yt-dlp/yt_dlp/utils.py

5466 lines
176 KiB
Python
Raw Normal View History

#!/usr/bin/env python3
2016-10-02 13:39:18 +02:00
# coding: utf-8
2014-11-02 11:37:49 +01:00
from __future__ import unicode_literals
import asyncio
import atexit
import base64
import binascii
2014-03-24 01:40:09 +01:00
import calendar
2014-04-04 23:00:51 +02:00
import codecs
import collections
import contextlib
import ctypes
import datetime
import email.utils
import email.header
import errno
import functools
import gzip
import hashlib
import hmac
import importlib.util
2012-11-28 00:09:17 +01:00
import io
import itertools
2012-12-20 13:13:24 +01:00
import json
import locale
import math
import operator
import os
import platform
import random
import re
import socket
import ssl
2013-12-09 18:29:07 +01:00
import subprocess
import sys
import tempfile
import time
import traceback
import xml.etree.ElementTree
import zlib
import mimetypes
from .compat import (
compat_HTMLParseError,
compat_HTMLParser,
compat_HTTPError,
2015-02-01 11:30:56 +01:00
compat_basestring,
compat_chr,
compat_cookiejar,
compat_ctypes_WINFUNCTYPE,
compat_etree_fromstring,
2017-03-25 20:30:10 +01:00
compat_expanduser,
compat_html_entities,
compat_html_entities_html5,
compat_http_client,
2019-12-15 17:15:24 +01:00
compat_integer_types,
compat_numeric_types,
compat_kwargs,
compat_os_name,
compat_parse_qs,
compat_shlex_split,
compat_shlex_quote,
compat_str,
compat_struct_pack,
compat_struct_unpack,
compat_urllib_error,
compat_urllib_parse,
compat_urllib_parse_urlencode,
compat_urllib_parse_urlparse,
compat_urllib_parse_urlunparse,
compat_urllib_parse_quote,
compat_urllib_parse_quote_plus,
compat_urllib_parse_unquote_plus,
compat_urllib_request,
compat_urlparse,
compat_websockets,
2016-03-17 21:52:23 +01:00
compat_xpath,
)
2016-04-23 15:30:06 +02:00
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_urlparse.uses_netloc:
compat_urlparse.uses_netloc.append(scheme)
# This is not clearly defined otherwise
compiled_regex_type = type(re.compile(''))
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 = (
'90.0.4430.212',
'90.0.4430.24',
'90.0.4430.70',
'90.0.4430.72',
'90.0.4430.85',
'90.0.4430.93',
'91.0.4472.101',
'91.0.4472.106',
'91.0.4472.114',
'91.0.4472.124',
'91.0.4472.164',
'91.0.4472.19',
'91.0.4472.77',
'92.0.4515.107',
'92.0.4515.115',
'92.0.4515.131',
'92.0.4515.159',
'92.0.4515.43',
'93.0.4556.0',
'93.0.4577.15',
'93.0.4577.63',
'93.0.4577.82',
'94.0.4606.41',
'94.0.4606.54',
'94.0.4606.61',
'94.0.4606.71',
'94.0.4606.81',
'94.0.4606.85',
'95.0.4638.17',
'95.0.4638.50',
'95.0.4638.54',
'95.0.4638.69',
'95.0.4638.74',
'96.0.4664.18',
'96.0.4664.45',
'96.0.4664.55',
'96.0.4664.93',
'97.0.4692.20',
)
return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS)
std_headers = {
'User-Agent': random_user_agent(),
2012-11-28 02:04:46 +01:00
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'en-us,en;q=0.5',
'Sec-Fetch-Mode': 'navigate',
}
2014-11-23 20:41:03 +01:00
2016-12-11 18:49:07 +01:00
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()
2015-02-13 08:14:23 +01:00
ENGLISH_MONTH_NAMES = [
'January', 'February', 'March', 'April', 'May', 'June',
'July', 'August', 'September', 'October', 'November', 'December']
MONTH_NAMES = {
'en': ENGLISH_MONTH_NAMES,
'fr': [
2016-09-14 18:57:01 +02:00
'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
}
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')))
2016-06-25 17:30:35 +02:00
DATE_FORMATS = (
'%d %B %Y',
'%d %b %Y',
'%B %d %Y',
2017-01-12 16:39:45 +01:00
'%B %dst %Y',
'%B %dnd %Y',
'%B %drd %Y',
2017-01-12 16:39:45 +01:00
'%B %dth %Y',
2016-06-25 17:30:35 +02:00
'%b %d %Y',
2017-01-12 16:39:45 +01:00
'%b %dst %Y',
'%b %dnd %Y',
'%b %drd %Y',
2017-01-12 16:39:45 +01:00
'%b %dth %Y',
2016-06-25 17:30:35 +02:00
'%b %dst %Y %I:%M',
'%b %dnd %Y %I:%M',
'%b %drd %Y %I:%M',
2016-06-25 17:30:35 +02:00
'%b %dth %Y %I:%M',
'%Y %m %d',
'%Y-%m-%d',
'%Y.%m.%d.',
2016-06-25 17:30:35 +02:00
'%Y/%m/%d',
'%Y/%m/%d %H:%M',
2016-06-25 17:30:35 +02:00
'%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',
2016-06-25 17:30:35 +02:00
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
'%Y-%m-%d %H:%M:%S:%f',
2016-06-25 17:30:35 +02:00
'%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',
2016-06-25 17:30:35 +02:00
)
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',
])
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',
])
2016-10-19 18:28:49 +02:00
PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
2018-10-10 23:47:21 +02:00
JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
2016-10-19 18:28:49 +02:00
2015-02-13 08:14:23 +01:00
def preferredencoding():
2012-11-28 02:04:46 +01:00
"""Get preferred encoding.
2012-11-28 02:04:46 +01:00
Returns the best encoding scheme for the system, based on
locale.getpreferredencoding() and some further tweaks.
"""
try:
pref = locale.getpreferredencoding()
2014-11-17 07:16:12 +01:00
'TEST'.encode(pref)
except Exception:
2012-11-28 02:04:46 +01:00
pref = 'UTF-8'
2012-11-28 02:04:46 +01:00
return pref
2012-12-20 13:13:24 +01:00
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(fn).decode(encoding)
# the same for os.path.dirname
path_dirname = lambda f: os.path.dirname(fn).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, ensure_ascii=False)
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):
2013-07-11 16:12:08 +02:00
""" Find the xpath xpath[@key=val] """
2015-09-04 19:57:27 +02:00
assert re.match(r'^[a-zA-Z_-]+$', key)
expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
2013-07-11 16:12:08 +02:00
return node.find(expr)
else:
def find_xpath_attr(node, xpath, key, val=None):
2016-03-17 21:52:23 +01:00
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:
2013-07-11 16:12:08 +02:00
return f
return None
# On python2.6 the xml.etree.ElementTree.Element methods don't support
# the namespace parameter
2014-11-23 20:41:03 +01:00
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):
2016-03-17 21:52:23 +01:00
return node.find(compat_xpath(xpath))
if isinstance(xpath, (str, compat_str)):
n = _find_xpath(xpath)
else:
for xp in xpath:
n = _find_xpath(xp)
if n is not None:
break
2015-09-04 20:34:49 +02:00
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):
2015-09-04 20:34:49 +02:00
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"""
2016-02-14 10:37:17 +01:00
return get_element_by_attribute('id', id, html)
def get_element_html_by_id(id, html):
"""Return the html of the tag with the specified ID in the passed HTML document"""
return get_element_html_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_html_by_class(class_name, html):
"""Return the html of the first tag with the specified class in the passed HTML document"""
retval = get_elements_html_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_element_html_by_attribute(attribute, value, html, escape_value=True):
retval = get_elements_html_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_html_by_class(class_name, html):
"""Return the html of all tags with the specified class in the passed HTML document as a list"""
return get_elements_html_by_attribute(
'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
html, escape_value=False)
def get_elements_by_attribute(*args, **kwargs):
"""Return the content of the tag with the specified attribute in the passed HTML document"""
return [content for content, _ in get_elements_text_and_html_by_attribute(*args, **kwargs)]
def get_elements_html_by_attribute(*args, **kwargs):
"""Return the html of the tag with the specified attribute in the passed HTML document"""
return [whole for _, whole in get_elements_text_and_html_by_attribute(*args, **kwargs)]
def get_elements_text_and_html_by_attribute(attribute, value, html, escape_value=True):
"""
Return the text (content) and the html (whole) of the tag with the specified
attribute in the passed HTML document
"""
value_quote_optional = '' if re.match(r'''[\s"'`=<>]''', value) else '?'
value = re.escape(value) if escape_value else value
partial_element_re = r'''(?x)
<(?P<tag>[a-zA-Z0-9:._-]+)
(?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?
