
'rm': 'roh',
'rn': 'run',
'ro': 'ron',
'ru': 'rus',
'rw': 'kin',
'sa': 'san',
'sc': 'srd',
'sd': 'snd',
'se': 'sme',
'sg': 'sag',
'si': 'sin',
'sk': 'slk',
'sl': 'slv',
'sm': 'smo',
'sn': 'sna',
'so': 'som',
'sq': 'sqi',
'sr': 'srp',
'ss': 'ssw',
'st': 'sot',
'su': 'sun',
'sv': 'swe',
'sw': 'swa',
'ta': 'tam',
'te': 'tel',
'tg': 'tgk',
'th': 'tha',
'ti': 'tir',
'tk': 'tuk',
'tl': 'tgl',
'tn': 'tsn',
'to': 'ton',
'tr': 'tur',
'ts': 'tso',
'tt': 'tat',
'tw': 'twi',
'ty': 'tah',
'ug': 'uig',
'uk': 'ukr',
'ur': 'urd',
'uz': 'uzb',
've': 'ven',
'vi': 'vie',
'vo': 'vol',
'wa': 'wln',
'wo': 'wol',
'xh': 'xho',
'yi': 'yid',
'ji': 'yid', # Replaced by yi in 1989 revision
'yo': 'yor',
'za': 'zha',
'zh': 'zho',
'zu': 'zul',
}
@classmethod
def short2long(cls, code):
"""Convert language code from ISO 639-1 to ISO 639-2/T"""
return cls._lang_map.get(code[:2])
@classmethod
def long2short(cls, code):
"""Convert language code from ISO 639-2/T to ISO 639-1"""
for short_name, long_name in cls._lang_map.items():
if long_name == code:
return short_name
class ISO3166Utils(object):
# From http://data.okfn.org/data/core/country-list
_country_map = {
'AF': 'Afghanistan',
'AX': 'Åland Islands',
'AL': 'Albania',
'DZ': 'Algeria',
'AS': 'American Samoa',
'AD': 'Andorra',
'AO': 'Angola',
'AI': 'Anguilla',
'AQ': 'Antarctica',
'AG': 'Antigua and Barbuda',
'AR': 'Argentina',
'AM': 'Armenia',
'AW': 'Aruba',
'AU': 'Australia',
'AT': 'Austria',
'AZ': 'Azerbaijan',
'BS': 'Bahamas',
'BH': 'Bahrain',
'BD': 'Bangladesh',
'BB': 'Barbados',
'BY': 'Belarus',
'BE': 'Belgium',
'BZ': 'Belize',
'BJ': 'Benin',
'BM': 'Bermuda',
'BT': 'Bhutan',
'BO': 'Bolivia, Plurinational State of',
'BQ': 'Bonaire, Sint Eustatius and Saba',
'BA': 'Bosnia and Herzegovina',
'BW': 'Botswana',
'BV': 'Bouvet Island',
'BR': 'Brazil',
'IO': 'British Indian Ocean Territory',
'BN': 'Brunei Darussalam',
'BG': 'Bulgaria',
'BF': 'Burkina Faso',
'BI': 'Burundi',
'KH': 'Cambodia',
'CM': 'Cameroon',
'CA': 'Canada',
'CV': 'Cape Verde',
'KY': 'Cayman Islands',
'CF': 'Central African Republic',
'TD': 'Chad',
'CL': 'Chile',
'CN': 'China',
'CX': 'Christmas Island',
'CC': 'Cocos (Keeling) Islands',
'CO': 'Colombia',
'KM': 'Comoros',
'CG': 'Congo',
'CD': 'Congo, the Democratic Republic of the',
'CK': 'Cook Islands',
'CR': 'Costa Rica',
'CI': 'Côte d\'Ivoire',
'HR': 'Croatia',
'CU': 'Cuba',
'CW': 'Curaçao',
'CY': 'Cyprus',
'CZ': 'Czech Republic',
'DK': 'Denmark',
'DJ': 'Djibouti',
'DM': 'Dominica',
'DO': 'Dominican Republic',
'EC': 'Ecuador',
'EG': 'Egypt',
'SV': 'El Salvador',
'GQ': 'Equatorial Guinea',
'ER': 'Eritrea',
'EE': 'Estonia',
'ET': 'Ethiopia',
'FK': 'Falkland Islands (Malvinas)',
'FO': 'Faroe Islands',
'FJ': 'Fiji',
'FI': 'Finland',
'FR': 'France',
'GF': 'French Guiana',
'PF': 'French Polynesia',
'TF': 'French Southern Territories',
'GA': 'Gabon',
'GM': 'Gambia',
'GE': 'Georgia',
'DE': 'Germany',
'GH': 'Ghana',
'GI': 'Gibraltar',
'GR': 'Greece',
'GL': 'Greenland',
'GD': 'Grenada',
'GP': 'Guadeloupe',
'GU': 'Guam',
'GT': 'Guatemala',
'GG': 'Guernsey',
'GN': 'Guinea',
'GW': 'Guinea-Bissau',
'GY': 'Guyana',
'HT': 'Haiti',
'HM': 'Heard Island and McDonald Islands',
'VA': 'Holy See (Vatican City State)',
'HN': 'Honduras',
'HK': 'Hong Kong',
'HU': 'Hungary',
'IS': 'Iceland',
'IN': 'India',
'ID': 'Indonesia',
'IR': 'Iran, Islamic Republic of',
'IQ': 'Iraq',
'IE': 'Ireland',
'IM': 'Isle of Man',
'IL': 'Israel',
'IT': 'Italy',
'JM': 'Jamaica',
'JP': 'Japan',