\s%(attribute)s\s*=\s*(?P<_q>['"]%(vqo)s)(?-x:%(value)s)(?P=_q)
''' % {'attribute': re.escape(attribute), 'value': value, 'vqo': value_quote_optional}
for m in re.finditer(partial_element_re, html):
content, whole = get_element_text_and_html_by_tag(m.group('tag'), html[m.start():])
yield (
unescapeHTML(re.sub(r'^(?P<q>["\'])(?P<content>.*)(?P=q)$', r'\g<content>', content, flags=re.DOTALL)),
whole
)
2016-03-16 16:50:04 +01:00
class HTMLBreakOnClosingTagParser(compat_HTMLParser):
"""
HTML parser which raises HTMLBreakOnClosingTagException upon reaching the
closing tag for the first opening tag it has encountered, and can be used
as a context manager
"""
class HTMLBreakOnClosingTagException(Exception):
pass
def __init__(self):
self.tagstack = collections.deque()
compat_HTMLParser.__init__(self)
def __enter__(self):
return self
def __exit__(self, *_):
self.close()
def close(self):
# handle_endtag does not return upon raising HTMLBreakOnClosingTagException,
# so data remains buffered; we no longer have any interest in it, thus
# override this method to discard it
pass
def handle_starttag(self, tag, _):
self.tagstack.append(tag)
def handle_endtag(self, tag):
if not self.tagstack:
raise compat_HTMLParseError('no tags in the stack')
while self.tagstack:
inner_tag = self.tagstack.pop()
if inner_tag == tag:
break
else:
raise compat_HTMLParseError(f'matching opening tag for closing {tag} tag not found')
if not self.tagstack:
raise self.HTMLBreakOnClosingTagException()
def get_element_text_and_html_by_tag(tag, html):
"""
For the first element with the specified tag in the passed HTML document
return its' content (text) and the whole element (html)
"""
def find_or_raise(haystack, needle, exc):
try:
return haystack.index(needle)
except ValueError:
raise exc
closing_tag = f'</{tag}>'
whole_start = find_or_raise(
html, f'<{tag}', compat_HTMLParseError(f'opening {tag} tag not found'))
content_start = find_or_raise(
html[whole_start:], '>', compat_HTMLParseError(f'malformed opening {tag} tag'))
content_start += whole_start + 1
with HTMLBreakOnClosingTagParser() as parser:
parser.feed(html[whole_start:content_start])
if not parser.tagstack or parser.tagstack[0] != tag:
raise compat_HTMLParseError(f'parser did not match opening {tag} tag')
offset = content_start
while offset < len(html):
next_closing_tag_start = find_or_raise(
html[offset:], closing_tag,
compat_HTMLParseError(f'closing {tag} tag not found'))
next_closing_tag_end = next_closing_tag_start + len(closing_tag)
try:
parser.feed(html[offset:offset + next_closing_tag_end])
offset += next_closing_tag_end
except HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException:
return html[content_start:offset + next_closing_tag_start], \
html[whole_start:offset + next_closing_tag_end]
raise compat_HTMLParseError('unexpected end of html')
class HTMLAttributeParser(compat_HTMLParser):
"""Trivial HTML parser to gather the attributes for a single element"""
2020-10-09 07:06:49 +02:00
def __init__(self):
2016-03-16 16:50:04 +01:00
self.attrs = {}
compat_HTMLParser.__init__(self)
def handle_starttag(self, tag, attrs):
self.attrs = dict(attrs)
2016-03-16 16:50:04 +01:00
class HTMLListAttrsParser(compat_HTMLParser):
"""HTML parser to gather the attributes for the elements of a list"""
def __init__(self):
compat_HTMLParser.__init__(self)
self.items = []
self._level = 0
def handle_starttag(self, tag, attrs):
if tag == 'li' and self._level == 0:
self.items.append(dict(attrs))
self._level += 1
def handle_endtag(self, tag):
self._level -= 1
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
2016-03-16 16:50:04 +01:00
def parse_list(webpage):
"""Given a string for an series of HTML <li> elements,
return a dictionary of their attributes"""
parser = HTMLListAttrsParser()
parser.feed(webpage)
parser.close()
return parser.items
def clean_html(html):
2012-11-28 02:04:46 +01:00
"""Clean an HTML snippet into a readable string"""
if html is None: # Convenience for sanitizing descriptions etc.
return html
html = re.sub(r'\s+', ' ', html)
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)
2012-11-28 02:04:46 +01:00
# Strip html tags
html = re.sub('<.*?>', '', html)
# Replace html entities
html = unescapeHTML(html)
2013-03-29 15:59:13 +01:00
return html.strip()
def sanitize_open(filename, open_mode):
2012-11-28 02:04:46 +01:00
"""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:
2014-11-17 07:16:12 +01:00
if filename == '-':
2012-11-28 02:04:46 +01:00
if sys.platform == 'win32':
import msvcrt
msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2013-03-28 13:13:03 +01:00
return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
stream = locked_file(filename, open_mode, block=False).open()
2012-11-28 02:04:46 +01:00
return (stream, filename)
except (IOError, OSError) as err:
if err.errno in (errno.EACCES,):
raise
2012-11-28 02:04:46 +01:00
# In case of error, try to remove win32 forbidden chars
2015-03-08 15:56:28 +01:00
alt_filename = sanitize_path(filename)
if alt_filename == filename:
raise
else:
# An exception here should be caught in the caller
stream = locked_file(filename, open_mode, block=False).open()
return (stream, alt_filename)
def timeconvert(timestr):
2012-11-28 02:04:46 +01:00
"""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
2012-11-26 23:58:46 +01:00
2014-11-23 20:41:03 +01:00
def sanitize_filename(s, restricted=False, is_id=False):
2012-11-28 02:04:46 +01:00
"""Sanitizes a string so it could be used as part of a filename.
If restricted is set, use a stricter subset of allowed characters.
2017-03-01 17:04:02 +01:00
Set is_id if this is not an arbitrary string, but an ID that should be kept
if possible.
2012-11-28 02:04:46 +01:00
"""
def replace_insane(char):
if restricted and char in ACCENT_CHARS:
return ACCENT_CHARS[char]
elif not restricted and char == '\n':
return ' '
elif char == '?' or ord(char) < 32 or ord(char) == 127:
2012-11-28 02:04:46 +01:00
return ''
elif char == '"':
return '' if restricted else '\''
elif char == ':':
return '_-' if restricted else ' -'
elif char in '\\/|*<>':
return '_'
2012-11-28 12:59:27 +01:00
if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()):
2012-11-28 02:04:46 +01:00
return '_'
if restricted and ord(char) > 127:
return '_'
return char
if s == '':
return ''
# Handle timestamps
s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
2014-11-17 07:16:12 +01:00
result = ''.join(map(replace_insane, s))
if not is_id:
while '__' in result:
result = result.replace('__', '_')
result = result.strip('_')
# Common case of "Foreign band name - English song title"
if restricted and result.startswith('-_'):
result = result[2:]
if result.startswith('-'):
result = '_' + result[len('-'):]
result = result.lstrip('.')
if not result:
result = '_'
2012-11-28 02:04:46 +01:00
return result
2014-11-23 20:41:03 +01:00
def sanitize_path(s, force=False):
2015-03-08 15:55:22 +01:00
"""Sanitizes and normalizes path on Windows"""
if sys.platform == 'win32':
force = False
drive_or_unc, _ = os.path.splitdrive(s)
if sys.version_info < (2, 7) and not drive_or_unc:
drive_or_unc, _ = os.path.splitunc(s)
elif force:
drive_or_unc = ''
else:
2015-03-08 15:55:22 +01:00
return s
norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep)
if drive_or_unc:
2015-03-08 15:55:22 +01:00
norm_path.pop(0)
sanitized_path = [
path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
2015-03-08 15:55:22 +01:00
for path_part in norm_path]
if drive_or_unc:
sanitized_path.insert(0, drive_or_unc + os.path.sep)
elif force and s[0] == os.path.sep:
sanitized_path.insert(0, os.path.sep)
2015-03-08 15:55:22 +01:00
return os.path.join(*sanitized_path)
2016-03-26 14:33:57 +01:00
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 = (
2019-03-09 13:14:41 +01:00
# 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
2016-03-26 14:33:57 +01:00
def extract_basic_auth(url):
parts = compat_urlparse.urlsplit(url)
if parts.username is None:
return url, None
url = compat_urlparse.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 ' + auth_payload.decode('utf-8')
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) >= 2 else kwargs.setdefault('headers', {})
headers['Authorization'] = auth_header
return compat_urllib_request.Request(url, *args, **kwargs)
2017-03-25 20:30:10 +01:00
def expand_path(s):
"""Expand shell variables and ~"""
return os.path.expandvars(compat_expanduser(s))
def orderedSet(iterable):
2012-11-28 02:04:46 +01:00
""" Remove all duplicates from the input iterable """
res = []
for el in iterable:
if el not in res:
res.append(el)
return res
2014-03-24 01:40:09 +01:00
def _htmlentity_transform(entity_with_semicolon):
2014-08-27 19:11:45 +02:00
"""Transforms an HTML entity to a character."""
entity = entity_with_semicolon[:-1]
2014-08-27 19:11:45 +02:00
# 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)
2014-08-27 19:11:45 +02:00
if mobj is not None:
numstr = mobj.group(1)
2014-11-17 07:16:12 +01:00
if numstr.startswith('x'):
2014-08-27 19:11:45 +02:00
base = 16
2014-11-17 07:16:12 +01:00
numstr = '0%s' % numstr
2014-08-27 19:11:45 +02:00
else:
base = 10
2019-03-09 13:14:41 +01:00
# See https://github.com/ytdl-org/youtube-dl/issues/7518
try:
return compat_chr(int(numstr, base))
except ValueError:
pass
2014-08-27 19:11:45 +02:00
# Unknown entity in name, return its literal representation
2015-11-16 15:24:09 +01:00
return '&%s;' % entity
2014-08-27 19:11:45 +02:00
def unescapeHTML(s):
2014-03-24 01:40:09 +01:00
if s is None:
return None
assert type(s) == compat_str
2014-08-27 19:11:45 +02:00
return re.sub(
r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
def escapeHTML(text):
return (
text
.replace('&', '&amp;')
.replace('<', '&lt;')
.replace('>', '&gt;')
.replace('"', '&quot;')
.replace("'", '&#39;')
)
def process_communicate_or_kill(p, *args, **kwargs):
try:
return p.communicate(*args, **kwargs)
except BaseException: # Including KeyboardInterrupt
p.kill()
p.wait()
raise
class Popen(subprocess.Popen):
if sys.platform == 'win32':
_startupinfo = subprocess.STARTUPINFO()
_startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW
else:
_startupinfo = None
def __init__(self, *args, **kwargs):
super(Popen, self).__init__(*args, **kwargs, startupinfo=self._startupinfo)
def communicate_or_kill(self, *args, **kwargs):
return process_communicate_or_kill(self, *args, **kwargs)
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
def encodeFilename(s, for_subprocess=False):
2012-11-28 02:04:46 +01:00
"""
@param s The name of the file
"""
assert type(s) == compat_str
2012-11-28 02:04:46 +01:00
# Python 3 has a Unicode API
if sys.version_info >= (3, 0):
return s
# 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:
return s
# Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
if sys.platform.startswith('java'):
return s
return s.encode(get_subprocess_encoding(), 'ignore')
def decodeFilename(b, for_subprocess=False):
if sys.version_info >= (3, 0):
return b
if not isinstance(b, bytes):
return b
return b.decode(get_subprocess_encoding(), 'ignore')
def encodeArgument(s):
if not isinstance(s, compat_str):
# Legacy code that uses byte strings
# Uncomment the following line after fixing all post processors
2014-12-17 00:06:41 +01:00
# 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
if isinstance(optval, bytes):
optval = optval.decode(preferredencoding())
assert isinstance(optval, compat_str)
return optval
2014-11-23 20:41:03 +01:00
_timetuple = collections.namedtuple('Time', ('hours', 'minutes', 'seconds', 'milliseconds'))
def timetuple_from_msec(msec):
secs, msec = divmod(msec, 1000)
mins, secs = divmod(secs, 60)
hrs, mins = divmod(mins, 60)
return _timetuple(hrs, mins, secs, msec)
def formatSeconds(secs, delim=':', msec=False):
time = timetuple_from_msec(secs * 1000)
if time.hours:
ret = '%d%s%02d%s%02d' % (time.hours, delim, time.minutes, delim, time.seconds)
elif time.minutes:
ret = '%d%s%02d' % (time.minutes, delim, time.seconds)
else:
ret = '%d' % time.seconds
return '%s.%03d' % (ret, time.milliseconds) if msec else ret
2013-12-29 15:28:32 +01:00
def _ssl_load_windows_store_certs(ssl_context, storename):
# Code adapted from _load_windows_store_certs in https://github.com/python/cpython/blob/main/Lib/ssl.py
try:
certs = [cert for cert, encoding, trust in ssl.enum_certificates(storename)
if encoding == 'x509_asn' and (
trust is True or ssl.Purpose.SERVER_AUTH.oid in trust)]
except PermissionError:
return
for cert in certs:
2014-12-13 23:27:21 +01:00
try:
ssl_context.load_verify_locations(cadata=cert)
except ssl.SSLError:
2014-12-13 23:27:21 +01:00
pass
def make_HTTPS_handler(params, **kwargs):
opts_check_certificate = not params.get('nocheckcertificate')
context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT)
context.check_hostname = opts_check_certificate
if params.get('legacyserverconnect'):
context.options |= 4 # SSL_OP_LEGACY_SERVER_CONNECT
context.verify_mode = ssl.CERT_REQUIRED if opts_check_certificate else ssl.CERT_NONE
if opts_check_certificate:
try:
context.load_default_certs()
# Work around the issue in load_default_certs when there are bad certificates. See:
# https://github.com/yt-dlp/yt-dlp/issues/1060,
# https://bugs.python.org/issue35665, https://bugs.python.org/issue45312
except ssl.SSLError:
# enum_certificates is not present in mingw python. See https://github.com/yt-dlp/yt-dlp/issues/1151
if sys.platform == 'win32' and hasattr(ssl, 'enum_certificates'):
# Create a new context to discard any certificates that were already loaded
context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT)
context.check_hostname, context.verify_mode = True, ssl.CERT_REQUIRED
for storename in ('CA', 'ROOT'):
_ssl_load_windows_store_certs(context, storename)
context.set_default_verify_paths()
return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
2013-05-04 12:19:02 +02:00
def bug_reports_message(before=';'):
msg = ('please report this issue on https://github.com/yt-dlp/yt-dlp , '
'filling out the "Broken site" issue template properly. '
'Confirm you are on the latest version using -U')
before = before.rstrip()
if not before or before.endswith(('.', '!', '?')):
msg = msg[0].title() + msg[1:]
return (before + ' ' if before else '') + msg
class YoutubeDLError(Exception):
"""Base exception for YoutubeDL errors."""
msg = None
def __init__(self, msg=None):
if msg is not None:
self.msg = msg
elif self.msg is None:
self.msg = type(self).__name__
super().__init__(self.msg)
2021-05-04 19:06:18 +02:00
network_exceptions = [compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error]
if hasattr(ssl, 'CertificateError'):
network_exceptions.append(ssl.CertificateError)
network_exceptions = tuple(network_exceptions)
class ExtractorError(YoutubeDLError):
"""Error during info extraction."""
2014-11-23 20:41:03 +01:00
def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None, ie=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 yt-dlp.
"""
2021-05-04 19:06:18 +02:00
if sys.exc_info()[0] in network_exceptions:
expected = True
self.orig_msg = str(msg)
self.traceback = tb
self.expected = expected
2013-08-28 04:25:38 +02:00
self.cause = cause
self.video_id = video_id
self.ie = ie
self.exc_info = sys.exc_info() # preserve original exception
super(ExtractorError, self).__init__(''.join((
format_field(ie, template='[%s] '),
format_field(video_id, template='%s: '),
msg,
format_field(cause, template=' (caused by %r)'),
'' if expected else bug_reports_message())))
def format_traceback(self):
if self.traceback is None:
return None
2014-11-17 07:16:12 +01:00
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.
"""
2020-10-09 07:06:49 +02:00
def __init__(self, msg, countries=None, **kwargs):
kwargs['expected'] = True
super(GeoRestrictedError, self).__init__(msg, **kwargs)
self.countries = countries
class DownloadError(YoutubeDLError):
2012-11-28 02:04:46 +01:00
"""Download Error exception.
2012-11-28 02:04:46 +01:00
This exception may be thrown by FileDownloader objects if they are not
configured to continue on errors. They will contain the appropriate
error message.
"""
2014-11-23 20:41:03 +01:00
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 EntryNotInPlaylist(YoutubeDLError):
"""Entry not in playlist exception.
This exception will be thrown by YoutubeDL when a requested entry
is not found in the playlist info_dict
"""
msg = 'Entry not found in info'
class SameFileError(YoutubeDLError):
2012-11-28 02:04:46 +01:00
"""Same File exception.
2012-11-28 02:04:46 +01:00
This exception will be thrown by FileDownloader objects if they detect
multiple files would have to be downloaded to the same file on disk.
"""
msg = 'Fixed output name but more than one file to download'
def __init__(self, filename=None):
if filename is not None:
self.msg += f': {filename}'
super().__init__(self.msg)
class PostProcessingError(YoutubeDLError):
2012-11-28 02:04:46 +01:00
"""Post Processing exception.
2012-11-28 02:04:46 +01:00
This exception may be raised by PostProcessor's .run() method to
indicate an error in the postprocessing task.
"""
2014-11-23 20:41:03 +01:00
class DownloadCancelled(YoutubeDLError):
""" Exception raised when the download queue should be interrupted """
msg = 'The download was cancelled'
class ExistingVideoReached(DownloadCancelled):
""" --break-on-existing triggered """
msg = 'Encountered a video that is already in the archive, stopping due to --break-on-existing'
class RejectedVideoReached(DownloadCancelled):
""" --break-on-reject triggered """
msg = 'Encountered a video that did not match filter, stopping due to --break-on-reject'
class MaxDownloadsReached(DownloadCancelled):
2012-11-28 02:04:46 +01:00
""" --max-downloads limit has been reached. """
msg = 'Maximum number of downloads reached, stopping due to --max-downloads'
class ReExtractInfo(YoutubeDLError):
""" Video info needs to be re-extracted. """
def __init__(self, msg, expected=False):
super().__init__(msg)
self.expected = expected
class ThrottledDownload(ReExtractInfo):
""" Download speed below --throttled-rate. """
msg = 'The download speed is below throttle limit'
2021-12-02 22:22:03 +01:00
def __init__(self):
super().__init__(self.msg, expected=False)
class UnavailableVideoError(YoutubeDLError):
2012-11-28 02:04:46 +01:00
"""Unavailable Format exception.
2012-11-28 02:04:46 +01:00
This exception will be thrown when a video is requested
in a format that is not available for that video.
"""
msg = 'Unable to download video'
def __init__(self, err=None):
if err is not None:
self.msg += f': {err}'
super().__init__(self.msg)
class ContentTooShortError(YoutubeDLError):
2012-11-28 02:04:46 +01:00
"""Content Too Short exception.
2012-11-28 02:04:46 +01:00
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.