'JE': 'Jersey',
'JO': 'Jordan',
'KZ': 'Kazakhstan',
'KE': 'Kenya',
'KI': 'Kiribati',
'KP': 'Korea, Democratic People\'s Republic of',
'KR': 'Korea, Republic of',
'KW': 'Kuwait',
'KG': 'Kyrgyzstan',
'LA': 'Lao People\'s Democratic Republic',
'LV': 'Latvia',
'LB': 'Lebanon',
'LS': 'Lesotho',
'LR': 'Liberia',
'LY': 'Libya',
'LI': 'Liechtenstein',
'LT': 'Lithuania',
'LU': 'Luxembourg',
'MO': 'Macao',
'MK': 'Macedonia, the Former Yugoslav Republic of',
'MG': 'Madagascar',
'MW': 'Malawi',
'MY': 'Malaysia',
'MV': 'Maldives',
'ML': 'Mali',
'MT': 'Malta',
'MH': 'Marshall Islands',
'MQ': 'Martinique',
'MR': 'Mauritania',
'MU': 'Mauritius',
'YT': 'Mayotte',
'MX': 'Mexico',
'FM': 'Micronesia, Federated States of',
'MD': 'Moldova, Republic of',
'MC': 'Monaco',
'MN': 'Mongolia',
'ME': 'Montenegro',
'MS': 'Montserrat',
'MA': 'Morocco',
'MZ': 'Mozambique',
'MM': 'Myanmar',
'NA': 'Namibia',
'NR': 'Nauru',
'NP': 'Nepal',
'NL': 'Netherlands',
'NC': 'New Caledonia',
'NZ': 'New Zealand',
'NI': 'Nicaragua',
'NE': 'Niger',
'NG': 'Nigeria',
'NU': 'Niue',
'NF': 'Norfolk Island',
'MP': 'Northern Mariana Islands',
'NO': 'Norway',
'OM': 'Oman',
'PK': 'Pakistan',
'PW': 'Palau',
'PS': 'Palestine, State of',
'PA': 'Panama',
'PG': 'Papua New Guinea',
'PY': 'Paraguay',
'PE': 'Peru',
'PH': 'Philippines',
'PN': 'Pitcairn',
'PL': 'Poland',
'PT': 'Portugal',
'PR': 'Puerto Rico',
'QA': 'Qatar',
'RE': 'Réunion',
'RO': 'Romania',
'RU': 'Russian Federation',
'RW': 'Rwanda',
'BL': 'Saint Barthélemy',
'SH': 'Saint Helena, Ascension and Tristan da Cunha',
'KN': 'Saint Kitts and Nevis',
'LC': 'Saint Lucia',
'MF': 'Saint Martin (French part)',
'PM': 'Saint Pierre and Miquelon',
'VC': 'Saint Vincent and the Grenadines',
'WS': 'Samoa',
'SM': 'San Marino',
'ST': 'Sao Tome and Principe',
'SA': 'Saudi Arabia',
'SN': 'Senegal',
'RS': 'Serbia',
'SC': 'Seychelles',
'SL': 'Sierra Leone',
'SG': 'Singapore',
'SX': 'Sint Maarten (Dutch part)',
'SK': 'Slovakia',
'SI': 'Slovenia',
'SB': 'Solomon Islands',
'SO': 'Somalia',
'ZA': 'South Africa',
'GS': 'South Georgia and the South Sandwich Islands',
'SS': 'South Sudan',
'ES': 'Spain',
'LK': 'Sri Lanka',
'SD': 'Sudan',
'SR': 'Suriname',
'SJ': 'Svalbard and Jan Mayen',
'SZ': 'Swaziland',
'SE': 'Sweden',
'CH': 'Switzerland',
'SY': 'Syrian Arab Republic',
'TW': 'Taiwan, Province of China',
'TJ': 'Tajikistan',
'TZ': 'Tanzania, United Republic of',
'TH': 'Thailand',
'TL': 'Timor-Leste',
'TG': 'Togo',
'TK': 'Tokelau',
'TO': 'Tonga',
'TT': 'Trinidad and Tobago',
'TN': 'Tunisia',
'TR': 'Turkey',
'TM': 'Turkmenistan',
'TC': 'Turks and Caicos Islands',
'TV': 'Tuvalu',
'UG': 'Uganda',
'UA': 'Ukraine',
'AE': 'United Arab Emirates',
'GB': 'United Kingdom',
'US': 'United States',
'UM': 'United States Minor Outlying Islands',
'UY': 'Uruguay',
'UZ': 'Uzbekistan',
'VU': 'Vanuatu',
'VE': 'Venezuela, Bolivarian Republic of',
'VN': 'Viet Nam',
'VG': 'Virgin Islands, British',
'VI': 'Virgin Islands, U.S.',
'WF': 'Wallis and Futuna',
'EH': 'Western Sahara',
'YE': 'Yemen',
'ZM': 'Zambia',
'ZW': 'Zimbabwe',
}
@classmethod
def short2full(cls, code):
"""Convert an ISO 3166-2 country code to the corresponding full name"""
return cls._country_map.get(code.upper())
class GeoUtils(object):
# Major IPv4 address blocks per country
_country_ip_map = {
'AD': '46.172.224.0/19',
'AE': '94.200.0.0/13',
'AF': '149.54.0.0/17',
'AG': '209.59.64.0/18',
'AI': '204.14.248.0/21',
'AL': '46.99.0.0/16',
'AM': '46.70.0.0/15',
'AO': '105.168.0.0/13',
'AP': '182.50.184.0/21',
'AQ': '23.154.160.0/24',
'AR': '181.0.0.0/12',
'AS': '202.70.112.0/20',
'AT': '77.116.0.0/14',
'AU': '1.128.0.0/11',
'AW': '181.41.0.0/18',
'AX': '185.217.4.0/22',
'AZ': '5.197.0.0/16',
'BA': '31.176.128.0/17',
'BB': '65.48.128.0/17',
'BD': '114.130.0.0/16',
'BE': '57.0.0.0/8',
'BF': '102.178.0.0/15',
'BG': '95.42.0.0/15',
'BH': '37.131.0.0/17',
'BI': '154.117.192.0/18',
'BJ': '137.255.0.0/16',
'BL': '185.212.72.0/23',
'BM': '196.12.64.0/18',
'BN': '156.31.0.0/16',
'BO': '161.56.0.0/16',
'BQ': '161.0.80.0/20',
'BR': '191.128.0.0/12',
'BS': '24.51.64.0/18',
'BT': '119.2.96.0/19',
'BW': '168.167.0.0/16',
'BY': '178.120.0.0/13',
'BZ': '179.42.192.0/18',
'CA': '99.224.0.0/11',
'CD': '41.243.0.0/16',
'CF': '197.242.176.0/21',
'CG': '160.113.0.0/16',
'CH': '85.0.0.0/13',
'CI': '102.136.0.0/14',
'CK': '202.65.32.0/19',
'CL': '152.172.0.0/14',
'CM': '102.244.0.0/14',
'CN': '36.128.0.0/10',
'CO': '181.240.0.0/12',
'CR': '201.192.0.0/12',
'CU': '152.206.0.0/15',
'CV': '165.90.96.0/19',
'CW': '190.88.128.0/17',
'CY': '31.153.0.0/16',
'CZ': '88.100.0.0/14',
'DE': '53.0.0.0/8',
'DJ': '197.241.0.0/17',
'DK': '87.48.0.0/12',
'DM': '192.243.48.0/20',
'DO': '152.166.0.0/15',
'DZ': '41.96.0.0/12',
'EC': '186.68.0.0/15',
'EE': '90.190.0.0/15',
'EG': '156.160.0.0/11',
'ER': '196.200.96.0/20',
'ES': '88.0.0.0/11',
'ET': '196.188.0.0/14',
'EU': '2.16.0.0/13',
'FI': '91.152.0.0/13',
'FJ': '144.120.0.0/16',
'FK': '80.73.208.0/21',
'FM': '119.252.112.0/20',
'FO': '88.85.32.0/19',
'FR': '90.0.0.0/9',
'GA': '41.158.0.0/15',
'GB': '25.0.0.0/8',
'GD': '74.122.88.0/21',
'GE': '31.146.0.0/16',
'GF': '161.22.64.0/18',
'GG': '62.68.160.0/19',
'GH': '154.160.0.0/12',
'GI': '95.164.0.0/16',
'GL': '88.83.0.0/19',
'GM': '160.182.0.0/15',
'GN': '197.149.192.0/18',
'GP': '104.250.0.0/19',
'GQ': '105.235.224.0/20',
'GR': '94.64.0.0/13',
'GT': '168.234.0.0/16',
'GU': '168.123.0.0/16',
'GW': '197.214.80.0/20',
'GY': '181.41.64.0/18',
'HK': '113.252.0.0/14',
'HN': '181.210.0.0/16',
'HR': '93.136.0.0/13',
'HT': '148.102.128.0/17',
'HU': '84.0.0.0/14',
'ID': '39.192.0.0/10',
'IE': '87.32.0.0/12',
'IL': '79.176.0.0/13',
'IM': '5.62.80.0/20',
'IN': '117.192.0.0/10',
'IO': '203.83.48.0/21',
'IQ': '37.236.0.0/14',
'IR': '2.176.0.0/12',
'IS': '82.221.0.0/16',
'IT': '79.0.0.0/10',
'JE': '87.244.64.0/18',
'JM': '72.27.0.0/17',
'JO': '176.29.0.0/16',
'JP': '133.0.0.0/8',
'KE': '105.48.0.0/12',
'KG': '158.181.128.0/17',
'KH': '36.37.128.0/17',
'KI': '103.25.140.0/22',
'KM': '197.255.224.0/20',
'KN': '198.167.192.0/19',
'KP': '175.45.176.0/22',
'KR': '175.192.0.0/10',
'KW': '37.36.0.0/14',
'KY': '64.96.0.0/15',
'KZ': '2.72.0.0/13',
'LA': '115.84.64.0/18',
'LB': '178.135.0.0/16',
'LC': '24.92.144.0/20',
'LI': '82.117.0.0/19',
'LK': '112.134.0.0/15',
'LR': '102.183.0.0/16',
'LS': '129.232.0.0/17',
'LT': '78.56.0.0/13',
'LU': '188.42.0.0/16',
'LV': '46.109.0.0/16',
'LY': '41.252.0.0/14',
'MA': '105.128.0.0/11',
'MC': '88.209.64.0/18',
'MD': '37.246.0.0/16',
'ME': '178.175.0.0/17',
'MF': '74.112.232.0/21',
'MG': '154.126.0.0/17',
'MH': '117.103.88.0/21',
'MK': '77.28.0.0/15',
'ML': '154.118.128.0/18',
'MM': '37.111.0.0/17',
'MN': '49.0.128.0/17',
'MO': '60.246.0.0/16',
'MP': '202.88.64.0/20',
'MQ': '109.203.224.0/19',
'MR': '41.188.64.0/18',
'MS': '208.90.112.0/22',
'MT': '46.11.0.0/16',
'MU': '105.16.0.0/12',
'MV': '27.114.128.0/18',
'MW': '102.70.0.0/15',
'MX': '187.192.0.0/11',
'MY': '175.136.0.0/13',