"""
2012-11-28 02:04:46 +01:00
def __init__(self, downloaded, expected):
super(ContentTooShortError, self).__init__(
'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
)
# Both in bytes
2012-11-28 02:04:46 +01:00
self.downloaded = downloaded
self.expected = expected
2014-11-23 20:41:03 +01:00
class XAttrMetadataError(YoutubeDLError):
def __init__(self, code=None, msg='Unknown error'):
super(XAttrMetadataError, self).__init__(msg)
self.code = code
2016-10-01 21:03:41 +02:00
self.msg = msg
# Parsing code and msg
if (self.code in (errno.ENOSPC, errno.EDQUOT)
2020-11-21 15:50:42 +01:00
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
2019-03-09 13:14:41 +01:00
# https://github.com/ytdl-org/youtube-dl/issues/6727)
if sys.version_info < (3, 0):
2018-01-27 21:04:39 +01:00
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 = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding')
del filtered_headers['Youtubedl-no-compression']
return filtered_headers
class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
2012-11-28 02:04:46 +01:00
"""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
2012-11-28 02:04:46 +01:00
removed before making the real request.
Part of this code was copied from:
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):
2016-04-23 15:30:06 +02:00
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(
2016-04-23 15:30:06 +02:00
_create_http_connection, self, conn_class, False),
req)
2012-11-28 02:04:46 +01:00
@staticmethod
def deflate(data):
if not data:
return data
2012-11-28 02:04:46 +01:00
try:
return zlib.decompress(data, -zlib.MAX_WBITS)
except zlib.error:
return zlib.decompress(data)
def http_request(self, req):
# 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 = req.get_full_url()
url_escaped = escape_url(url)
# Substitute URL if any change after escaping
if url != url_escaped:
req = update_Request(req, url=url_escaped)
for h, v in self._params.get('http_headers', 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)
req.headers = handle_youtubedl_headers(req.headers)
if sys.version_info < (2, 7) and '#' 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]
2012-11-28 02:04:46 +01:00
return req
def http_response(self, req, resp):
2012-11-28 02:04:46 +01:00
old_resp = resp
# gzip
if resp.headers.get('Content-encoding', '') == 'gzip':
content = resp.read()
gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb')
try:
uncompressed = io.BytesIO(gz.read())
except IOError as original_ioerror:
# There may be junk add the end of the file
# See http://stackoverflow.com/q/4928560/35070 for details
for i in range(1, 1024):
try:
gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb')
uncompressed = io.BytesIO(gz.read())
except IOError:
continue
break
else:
raise original_ioerror
resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
2012-11-28 02:04:46 +01:00
resp.msg = old_resp.msg
[utils] Remove Content-encoding from headers after decompression With cn_verification_proxy, our http_response() is called twice, one from PerRequestProxyHandler.proxy_open() and another from normal YoutubeDL.urlopen(). As a result, for proxies honoring Accept-Encoding, the following bug occurs: $ youtube-dl -vs --cn-verification-proxy https://secure.uku.im:993 "test:letv" [debug] System config: [] [debug] User config: [] [debug] Command-line args: ['-vs', '--cn-verification-proxy', 'https://secure.uku.im:993', 'test:letv'] [debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8 [debug] youtube-dl version 2015.12.23 [debug] Git HEAD: 97f18fa [debug] Python version 3.5.1 - Linux-4.3.3-1-ARCH-x86_64-with-arch-Arch-Linux [debug] exe versions: ffmpeg 2.8.4, ffprobe 2.8.4, rtmpdump 2.4 [debug] Proxy map: {} [TestURL] Test URL: http://www.letv.com/ptv/vplay/22005890.html [Letv] 22005890: Downloading webpage [Letv] 22005890: Downloading playJson data ERROR: Unable to download JSON metadata: Not a gzipped file (b'{"') (caused by OSError('Not a gzipped file (b\'{"\')',)); please report this issue on https://yt-dl.org/bug . Make sure you are using the latest version; see https://yt-dl.org/update on how to update. Be sure to call youtube-dl with the --verbose flag and include its complete output. File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/extractor/common.py", line 330, in _request_webpage return self._downloader.urlopen(url_or_request) File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/YoutubeDL.py", line 1886, in urlopen return self._opener.open(req, timeout=self._socket_timeout) File "/usr/lib/python3.5/urllib/request.py", line 471, in open response = meth(req, response) File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/utils.py", line 773, in http_response raise original_ioerror File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/utils.py", line 761, in http_response uncompressed = io.BytesIO(gz.read()) File "/usr/lib/python3.5/gzip.py", line 274, in read return self._buffer.read(size) File "/usr/lib/python3.5/gzip.py", line 461, in read if not self._read_gzip_header(): File "/usr/lib/python3.5/gzip.py", line 409, in _read_gzip_header raise OSError('Not a gzipped file (%r)' % magic)
2015-12-27 18:09:18 +01:00
del resp.headers['Content-encoding']
2012-11-28 02:04:46 +01:00
# deflate
if resp.headers.get('Content-encoding', '') == 'deflate':
gz = io.BytesIO(self.deflate(resp.read()))
resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
2012-11-28 02:04:46 +01:00
resp.msg = old_resp.msg
[utils] Remove Content-encoding from headers after decompression With cn_verification_proxy, our http_response() is called twice, one from PerRequestProxyHandler.proxy_open() and another from normal YoutubeDL.urlopen(). As a result, for proxies honoring Accept-Encoding, the following bug occurs: $ youtube-dl -vs --cn-verification-proxy https://secure.uku.im:993 "test:letv" [debug] System config: [] [debug] User config: [] [debug] Command-line args: ['-vs', '--cn-verification-proxy', 'https://secure.uku.im:993', 'test:letv'] [debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8 [debug] youtube-dl version 2015.12.23 [debug] Git HEAD: 97f18fa [debug] Python version 3.5.1 - Linux-4.3.3-1-ARCH-x86_64-with-arch-Arch-Linux [debug] exe versions: ffmpeg 2.8.4, ffprobe 2.8.4, rtmpdump 2.4 [debug] Proxy map: {} [TestURL] Test URL: http://www.letv.com/ptv/vplay/22005890.html [Letv] 22005890: Downloading webpage [Letv] 22005890: Downloading playJson data ERROR: Unable to download JSON metadata: Not a gzipped file (b'{"') (caused by OSError('Not a gzipped file (b\'{"\')',)); please report this issue on https://yt-dl.org/bug . Make sure you are using the latest version; see https://yt-dl.org/update on how to update. Be sure to call youtube-dl with the --verbose flag and include its complete output. File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/extractor/common.py", line 330, in _request_webpage return self._downloader.urlopen(url_or_request) File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/YoutubeDL.py", line 1886, in urlopen return self._opener.open(req, timeout=self._socket_timeout) File "/usr/lib/python3.5/urllib/request.py", line 471, in open response = meth(req, response) File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/utils.py", line 773, in http_response raise original_ioerror File "/home/yen/Executables/Multimedia/youtube-dl/youtube_dl/utils.py", line 761, in http_response uncompressed = io.BytesIO(gz.read()) File "/usr/lib/python3.5/gzip.py", line 274, in read return self._buffer.read(size) File "/usr/lib/python3.5/gzip.py", line 461, in read if not self._read_gzip_header(): File "/usr/lib/python3.5/gzip.py", line 409, in _read_gzip_header raise OSError('Not a gzipped file (%r)' % magic)
2015-12-27 18:09:18 +01:00
del resp.headers['Content-encoding']
# Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2019-03-09 13:14:41 +01:00
# 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').decode('utf-8')
else:
location = location.decode('utf-8')
location_escaped = escape_url(location)
if location != location_escaped:
del resp.headers['Location']
if sys.version_info < (3, 0):
location_escaped = location_escaped.encode('utf-8')
resp.headers['Location'] = location_escaped
2012-11-28 02:04:46 +01:00
return resp
https_request = http_request
https_response = http_response
2014-02-06 11:29:46 +01:00
2016-04-23 15:30:06 +02:00
def make_socks_conn_class(base_class, socks_proxy):
assert issubclass(base_class, (
compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection))
url_components = compat_urlparse.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
2016-04-23 15:30:06 +02:00
def unquote_if_non_empty(s):
if not s:
return s
return compat_urllib_parse_unquote_plus(s)
2016-04-23 15:30:06 +02:00
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),
2016-04-23 15:30:06 +02:00
)
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 = {}
2016-04-23 15:30:06 +02:00
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
2016-04-23 15:30:06 +02:00
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(
2016-04-23 15:30:06 +02:00
_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 yt-dlp. 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
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
2019-03-09 13:14:41 +01:00
# 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 solves two issues:
- ensures redirect URL is always unicode under python 2
- introduces support for experimental HTTP response status code
308 Permanent Redirect [2] used by some sites [3]
1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py
2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308
3. https://github.com/ytdl-org/youtube-dl/issues/28768
"""
http_error_301 = http_error_303 = http_error_307 = 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.
"""
m = req.get_method()
if (not (code in (301, 302, 303, 307, 308) and m in ("GET", "HEAD")
or code in (301, 302, 303) and m == "POST")):
raise compat_HTTPError(req.full_url, code, msg, headers, fp)
# Strictly (according to RFC 2616), 301 or 302 in response to
# a POST MUST NOT cause a redirection without confirmation
# from the user (of urllib.request, in this case). In practice,
# essentially all clients do redirect in this case, so we do
# the same.
# On python 2 urlh.geturl() may sometimes return redirect URL
# as byte string instead of unicode. This workaround allows
# to force it always return unicode.
if sys.version_info[0] < 3:
newurl = 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')
CONTENT_HEADERS = ("content-length", "content-type")
# NB: don't use dict comprehension for python 2.6 compatibility
newheaders = dict((k, v) for k, v in req.headers.items()
if k.lower() not in CONTENT_HEADERS)
return compat_urllib_request.Request(
newurl, headers=newheaders, origin_req_host=req.origin_req_host,
unverifiable=True)
2016-06-25 17:30:35 +02:00
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)
2016-06-25 17:30:35 +02:00
if not m:
timezone = datetime.timedelta()
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):
2014-03-24 01:40:09 +01:00
""" Return a UNIX timestamp from the given date """
if date_str is None:
return None
2015-10-28 16:40:22 +01:00
date_str = re.sub(r'\.[0-9]+', '', date_str)
if timezone is None:
2016-06-25 17:30:35 +02:00
timezone, date_str = extract_timezone(date_str)
2015-10-28 16:40:22 +01:00
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
2014-03-24 01:40:09 +01:00
2016-06-25 17:30:35 +02:00
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"""
2014-03-21 14:38:37 +01:00
if date_str is None:
return None
upload_date = None
2014-11-23 20:41:03 +01:00
# Replace commas
2014-02-09 18:09:57 +01:00
date_str = date_str.replace(',', ' ')
# Remove AM/PM + timezone
2015-02-03 10:58:28 +01:00
date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
2016-06-25 17:30:35 +02:00
_, date_str = extract_timezone(date_str)
2016-06-25 17:30:35 +02:00
for expression in date_formats(day_first):
try:
upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
2014-02-06 11:29:46 +01:00
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)
2014-11-23 20:41:03 +01:00
2016-06-25 17:30:35 +02:00
def unified_timestamp(date_str, day_first=True):
if date_str is None:
return None
2017-06-11 16:27:22 +02:00
date_str = re.sub(r'[,|]', '', date_str)
2016-06-25 17:30:35 +02:00
pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
2016-06-25 17:30:35 +02:00
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]+$', date_str)
if m:
date_str = m.group(1)
2016-06-25 17:30:35 +02:00
for expression in date_formats(day_first):
try:
dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
2016-06-25 17:30:35 +02:00
return calendar.timegm(dt.timetuple())
except ValueError:
pass
timetuple = email.utils.parsedate_tz(date_str)
if timetuple:
return calendar.timegm(timetuple) + pm_delta * 3600
2016-06-25 17:30:35 +02:00
2014-11-17 07:16:12 +01:00
def determine_ext(url, default_ext='unknown_video'):
2018-06-01 19:16:22 +02:00
if url is None or '.' not in url:
2014-08-01 14:08:09 +02:00
return default_ext
guess = url.partition('?')[0].rpartition('.')[2]
2013-07-08 01:13:55 +02:00
if re.match(r'^[A-Za-z0-9]+$', 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('/')
2013-07-08 01:13:55 +02:00
else:
return default_ext
2013-07-08 01:13:55 +02:00
2014-11-23 20:41:03 +01:00
def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None):
return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext)
2014-11-23 20:41:03 +01:00
def datetime_from_str(date_str, precision='auto', format='%Y%m%d'):
"""
Return a datetime object from a string in the format YYYYMMDD or
(now|today|yesterday|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
format: string date format used to return datetime object from
precision: round the time portion of a datetime object.
auto|microsecond|second|minute|hour|day.
auto: round to the unit provided in date_str (if applicable).
"""
auto_precision = False
if precision == 'auto':
auto_precision = True
precision = 'microsecond'
today = datetime_round(datetime.datetime.utcnow(), precision)
if date_str in ('now', 'today'):
return today
if date_str == 'yesterday':
return today - datetime.timedelta(days=1)
match = re.match(
r'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)(s)?',
date_str)
if match is not None:
start_time = datetime_from_str(match.group('start'), precision, format)
time = int(match.group('time')) * (-1 if match.group('sign') == '-' else 1)
unit = match.group('unit')
if unit == 'month' or unit == 'year':
new_date = datetime_add_months(start_time, time * 12 if unit == 'year' else time)
unit = 'day'
else:
if unit == 'week':
unit = 'day'
time *= 7
delta = datetime.timedelta(**{unit + 's': time})
new_date = start_time + delta
if auto_precision:
return datetime_round(new_date, unit)
return new_date
return datetime_round(datetime.datetime.strptime(date_str, format), precision)
def date_from_str(date_str, format='%Y%m%d', strict=False):
"""
Return a datetime object from a string in the format YYYYMMDD or
(now|today|yesterday|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
If "strict", only (now|today)[+-][0-9](day|week|month|year)(s)? is allowed
format: string date format used to return datetime object from
"""
if strict and not re.fullmatch(r'\d{8}|(now|today)[+-]\d+(day|week|month|year)(s)?', date_str):
raise ValueError(f'Invalid date format {date_str}')
return datetime_from_str(date_str, precision='microsecond', format=format).date()
def datetime_add_months(dt, months):
"""Increment/Decrement a datetime object by months."""
month = dt.month + months - 1
year = dt.year + month // 12
month = month % 12 + 1
day = min(dt.day, calendar.monthrange(year, month)[1])
return dt.replace(year, month, day)
def datetime_round(dt, precision='day'):
"""
Round a datetime object's time to a specific precision
"""
if precision == 'microsecond':
return dt
unit_seconds = {
'day': 86400,
'hour': 3600,
'minute': 60,
'second': 1,
}
roundto = lambda x, n: ((x + n / 2) // n) * n
timestamp = calendar.timegm(dt.timetuple())
return datetime.datetime.utcfromtimestamp(roundto(timestamp, unit_seconds[precision]))
2014-11-23 20:41:03 +01:00
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
2014-11-23 20:41:03 +01:00
class DateRange(object):
"""Represents a time interval between two dates"""
2014-11-23 20:41:03 +01:00
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, strict=True)
else:
self.start = datetime.datetime.min.date()
if end is not None:
self.end = date_from_str(end, strict=True)
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)
2014-11-23 20:41:03 +01:00
@classmethod
def day(cls, day):
"""Returns a range that only contains the given day"""
2014-11-23 20:41:03 +01:00
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
2014-11-23 20:41:03 +01:00
def __str__(self):
2014-11-23 20:41:03 +01:00
return '%s - %s' % (self.start.isoformat(), self.end.isoformat())
def platform_name():
""" Returns the platform name as a compat_str """
res = platform.platform()
if isinstance(res, bytes):
res = res.decode(preferredencoding())
assert isinstance(res, compat_str)
return res
def get_windows_version():
''' Get Windows version. None if it's not running on Windows '''
if compat_os_name == 'nt':
return version_tuple(platform.win32_ver()[1])
else:
return None
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.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
2014-04-07 19:57:42 +02:00
def write_string(s, out=None, encoding=None):
if out is None:
out = sys.stderr
assert type(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()
2013-08-28 14:28:55 +02:00
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]
2013-08-28 15:59:07 +02:00
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, block):
overlapped = OVERLAPPED()
overlapped.Offset = 0
overlapped.OffsetHigh = 0
overlapped.hEvent = 0
f._lock_file_overlapped_p = ctypes.pointer(overlapped)
if not LockFileEx(msvcrt.get_osfhandle(f.fileno()),
(0x2 if exclusive else 0x0) | (0x0 if block else 0x1),
0, whole_low, whole_high, f._lock_file_overlapped_p):
raise BlockingIOError('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:
try:
import fcntl
def _lock_file(f, exclusive, block):
try:
fcntl.flock(f,
fcntl.LOCK_SH if not exclusive
else fcntl.LOCK_EX if block
else fcntl.LOCK_EX | fcntl.LOCK_NB)
except BlockingIOError:
raise
except OSError: # AOSP does not have flock()
fcntl.lockf(f,
fcntl.LOCK_SH if not exclusive
else fcntl.LOCK_EX if block
else fcntl.LOCK_EX | fcntl.LOCK_NB)
def _unlock_file(f):
try:
fcntl.flock(f, fcntl.LOCK_UN)
except OSError:
fcntl.lockf(f, fcntl.LOCK_UN)
except ImportError:
UNSUPPORTED_MSG = 'file locking is not supported on this platform'
def _lock_file(f, exclusive, block):
raise IOError(UNSUPPORTED_MSG)
def _unlock_file(f):
raise IOError(UNSUPPORTED_MSG)
class locked_file(object):
_closed = False
def __init__(self, filename, mode, block=True, encoding=None):
assert mode in ['r', 'rb', 'a', 'ab', 'w', 'wb']
self.f = io.open(filename, mode, encoding=encoding)