'MZ': '197.218.0.0/15',
'NA': '41.182.0.0/16',
'NC': '101.101.0.0/18',
'NE': '197.214.0.0/18',
'NF': '203.17.240.0/22',
'NG': '105.112.0.0/12',
'NI': '186.76.0.0/15',
'NL': '145.96.0.0/11',
'NO': '84.208.0.0/13',
'NP': '36.252.0.0/15',
'NR': '203.98.224.0/19',
'NU': '49.156.48.0/22',
'NZ': '49.224.0.0/14',
'OM': '5.36.0.0/15',
'PA': '186.72.0.0/15',
'PE': '186.160.0.0/14',
'PF': '123.50.64.0/18',
'PG': '124.240.192.0/19',
'PH': '49.144.0.0/13',
'PK': '39.32.0.0/11',
'PL': '83.0.0.0/11',
'PM': '70.36.0.0/20',
'PR': '66.50.0.0/16',
'PS': '188.161.0.0/16',
'PT': '85.240.0.0/13',
'PW': '202.124.224.0/20',
'PY': '181.120.0.0/14',
'QA': '37.210.0.0/15',
'RE': '102.35.0.0/16',
'RO': '79.112.0.0/13',
'RS': '93.86.0.0/15',
'RU': '5.136.0.0/13',
'RW': '41.186.0.0/16',
'SA': '188.48.0.0/13',
'SB': '202.1.160.0/19',
'SC': '154.192.0.0/11',
'SD': '102.120.0.0/13',
'SE': '78.64.0.0/12',
'SG': '8.128.0.0/10',
'SI': '188.196.0.0/14',
'SK': '78.98.0.0/15',
'SL': '102.143.0.0/17',
'SM': '89.186.32.0/19',
'SN': '41.82.0.0/15',
'SO': '154.115.192.0/18',
'SR': '186.179.128.0/17',
'SS': '105.235.208.0/21',
'ST': '197.159.160.0/19',
'SV': '168.243.0.0/16',
'SX': '190.102.0.0/20',
'SY': '5.0.0.0/16',
'SZ': '41.84.224.0/19',
'TC': '65.255.48.0/20',
'TD': '154.68.128.0/19',
'TG': '196.168.0.0/14',
'TH': '171.96.0.0/13',
'TJ': '85.9.128.0/18',
'TK': '27.96.24.0/21',
'TL': '180.189.160.0/20',
'TM': '95.85.96.0/19',
'TN': '197.0.0.0/11',
'TO': '175.176.144.0/21',
'TR': '78.160.0.0/11',
'TT': '186.44.0.0/15',
'TV': '202.2.96.0/19',
'TW': '120.96.0.0/11',
'TZ': '156.156.0.0/14',
'UA': '37.52.0.0/14',
'UG': '102.80.0.0/13',
'US': '6.0.0.0/8',
'UY': '167.56.0.0/13',
'UZ': '84.54.64.0/18',
'VA': '212.77.0.0/19',
'VC': '207.191.240.0/21',
'VE': '186.88.0.0/13',
'VG': '66.81.192.0/20',
'VI': '146.226.0.0/16',
'VN': '14.160.0.0/11',
'VU': '202.80.32.0/20',
'WF': '117.20.32.0/21',
'WS': '202.4.32.0/19',
'YE': '134.35.0.0/16',
'YT': '41.242.116.0/22',
'ZA': '41.0.0.0/11',
'ZM': '102.144.0.0/13',
'ZW': '102.177.192.0/18',
}
@classmethod
def random_ipv4(cls, code_or_block):
if len(code_or_block) == 2:
block = cls._country_ip_map.get(code_or_block.upper())
if not block:
return None
else:
block = code_or_block
addr, preflen = block.split('/')
addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
addr_max = addr_min | (0xffffffff >> int(preflen))
return compat_str(socket.inet_ntoa(
compat_struct_pack('!L', random.randint(addr_min, addr_max))))
class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
def __init__(self, proxies=None):
# Set default handlers
for type in ('http', 'https'):
setattr(self, '%s_open' % type,
lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
meth(r, proxy, type))
compat_urllib_request.ProxyHandler.__init__(self, proxies)
def proxy_open(self, req, proxy, type):
req_proxy = req.headers.get('Ytdl-request-proxy')
if req_proxy is not None:
proxy = req_proxy
del req.headers['Ytdl-request-proxy']
if proxy == '__noproxy__':
return None # No Proxy
if compat_urllib_parse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
req.add_header('Ytdl-socks-proxy', proxy)
# youtube-dl's http/https handlers do wrapping the socket with socks
return None
return compat_urllib_request.ProxyHandler.proxy_open(
self, req, proxy, type)
# Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
# released into Public Domain
# https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
def long_to_bytes(n, blocksize=0):