self.mode = mode
self.block = block
def __enter__(self):
exclusive = 'r' not in self.mode
try:
_lock_file(self.f, exclusive, self.block)
except IOError:
self.f.close()
raise
return self
def __exit__(self, etype, value, traceback):
try:
if not self._closed:
_unlock_file(self.f)
finally:
self.f.close()
self._closed = True
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 flush(self):
self.f.flush()
def open(self):
return self.__enter__()
def close(self, *args):
self.__exit__(self, *args, value=False, traceback=False)
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:
if isinstance(a, bytes):
# We may get a filename encoded with 'encodeFilename'
a = a.decode(encoding)
quoted_args.append(compat_shlex_quote(a))
2014-11-17 07:16:12 +01:00
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(
2014-11-17 07:16:12 +01:00
{'__youtubedl_smuggle': json.dumps(data)})
return url + '#' + sdata
def unsmuggle_url(smug_url, default=None):
2014-12-09 23:11:26 +01:00
if '#__youtubedl_smuggle' not in smug_url:
return smug_url, default
2014-11-17 07:16:12 +01:00
url, _, sdata = smug_url.rpartition('#')
jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
data = json.loads(jsond)
return url, data
def format_decimal_suffix(num, fmt='%d%s', *, factor=1000):
""" Formats numbers with decimal sufixes like K, M, etc """
num, factor = float_or_none(num), float(factor)
if num is None or num < 0:
return None
exponent = 0 if num == 0 else int(math.log(num, factor))
suffix = ['', *'kMGTPEZY'][exponent]
if factor == 1024:
suffix = {'k': 'Ki', '': ''}.get(suffix, f'{suffix}i')
converted = num / (factor ** exponent)
return fmt % (converted, suffix)
def format_bytes(bytes):
return format_decimal_suffix(bytes, '%.2f%sB', factor=1024) or 'N/A'
2013-12-09 18:29:07 +01:00
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)
2014-11-25 09:54:54 +01:00
def parse_filesize(s):
if s is None:
return None
2016-01-10 16:17:47 +01:00
# The lower-case forms are of course incorrect and unofficial,
2014-11-25 09:54:54 +01:00
# but we support those too
_UNIT_TABLE = {
'B': 1,
'b': 1,
'bytes': 1,
2014-11-25 09:54:54 +01:00
'KiB': 1024,
'KB': 1000,
'kB': 1024,
'Kb': 1000,
'kb': 1000,
'kilobytes': 1000,
'kibibytes': 1024,
2014-11-25 09:54:54 +01:00
'MiB': 1024 ** 2,
'MB': 1000 ** 2,
'mB': 1024 ** 2,
'Mb': 1000 ** 2,
'mb': 1000 ** 2,
'megabytes': 1000 ** 2,
'mebibytes': 1024 ** 2,
2014-11-25 09:54:54 +01:00
'GiB': 1024 ** 3,
'GB': 1000 ** 3,
'gB': 1024 ** 3,
'Gb': 1000 ** 3,
'gb': 1000 ** 3,
'gigabytes': 1000 ** 3,
'gibibytes': 1024 ** 3,
2014-11-25 09:54:54 +01:00
'TiB': 1024 ** 4,
'TB': 1000 ** 4,
'tB': 1024 ** 4,
'Tb': 1000 ** 4,
'tb': 1000 ** 4,
'terabytes': 1000 ** 4,
'tebibytes': 1024 ** 4,
2014-11-25 09:54:54 +01:00
'PiB': 1024 ** 5,
'PB': 1000 ** 5,
'pB': 1024 ** 5,
'Pb': 1000 ** 5,
'pb': 1000 ** 5,
'petabytes': 1000 ** 5,
'pebibytes': 1024 ** 5,
2014-11-25 09:54:54 +01:00
'EiB': 1024 ** 6,
'EB': 1000 ** 6,
'eB': 1024 ** 6,
'Eb': 1000 ** 6,
'eb': 1000 ** 6,
'exabytes': 1000 ** 6,
'exbibytes': 1024 ** 6,
2014-11-25 09:54:54 +01:00
'ZiB': 1024 ** 7,
'ZB': 1000 ** 7,
'zB': 1024 ** 7,
'Zb': 1000 ** 7,
'zb': 1000 ** 7,
'zettabytes': 1000 ** 7,
'zebibytes': 1024 ** 7,
2014-11-25 09:54:54 +01:00
'YiB': 1024 ** 8,
'YB': 1000 ** 8,
'yB': 1024 ** 8,
'Yb': 1000 ** 8,
'yb': 1000 ** 8,
'yottabytes': 1000 ** 8,
'yobibytes': 1024 ** 8,
2014-11-25 09:54:54 +01:00
}
return lookup_unit_table(_UNIT_TABLE, s)
def parse_count(s):
if s is None:
2014-11-25 09:54:54 +01:00
return None
2021-12-23 22:32:50 +01:00
s = re.sub(r'^[^\d]+\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,
2021-12-23 22:32:50 +01:00
'b': 1000 ** 3,
'B': 1000 ** 3,
}
2014-11-25 09:54:54 +01:00
2021-12-23 22:32:50 +01:00
ret = lookup_unit_table(_UNIT_TABLE, s)
if ret is not None:
return ret
mobj = re.match(r'([\d,.]+)(?:$|\s)', s)
if mobj:
return str_to_int(mobj.group(1))
2014-11-25 09:54:54 +01:00
2016-03-13 12:23:08 +01:00
2018-03-02 17:39:04 +01:00
def parse_resolution(s):
if s is None:
return {}
mobj = re.search(r'(?<![a-zA-Z0-9])(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)(?![a-zA-Z0-9])', s)
2018-03-02 17:39:04 +01:00
if mobj:
return {
'width': int(mobj.group('w')),
'height': int(mobj.group('h')),
}
mobj = re.search(r'(?<![a-zA-Z0-9])(\d+)[pPiI](?![a-zA-Z0-9])', s)
2018-03-02 17:39:04 +01:00
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 {}
2019-03-17 03:07:47 +01:00
def parse_bitrate(s):
if not isinstance(s, compat_str):
return
mobj = re.search(r'\b(\d+)\s*kbps', s)
if mobj:
return int(mobj.group(1))
def month_by_name(name, lang='en'):
2013-12-09 19:39:41 +01:00
""" Return the number of a month by (locale-independently) English name """
month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en'])
2013-12-09 19:39:41 +01:00
try:
return month_names.index(name) + 1
2015-02-13 08:14:23 +01:00
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
2013-12-09 19:39:41 +01:00
except ValueError:
return None
2013-12-10 21:03:53 +01:00
2014-01-20 22:11:34 +01:00
def fix_xml_ampersands(xml_str):
2013-12-10 21:03:53 +01:00
"""Replace all the '&' by '&amp;' in XML"""
2014-01-20 22:11:34 +01:00
return re.sub(
r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
2014-11-17 07:16:12 +01:00
'&amp;',
2014-01-20 22:11:34 +01:00
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:
2016-02-14 10:37:17 +01:00
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
2014-03-23 14:28:22 +01:00
title_bytes = title.encode('utf-8')
buf = ctypes.create_string_buffer(len(title_bytes))
buf.value = title_bytes
try:
2014-03-23 14:28:22 +01:00
libc.prctl(15, buf, 0, 0, 0)
except AttributeError:
return # Strange libc, just skip this
2013-12-16 13:56:13 +01:00
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
2015-12-14 16:30:58 +01:00
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
2020-10-09 07:06:49 +02:00
def get_domain(url):
domain = re.match(r'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url)
return domain.group('domain') if domain else None
def url_basename(url):
path = compat_urlparse.urlparse(url).path
2014-11-17 07:16:12 +01:00
return path.strip('/').split('/')[-1]
2013-12-20 17:05:28 +01:00
2016-11-01 20:14:01 +01:00
def base_url(url):
return re.match(r'https?://[^?#&]+/', url).group()
2016-12-12 20:23:49 +01:00
def urljoin(base, path):
if isinstance(path, bytes):
path = path.decode('utf-8')
2016-12-12 20:23:49 +01:00
if not isinstance(path, compat_str) or not path:
return None
if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path):
2016-12-12 20:23:49 +01:00
return path
if isinstance(base, bytes):
base = base.decode('utf-8')
if not isinstance(base, compat_str) or not re.match(
r'^(?:https?:)?//', base):
2016-12-12 20:23:49 +01:00
return None
return compat_urlparse.urljoin(base, path)
2013-12-20 17:05:28 +01:00
class HEADRequest(compat_urllib_request.Request):
def get_method(self):
2016-02-14 10:37:17 +01:00
return 'HEAD'
2016-07-02 21:21:32 +02:00
class PUTRequest(compat_urllib_request.Request):
def get_method(self):
return 'PUT'
2014-07-21 12:02:44 +02:00
def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
2022-01-03 20:37:24 +01:00
if get_attr and v is not None:
v = getattr(v, get_attr, None)
try:
return int(v) * invscale // scale
except (ValueError, TypeError, OverflowError):
return default
2014-07-21 12:02:44 +02:00
def str_or_none(v, default=None):
return default if v is None else compat_str(v)
2014-07-21 12:02:44 +02:00
def str_to_int(int_str):
2014-08-31 23:51:36 +02:00
""" A more relaxed version of int_or_none """
2019-12-15 17:15:24 +01:00
if isinstance(int_str, compat_integer_types):
return int_str
2019-12-15 17:15:24 +01:00
elif isinstance(int_str, compat_str):
int_str = re.sub(r'[,\.\+]', '', int_str)
return int_or_none(int_str)
2014-07-21 12:02:44 +02:00
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
2017-09-10 14:08:39 +02:00
def bool_or_none(v, default=None):
return v if isinstance(v, bool) else default
2019-05-23 18:58:35 +02:00
def strip_or_none(v, default=None):
return v.strip() if isinstance(v, compat_str) else default
2016-06-25 17:32:02 +02:00
2018-07-21 13:01:06 +02:00
def url_or_none(url):
if not url or not isinstance(url, compat_str):
return None
url = url.strip()
2021-01-01 13:26:37 +01:00
return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None
2018-07-21 13:01:06 +02:00
def request_to_url(req):
if isinstance(req, compat_urllib_request.Request):
return req.get_full_url()
else:
return req
def strftime_or_none(timestamp, date_format, default=None):
datetime_object = None
try:
if isinstance(timestamp, compat_numeric_types): # unix timestamp
datetime_object = datetime.datetime.utcfromtimestamp(timestamp)
elif isinstance(timestamp, compat_str): # assume YYYYMMDD
datetime_object = datetime.datetime.strptime(timestamp, '%Y%m%d')
return datetime_object.strftime(date_format)
except (ValueError, TypeError, AttributeError):
return default
def parse_duration(s):
2015-02-01 11:30:56 +01:00
if not isinstance(s, compat_basestring):
return None
2014-08-31 01:41:30 +02:00
s = s.strip()
if not s:
return None
2014-08-31 01:41:30 +02:00
days, hours, mins, secs, ms = [None] * 5
m = re.match(r'''(?x)
(?P<before_secs>
(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?