"""long_to_bytes(n:long, blocksize:int) : string
Convert a long integer to a byte string.
If optional blocksize is given and greater than zero, pad the front of the
byte string with binary zeros so that the length is a multiple of
blocksize.
"""
# after much testing, this algorithm was deemed to be the fastest
s = b''
n = int(n)
while n > 0:
s = compat_struct_pack('>I', n & 0xffffffff) + s
n = n >> 32
# strip off leading zeros
for i in range(len(s)):
if s[i] != b'\000'[0]:
break
else:
# only happens when n == 0
s = b'\000'
i = 0
s = s[i:]
# add back some pad bytes. this could be done more efficiently w.r.t. the
# de-padding being done above, but sigh...
if blocksize > 0 and len(s) % blocksize:
s = (blocksize - len(s) % blocksize) * b'\000' + s
return s
def bytes_to_long(s):
"""bytes_to_long(string) : long
Convert a byte string to a long integer.
This is (essentially) the inverse of long_to_bytes().
"""
acc = 0
length = len(s)
if length % 4:
extra = (4 - length % 4)
s = b'\000' * extra + s
length = length + extra
for i in range(0, length, 4):
acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
return acc
def ohdave_rsa_encrypt(data, exponent, modulus):
'''
Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/
Input:
data: data to encrypt, bytes-like object
exponent, modulus: parameter e and N of RSA algorithm, both integer
Output: hex string of encrypted data
Limitation: supports one block encryption only
'''
payload = int(binascii.hexlify(data[::-1]), 16)
encrypted = pow(payload, exponent, modulus)
return '%x' % encrypted
def pkcs1pad(data, length):
"""
Padding input data with PKCS#1 scheme
@param {int[]} data input data
@param {int} length target length
@returns {int[]} padded data
"""
if len(data) > length - 11:
raise ValueError('Input data too long for PKCS#1 padding')
pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)]
return [0, 2] + pseudo_random + [0] + data
def encode_base_n(num, n, table=None):
FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
if not table:
table = FULL_TABLE[:n]
if n > len(table):
raise ValueError('base %d exceeds table length %d' % (n, len(table)))
if num == 0:
return table[0]
ret = ''
while num:
ret = table[num % n] + ret
num = num // n
return ret
def decode_packed_codes(code):
mobj = re.search(PACKED_CODES_RE, code)
obfuscated_code, base, count, symbols = mobj.groups()
base = int(base)
count = int(count)
symbols = symbols.split('|')
symbol_table = {}
while count:
count -= 1
base_n_count = encode_base_n(count, base)
symbol_table[base_n_count] = symbols[count] or base_n_count
return re.sub(
r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
obfuscated_code)
def caesar(s, alphabet, shift):
if shift == 0:
return s
l = len(alphabet)
return ''.join(
alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c
for c in s)
def rot47(s):
return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
def parse_m3u8_attributes(attrib):
info = {}
for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):
if val.startswith('"'):
val = val[1:-1]
info[key] = val
return info
def urshift(val, n):
return val >> n if val >= 0 else (val + 0x100000000) >> n
# Based on png2str() written by @gdkchan and improved by @yokrysty
# Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
def decode_png(png_data):
# Reference: https://www.w3.org/TR/PNG/
header = png_data[8:]
if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR':
raise IOError('Not a valid PNG file.')
int_map = {1: '>B', 2: '>H', 4: '>I'}
unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0]
chunks = []
while header:
length = unpack_integer(header[:4])
header = header[4:]
chunk_type = header[:4]
header = header[4:]
chunk_data = header[:length]
header = header[length:]
header = header[4:] # Skip CRC
chunks.append({
'type': chunk_type,
'length': length,
'data': chunk_data
})
ihdr = chunks[0]['data']
width = unpack_integer(ihdr[:4])
height = unpack_integer(ihdr[4:8])
idat = b''
for chunk in chunks:
if chunk['type'] == b'IDAT':
idat += chunk['data']
if not idat:
raise IOError('Unable to read PNG data.')
decompressed_data = bytearray(zlib.decompress(idat))
stride = width * 3
pixels = []
def _get_pixel(idx):
x = idx % stride
y = idx // stride
return pixels[y][x]
for y in range(height):
basePos = y * (1 + stride)
filter_type = decompressed_data[basePos]
current_row = []
pixels.append(current_row)
for x in range(stride):
color = decompressed_data[1 + basePos + x]
basex = y * stride + x
left = 0
up = 0
if x > 2:
left = _get_pixel(basex - 3)
if y > 0:
up = _get_pixel(basex - stride)
if filter_type == 1: # Sub
color = (color + left) & 0xff
elif filter_type == 2: # Up
color = (color + up) & 0xff
elif filter_type == 3: # Average
color = (color + ((left + up) >> 1)) & 0xff
elif filter_type == 4: # Paeth
a = left
b = up
c = 0
if x > 2 and y > 0:
c = _get_pixel(basex - stride - 3)
p = a + b - c
pa = abs(p - a)
pb = abs(p - b)
pc = abs(p - c)
if pa <= pb and pa <= pc:
color = (color + a) & 0xff
elif pb <= pc:
color = (color + b) & 0xff
else:
color = (color + c) & 0xff
current_row.append(color)
return width, height, pixels
def write_xattr(path, key, value):
# This mess below finds the best xattr tool for the job
try:
# try the pyxattr module...
import xattr
if hasattr(xattr, 'set'): # pyxattr
# Unicode arguments are not supported in python-pyxattr until
# version 0.5.0
# See https://github.com/ytdl-org/youtube-dl/issues/5498
pyxattr_required_version = '0.5.0'
if version_tuple(xattr.__version__) < version_tuple(pyxattr_required_version):
# TODO: fallback to CLI tools
raise XAttrUnavailableError(
'python-pyxattr is detected but is too old. '
'youtube-dl requires %s or above while your version is %s. '
'Falling back to other xattr implementations' % (
pyxattr_required_version, xattr.__version__))
setxattr = xattr.set
else: # xattr
setxattr = xattr.setxattr
try:
setxattr(path, key, value)
except EnvironmentError as e:
raise XAttrMetadataError(e.errno, e.strerror)
except ImportError:
if compat_os_name == 'nt':
# Write xattrs to NTFS Alternate Data Streams:
# http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
assert ':' not in key
assert os.path.exists(path)
ads_fn = path + ':' + key
try:
with open(ads_fn, 'wb') as f:
f.write(value)
except EnvironmentError as e:
raise XAttrMetadataError(e.errno, e.strerror)
else:
user_has_setfattr = check_executable('setfattr', ['--version'])
user_has_xattr = check_executable('xattr', ['-h'])
if user_has_setfattr or user_has_xattr:
value = value.decode('utf-8')
if user_has_setfattr:
executable = 'setfattr'
opts = ['-n', key, '-v', value]
elif user_has_xattr:
executable = 'xattr'
opts = ['-w', key, value]
cmd = ([encodeFilename(executable, True)]
+ [encodeArgument(o) for o in opts]
+ [encodeFilename(path, True)])
try:
p = subprocess.Popen(
cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
except EnvironmentError as e:
raise XAttrMetadataError(e.errno, e.strerror)
stdout, stderr = process_communicate_or_kill(p)
stderr = stderr.decode('utf-8', 'replace')
if p.returncode != 0:
raise XAttrMetadataError(p.returncode, stderr)
else:
# On Unix, and can't find pyxattr, setfattr, or xattr.
if sys.platform.startswith('linux'):
raise XAttrUnavailableError(
"Couldn't find a tool to set the xattrs. "
"Install either the python 'pyxattr' or 'xattr' "
"modules, or the GNU 'attr' package "
"(which contains the 'setfattr' tool).")
else:
raise XAttrUnavailableError(
"Couldn't find a tool to set the xattrs. "
"Install either the python 'xattr' module, "
"or the 'xattr' binary.")