(?P<secs>(?(before_secs)[0-9]{1,2}|[0-9]+))
(?P<ms>[.:][0-9]+)?Z?$
''', s)
if m:
days, hours, mins, secs, ms = m.group('days', 'hours', 'mins', 'secs', 'ms')
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*
2017-01-26 17:23:08 +01:00
)?Z?$''', s)
if m:
days, hours, mins, secs, ms = m.groups()
else:
2017-01-26 17:23:08 +01:00
m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', 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.replace(':', '.'))
return duration
def prepend_extension(filename, ext, expected_real_ext=None):
2014-11-23 20:41:03 +01:00
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))
2015-05-02 19:23:06 +02:00
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:
Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate_or_kill()
except OSError:
return False
return exe
def _get_exe_version_output(exe, args):
try:
# STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
# SIGTTOU if yt-dlp is run in the background.
2019-03-09 13:14:41 +01:00
# See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
out, _ = Popen(
[encodeArgument(exe)] + args, stdin=subprocess.PIPE,
stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate_or_kill()
except OSError:
return False
if isinstance(out, bytes): # Python 2.x
out = out.decode('ascii', 'ignore')
return out
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
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 """
out = _get_exe_version_output(exe, args)
return detect_exe_version(out, version_re, unrecognized) if out else False
class LazyList(collections.abc.Sequence):
2021-05-28 18:49:13 +02:00
''' Lazy immutable list from an iterable
Note that slices of a LazyList are lists and not LazyList'''
2021-09-03 23:37:27 +02:00
class IndexError(IndexError):
pass
def __init__(self, iterable, *, reverse=False, _cache=None):
2021-05-28 18:49:13 +02:00
self.__iterable = iter(iterable)
self.__cache = [] if _cache is None else _cache
self.__reversed = reverse
2021-05-28 18:49:13 +02:00
def __iter__(self):
if self.__reversed:
# We need to consume the entire iterable to iterate in reverse
yield from self.exhaust()
return
yield from self.__cache
2021-05-28 18:49:13 +02:00
for item in self.__iterable:
self.__cache.append(item)
yield item
def __exhaust(self):
2021-05-28 18:49:13 +02:00
self.__cache.extend(self.__iterable)
# Discard the emptied iterable to make it pickle-able
self.__iterable = []
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 + 1)
2021-05-28 18:49:13 +02:00
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
2021-05-28 18:49:13 +02:00
elif isinstance(idx, int):
if self.__reversed:
idx = self.__reverse_index(idx)
start, stop, step = idx, idx, 0
2021-05-28 18:49:13 +02:00
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)):
2021-05-28 18:49:13 +02:00
# We need to consume the entire iterable to be able to slice from the end
# Obviously, never use this with infinite iterables
2021-09-03 23:37:27 +02:00
self.__exhaust()
try:
return self.__cache[idx]
except IndexError as e:
raise self.IndexError(e) from e
n = max(start or 0, stop or 0) - len(self.__cache) + 1
if n > 0:
self.__cache.extend(itertools.islice(self.__iterable, n))
2021-09-03 23:37:27 +02:00
try:
return self.__cache[idx]
except IndexError as e:
raise self.IndexError(e) from e
2021-05-28 18:49:13 +02:00
def __bool__(self):
try:
self[-1] if self.__reversed else self[0]
2021-09-03 23:37:27 +02:00
except self.IndexError:
2021-05-28 18:49:13 +02:00
return False
return True
def __len__(self):
2021-09-03 23:37:27 +02:00
self.__exhaust()
2021-05-28 18:49:13 +02:00
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())
2021-05-28 18:49:13 +02:00
class PagedList:
class IndexError(IndexError):
pass
2014-01-22 21:43:33 +01:00
def __len__(self):
# This is only useful for tests
return len(self.getslice())
def __init__(self, pagefunc, pagesize, use_cache=True):
self._pagefunc = pagefunc
self._pagesize = pagesize
self._pagecount = float('inf')
self._use_cache = use_cache
self._cache = {}
def getpage(self, pagenum):
2021-11-16 16:44:02 +01:00
page_results = self._cache.get(pagenum)
if page_results is None:
page_results = [] if pagenum > self._pagecount else list(self._pagefunc(pagenum))
if self._use_cache:
self._cache[pagenum] = page_results
return page_results
def getslice(self, start=0, end=None):
return list(self._getslice(start, end))
def _getslice(self, start, end):
raise NotImplementedError('This method must be implemented by subclasses')
def __getitem__(self, idx):
assert self._use_cache, 'Indexing PagedList requires cache'
if not isinstance(idx, int) or idx < 0:
raise TypeError('indices must be non-negative integers')
entries = self.getslice(idx, idx + 1)
2021-11-16 16:44:02 +01:00
if not entries:
raise self.IndexError()
2021-11-16 16:44:02 +01:00
return entries[0]
class OnDemandPagedList(PagedList):
def _getslice(self, start, end):
for pagenum in itertools.count(start // self._pagesize):
firstid = pagenum * self._pagesize
nextfirstid = pagenum * self._pagesize + self._pagesize
if start >= nextfirstid:
continue
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)
try:
page_results = self.getpage(pagenum)
except Exception:
self._pagecount = pagenum - 1
raise
if startv != 0 or endv is not None:
page_results = page_results[startv:endv]
yield from 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
class InAdvancePagedList(PagedList):
def __init__(self, pagefunc, pagecount, pagesize):
PagedList.__init__(self, pagefunc, pagesize, True)
self._pagecount = pagecount
def _getslice(self, start, end):
start_page = start // self._pagesize
2022-01-23 18:55:17 +01:00
end_page = self._pagecount if end is None else min(self._pagecount, 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_results = self.getpage(pagenum)
if skip_elems:
page_results = page_results[skip_elems:]
skip_elems = None
if only_more is not None:
if len(page_results) < only_more:
only_more -= len(page_results)
else:
yield from page_results[:only_more]
break
yield from page_results
def uppercase_escape(s):
2014-04-04 23:00:51 +02:00
unicode_escape = codecs.getdecoder('unicode_escape')
return re.sub(
r'\\U[0-9a-fA-F]{8}',
2014-04-04 23:00:51 +02:00
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)
2014-02-15 16:24:43 +01:00
def escape_rfc3986(s):
"""Escape non-ASCII characters as suggested by RFC 3986"""
2015-02-01 11:30:56 +01:00
if sys.version_info < (3, 0) and isinstance(s, compat_str):
s = s.encode('utf-8')
2014-11-02 11:37:49 +01:00
return 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()
2021-08-22 21:02:00 +02:00
def parse_qs(url):
return compat_parse_qs(compat_urllib_parse_urlparse(url).query)
def read_batch_urls(batch_fd):
def fixup(url):
if not isinstance(url, compat_str):
url = url.decode('utf-8', 'replace')
batch-file enumeration improvements (https://github.com/ytdl-org/youtube-dl/pull/26813) Co-authored by: glenn-slayden Modified from https://github.com/ytdl-org/youtube-dl/pull/26813/commits/c9a9ccf8a35e157e22afeaafc2851176ddd87e68 These improvements apply to reading the list of URLs from the file supplied via the `--batch-file` (`-a`) command line option. 1. Skip blank and empty lines in the file. Currently, lines with leading whitespace are only skipped when that whitespace is followed by a comment character (`#`, `;`, or `]`). This means that empty lines and lines consisting only of whitespace are returned as (trimmed) empty strings in the list of URLs to process. 2. [bug fix] Detect and remove the Unicode BOM when the file descriptor is already decoding Unicode. With Python 3, the `batch_fd` enumerator returns the lines of the file as Unicode. For UTF-8, this means that the raw BOM bytes from the file `\xef \xbb \xbf` show up converted into a single `\ufeff` character prefixed to the first enumerated text line. This fix solves several buggy interactions between the presence of BOM, the skipping of comments and/or blank lines, and ensuring the list of URLs is consistently trimmed. For example, if the first line of the file is blank, the BOM is incorrectly returned as a URL standing alone. If the first line contains a URL, it will be prefixed with this unwanted single character--but note that its being there will have inhibited the proper trimming of any leading whitespace. Currently, the `UnicodeBOMIE` helper attempts to recover from some of these error cases, but this fix prevents the error from happening in the first place (at least on Python3). In any case, the `UnicodeBOMIE` approach is flawed, because it is clearly illogical for a BOM to appear in the (non-batch) URL(s) specified directly on the command line (and for that matter, on URLs *after the first line* of a batch list, also) 3. Adds proper trimming of the " #" into the read_batch_urls processing so that the URLs it enumerates are cleaned and trimmed more consistently.