def random_birthday(year_field, month_field, day_field):
start_date = datetime.date(1950, 1, 1)
end_date = datetime.date(1995, 12, 31)
offset = random.randint(0, (end_date - start_date).days)
random_date = start_date + datetime.timedelta(offset)
return {
year_field: str(random_date.year),
month_field: str(random_date.month),
day_field: str(random_date.day),
}
def clean_podcast_url(url):
return re.sub(r'''(?x)
(?:
(?:
chtbl\.com/track|
media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
play\.podtrac\.com
)/[^/]+|
(?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
flex\.acast\.com|
pd(?:
cn\.co| # https://podcorn.com/analytics-prefix/
st\.fm # https://podsights.com/docs/
)/e
)/''', '', url)
if __debug__:
# Raise TypeError if args can't be bound
# needs compat owing to unstable inspect API, thanks PSF :-(
try:
inspect.signature
def _try_bind_args(fn, *args, **kwargs):
inspect.signature(fn).bind(*args, **kwargs)
except AttributeError:
# Py < 3.3
def _try_bind_args(fn, *args, **kwargs):
fn_args = inspect.getargspec(fn)
# Py2: ArgInfo(args, varargs, keywords, defaults)
# Py3: ArgSpec(args, varargs, keywords, defaults)
if not fn_args.keywords:
for k in kwargs:
if k not in (fn_args.args or []):
raise TypeError("got an unexpected keyword argument: '{0}'".format(k))
if not fn_args.varargs:
args_to_bind = len(args)
bindable = len(fn_args.args or [])
if args_to_bind > bindable:
raise TypeError('too many positional arguments')
bindable -= len(fn_args.defaults or [])
if args_to_bind < bindable:
if kwargs:
bindable -= len(set(fn_args.args or []) & set(kwargs))
if bindable > args_to_bind:
raise TypeError("missing a required argument: '{0}'".format(fn_args.args[args_to_bind]))
def traverse_obj(obj, *paths, **kwargs):
"""
Safely traverse nested `dict`s and `Iterable`s
>>> obj = [{}, {"key": "value"}]
>>> traverse_obj(obj, (1, "key"))
"value"
Each of the provided `paths` is tested and the first producing a valid result will be returned.
The next path will also be tested if the path branched but no results could be found.
Supported values for traversal are `Mapping`, `Iterable` and `re.Match`.
Unhelpful values (`{}`, `None`) are treated as the absence of a value and discarded.
The paths will be wrapped in `variadic`, so that `'key'` is conveniently the same as `('key', )`.
The keys in the path can be one of:
- `None`: Return the current object.
- `set`: Requires the only item in the set to be a type or function,
like `{type}`/`{func}`. If a `type`, returns only values
of this type. If a function, returns `func(obj)`.
- `str`/`int`: Return `obj[key]`. For `re.Match`, return `obj.group(key)`.
- `slice`: Branch out and return all values in `obj[key]`.
- `Ellipsis`: Branch out and return a list of all values.
- `tuple`/`list`: Branch out and return a list of all matching values.
Read as: `[traverse_obj(obj, branch) for branch in branches]`.
- `function`: Branch out and return values filtered by the function.
Read as: `[value for key, value in obj if function(key, value)]`.
For `Sequence`s, `key` is the index of the value.
For `Iterable`s, `key` is the enumeration count of the value.
For `re.Match`es, `key` is the group number (0 = full match)
as well as additionally any group names, if given.
- `dict` Transform the current object and return a matching dict.
Read as: `{key: traverse_obj(obj, path) for key, path in dct.items()}`.
`tuple`, `list`, and `dict` all support nested paths and branches.
@params paths Paths which to traverse by.
Keyword arguments:
@param default Value to return if the paths do not match.
If the last key in the path is a `dict`, it will apply to each value inside
the dict instead, depth first. Try to avoid if using nested `dict` keys.
@param expected_type If a `type`, only accept final values of this type.
If any other callable, try to call the function on each result.
If the last key in the path is a `dict`, it will apply to each value inside
the dict instead, recursively. This does respect branching paths.
@param get_all If `False`, return the first matching result, otherwise all matching ones.
@param casesense If `False`, consider string dictionary keys as case insensitive.
The following are only meant to be used by YoutubeDL.prepare_outtmpl and are not part of the API
@param _is_user_input Whether the keys are generated from user input.
If `True` strings get converted to `int`/`slice` if needed.
@param _traverse_string Whether to traverse into objects as strings.
If `True`, any non-compatible object will first be
converted into a string and then traversed into.
The return value of that path will be a string instead,
not respecting any further branching.
@returns The result of the object traversal.
If successful, `get_all=True`, and the path branches at least once,
then a list of results is returned instead.
A list is always returned if the last path branches and no `default` is given.
If a path ends on a `dict` that result will always be a `dict`.
"""
# parameter defaults
default = kwargs.get('default', NO_DEFAULT)
expected_type = kwargs.get('expected_type')
get_all = kwargs.get('get_all', True)
casesense = kwargs.get('casesense', True)
_is_user_input = kwargs.get('_is_user_input', False)
_traverse_string = kwargs.get('_traverse_string', False)
# instant compat
str = compat_str
casefold = lambda k: compat_casefold(k) if isinstance(k, str) else k
if isinstance(expected_type, type):
type_test = lambda val: val if isinstance(val, expected_type) else None
else:
type_test = lambda val: try_call(expected_type or IDENTITY, args=(val,))
def lookup_or_none(v, k, getter=None):
try:
return getter(v, k) if getter else v[k]
except IndexError:
return None
def from_iterable(iterables):
# chain.from_iterable(['ABC', 'DEF']) --> A B C D E F
for it in iterables:
for item in it:
yield item
def apply_key(key, obj, is_last):
branching = False
if obj is None and _traverse_string:
if key is Ellipsis or callable(key) or isinstance(key, slice):
branching = True
result = ()
else:
result = None
elif key is None:
result = obj
elif isinstance(key, set):
assert len(key) == 1, 'Set should only be used to wrap a single item'
item = next(iter(key))
if isinstance(item, type):
result = obj if isinstance(obj, item) else None
else:
result = try_call(item, args=(obj,))
elif isinstance(key, (list, tuple)):
branching = True
result = from_iterable(
apply_path(obj, branch, is_last)[0] for branch in key)
elif key is Ellipsis:
branching = True
if isinstance(obj, compat_collections_abc.Mapping):
result = obj.values()
elif is_iterable_like(obj):
result = obj
elif isinstance(obj, compat_re_Match):
result = obj.groups()
elif _traverse_string:
branching = False
result = str(obj)
else:
result = ()
elif callable(key):
branching = True
if isinstance(obj, compat_collections_abc.Mapping):
iter_obj = obj.items()
elif is_iterable_like(obj):
iter_obj = enumerate(obj)
elif isinstance(obj, compat_re_Match):
iter_obj = itertools.chain(
enumerate(itertools.chain((obj.group(),), obj.groups())),
obj.groupdict().items())
elif _traverse_string:
branching = False
iter_obj = enumerate(str(obj))
else:
iter_obj = ()
result = (v for k, v in iter_obj if try_call(key, args=(k, v)))
if not branching: # string traversal
result = ''.join(result)
elif isinstance(key, dict):
iter_obj = ((k, _traverse_obj(obj, v, False, is_last)) for k, v in key.items())
result = dict((k, v if v is not None else default) for k, v in iter_obj
if v is not None or default is not NO_DEFAULT) or None
elif isinstance(obj, compat_collections_abc.Mapping):
result = (try_call(obj.get, args=(key,))
if casesense or try_call(obj.__contains__, args=(key,))
else next((v for k, v in obj.items() if casefold(k) == key), None))
elif isinstance(obj, compat_re_Match):
result = None
if isinstance(key, int) or casesense:
# Py 2.6 doesn't have methods in the Match class/type
result = lookup_or_none(obj, key, getter=lambda _, k: obj.group(k))
elif isinstance(key, str):
result = next((v for k, v in obj.groupdict().items()
if casefold(k) == key), None)
else:
result = None
if isinstance(key, (int, slice)):
if is_iterable_like(obj, compat_collections_abc.Sequence):
branching = isinstance(key, slice)
result = lookup_or_none(obj, key)
elif _traverse_string:
result = lookup_or_none(str(obj), key)
return branching, result if branching else (result,)
def lazy_last(iterable):
iterator = iter(iterable)
prev = next(iterator, NO_DEFAULT)
if prev is NO_DEFAULT:
return
for item in iterator:
yield False, prev
prev = item
yield True, prev
def apply_path(start_obj, path, test_type):
objs = (start_obj,)
has_branched = False
key = None
for last, key in lazy_last(variadic(path, (str, bytes, dict, set))):
if _is_user_input and isinstance(key, str):
if key == ':':
key = Ellipsis
elif ':' in key:
key = slice(*map(int_or_none, key.split(':')))
elif int_or_none(key) is not None:
key = int(key)
if not casesense and isinstance(key, str):
key = compat_casefold(key)
if __debug__ and callable(key):
# Verify function signature
_try_bind_args(key, None, None)
new_objs = []
for obj in objs:
branching, results = apply_key(key, obj, last)
has_branched |= branching
new_objs.append(results)
objs = from_iterable(new_objs)
if test_type and not isinstance(key, (dict, list, tuple)):
objs = map(type_test, objs)
return objs, has_branched, isinstance(key, dict)
def _traverse_obj(obj, path, allow_empty, test_type):
results, has_branched, is_dict = apply_path(obj, path, test_type)
results = LazyList(x for x in results if x not in (None, {}))
if get_all and has_branched:
if results:
return results.exhaust()
if allow_empty:
return [] if default is NO_DEFAULT else default
return None
return results[0] if results else {} if allow_empty and is_dict else None
for index, path in enumerate(paths, 1):
result = _traverse_obj(obj, path, index == len(paths), True)
if result is not None:
return result
return None if default is NO_DEFAULT else default
def T(x):
""" For use in yt-dl instead of {type} or set((type,)) """
return set((x,))
def get_first(obj, keys, **kwargs):
return traverse_obj(obj, (Ellipsis,) + tuple(variadic(keys)), get_all=False, **kwargs)
def join_nonempty(*values, **kwargs):
# parameter defaults
delim = kwargs.get('delim', '-')
from_dict = kwargs.get('from_dict')
if from_dict is not None:
values = (traverse_obj(from_dict, variadic(v)) for v in values)
return delim.join(map(compat_str, filter(None, values)))
Reference in New Issue View Git Blame Copy Permalink