2021-01-09 13:38:03 +01:00
BOM_UTF8 = ('\xef\xbb\xbf', '\ufeff')
for bom in BOM_UTF8:
if url.startswith(bom):
url = url[len(bom):]
url = url.lstrip()
if not url or url.startswith(('#', ';', ']')):
return False
batch-file enumeration improvements (https://github.com/ytdl-org/youtube-dl/pull/26813) Co-authored by: glenn-slayden Modified from https://github.com/ytdl-org/youtube-dl/pull/26813/commits/c9a9ccf8a35e157e22afeaafc2851176ddd87e68 These improvements apply to reading the list of URLs from the file supplied via the `--batch-file` (`-a`) command line option. 1. Skip blank and empty lines in the file. Currently, lines with leading whitespace are only skipped when that whitespace is followed by a comment character (`#`, `;`, or `]`). This means that empty lines and lines consisting only of whitespace are returned as (trimmed) empty strings in the list of URLs to process. 2. [bug fix] Detect and remove the Unicode BOM when the file descriptor is already decoding Unicode. With Python 3, the `batch_fd` enumerator returns the lines of the file as Unicode. For UTF-8, this means that the raw BOM bytes from the file `\xef \xbb \xbf` show up converted into a single `\ufeff` character prefixed to the first enumerated text line. This fix solves several buggy interactions between the presence of BOM, the skipping of comments and/or blank lines, and ensuring the list of URLs is consistently trimmed. For example, if the first line of the file is blank, the BOM is incorrectly returned as a URL standing alone. If the first line contains a URL, it will be prefixed with this unwanted single character--but note that its being there will have inhibited the proper trimming of any leading whitespace. Currently, the `UnicodeBOMIE` helper attempts to recover from some of these error cases, but this fix prevents the error from happening in the first place (at least on Python3). In any case, the `UnicodeBOMIE` approach is flawed, because it is clearly illogical for a BOM to appear in the (non-batch) URL(s) specified directly on the command line (and for that matter, on URLs *after the first line* of a batch list, also) 3. Adds proper trimming of the " #" into the read_batch_urls processing so that the URLs it enumerates are cleaned and trimmed more consistently.
2021-01-09 13:38:03 +01:00
# "#" cannot be stripped out since it is part of the URI
# However, it can be safely stipped out if follwing a whitespace
return re.split(r'\s#', url, 1)[0].rstrip()
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')
2016-03-03 18:34:52 +01:00
def update_url_query(url, query):
if not query:
return url
2016-03-03 18:34:52 +01:00
parsed_url = compat_urlparse.urlparse(url)
qs = compat_parse_qs(parsed_url.query)
qs.update(query)
return compat_urlparse.urlunparse(parsed_url._replace(
query=compat_urllib_parse_urlencode(qs, True)))
2015-09-06 03:22:20 +02:00
2015-12-20 01:26:26 +01:00
2016-03-31 18:55:01 +02:00
def update_Request(req, url=None, data=None, headers={}, query={}):
req_headers = req.headers.copy()
req_headers.update(headers)
req_data = data or req.data
req_url = update_url_query(url or req.get_full_url(), query)
2016-07-02 21:21:32 +02:00
req_get_method = req.get_method()
if req_get_method == 'HEAD':
req_type = HEADRequest
elif req_get_method == 'PUT':
req_type = PUTRequest
else:
req_type = compat_urllib_request.Request
2016-03-31 18:55:01 +02:00
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'
if isinstance(k, compat_str):
k = k.encode('utf-8')
if isinstance(v, compat_str):
v = v.encode('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 dict_get(d, key_or_keys, default=None, skip_false_values=True):
if isinstance(key_or_keys, (list, tuple)):
for key in key_or_keys:
if key not in d or d[key] is None or skip_false_values and not d[key]:
continue
return d[key]
return default
return d.get(key_or_keys, default)
def try_get(src, getter, expected_type=None):
2021-07-10 23:59:44 +02:00
for get in variadic(getter):
try:
v = get(src)
except (AttributeError, KeyError, TypeError, IndexError):
pass
else:
if expected_type is None or isinstance(v, expected_type):
return v
2018-04-27 21:47:17 +02:00
def merge_dicts(*dicts):
merged = {}
for a_dict in dicts:
for k, v in a_dict.items():
if v is None:
continue
if (k not in merged
or (isinstance(v, compat_str) and v
and isinstance(merged[k], compat_str)
and not merged[k])):
2018-04-27 21:47:17 +02:00
merged[k] = v
return merged
2015-12-20 01:26:26 +01:00
def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)
2015-09-06 03:22:20 +02:00
2014-03-21 00:59:51 +01:00
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,
}
2014-10-03 14:37:25 +02:00
def parse_age_limit(s):
if type(s) == int:
return s if 0 <= s <= 21 else None
if not isinstance(s, compat_basestring):
return None
2014-10-03 14:37:25 +02:00
m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
if m:
return int(m.group('age'))
s = s.upper()
if s in US_RATINGS:
return US_RATINGS[s]
m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s)
if m:
return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)]
return None
2014-10-03 14:37:25 +02:00
def strip_jsonp(code):
2014-11-13 16:28:05 +01:00
return re.sub(
r'''(?sx)^
(?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
(?:\s*&&\s*(?P=func_name))?
\s*\(\s*(?P<callback_data>.*)\);?
\s*?(?://[^\n]*)*$''',
r'\g<callback_data>', code)
2014-04-21 07:12:02 +02:00
def js_to_json(code, vars={}):
# vars is a dict of var, val pairs to substitute
COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n'
SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
INTEGER_TABLE = (
(r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
(r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
)
2014-08-22 02:33:29 +02:00
def fix_kv(m):
2014-09-30 11:12:59 +02:00
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.startswith('!') or v == ',':
return ""
if v[0] in ("'", '"'):
v = re.sub(r'(?s)\\.|"', lambda m: {
2014-09-30 11:12:59 +02:00
'"': '\\"',
"\\'": "'",
'\\\n': '',
'\\x': '\\u00',
}.get(m.group(0), m.group(0)), v[1:-1])
else:
for regex, base in INTEGER_TABLE:
im = re.match(regex, v)
if im:
i = int(im.group(1), base)
return '"%d":' % i if v.endswith(':') else '%d' % i
if v in vars:
return vars[v]
2014-09-30 11:12:59 +02:00
return '"%s"' % v
2014-08-22 02:33:29 +02:00
code = re.sub(r'new Date\((".+")\)', r'\g<1>', code)
return re.sub(r'''(?sx)
"(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
'(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
{comment}|,(?={skip}[\]}}])|
void\s0|(?:(?<![0-9])[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*|
\b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
[0-9]+(?={skip}:)|
!+
'''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
2014-08-22 02:33:29 +02:00
2014-04-21 07:12:02 +02:00
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
POSTPROCESS_WHEN = {'pre_process', 'after_filter', 'before_dl', 'after_move', 'post_process', 'after_video', 'playlist'}
DEFAULT_OUTTMPL = {
'default': '%(title)s [%(id)s].%(ext)s',
'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s',
}
OUTTMPL_TYPES = {
'chapter': None,
'subtitle': None,
'thumbnail': None,
'description': 'description',
'annotation': 'annotations.xml',
'infojson': 'info.json',
'link': None,
'pl_video': None,
'pl_thumbnail': None,
'pl_description': 'description',
'pl_infojson': 'info.json',
}
# As of [1] format syntax is:
# %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
# 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
STR_FORMAT_RE_TMPL = r'''(?x)
(?<!%)(?P<prefix>(?:%%)*)
%
(?P<has_key>\((?P<key>{0})\))?
(?P<format>
(?P<conversion>[#0\-+ ]+)?
(?P<min_width>\d+)?
(?P<precision>\.\d+)?
(?P<len_mod>[hlL])? # unused in python
{1} # conversion type
)
'''
STR_FORMAT_TYPES = 'diouxXeEfFgGcrs'
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 yt-dlp can be updated with -U """
2021-01-06 12:58:30 +01:00
from .update import is_non_updateable
return not is_non_updateable()
def args_to_str(args):
# Get a short string representation for a subprocess command
return ' '.join(compat_shlex_quote(a) for a in args)
2015-01-04 02:20:45 +01:00
def error_to_compat_str(err):
2015-12-20 00:26:47 +01:00
err_str = str(err)
# On python 2 error byte string must be decoded with proper
# encoding rather than ascii
if sys.version_info[0] < 3:
err_str = err_str.decode(preferredencoding())
return err_str
2015-02-19 00:31:01 +01:00
def mimetype2ext(mt):
if mt is None:
return None
mt, _, params = mt.partition(';')
mt = mt.strip()
FULL_MAP = {
'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',
'audio/x-wav': 'wav',
'audio/wav': 'wav',
'audio/wave': 'wav',
}
ext = FULL_MAP.get(mt)
if ext is not None:
return ext
SUBTYPE_MAP = {
'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',
2016-07-14 15:13:57 +02:00
'vnd.ms-sstr+xml': 'ism',
2016-09-02 18:50:42 +02:00
'quicktime': 'mov',
2017-04-30 10:01:05 +02:00
'mp2t': 'ts',
2020-09-12 05:08:57 +02:00
'x-wav': 'wav',
'filmstrip+json': 'fs',
'svg+xml': 'svg',
}
_, _, subtype = mt.rpartition('/')
ext = SUBTYPE_MAP.get(subtype.lower())
if ext is not None:
return ext
SUFFIX_MAP = {
'json': 'json',
'xml': 'xml',
'zip': 'zip',
'gzip': 'gz',
}
_, _, suffix = subtype.partition('+')
ext = SUFFIX_MAP.get(suffix)
if ext is not None:
return ext
return subtype.replace('+', '.')
2015-02-19 00:31:01 +01:00
def ext2mimetype(ext_or_url):
if not ext_or_url:
return None
if '.' not in ext_or_url:
ext_or_url = f'file.{ext_or_url}'
return mimetypes.guess_type(ext_or_url)[0]
def parse_codecs(codecs_str):
# http://tools.ietf.org/html/rfc6381
if not codecs_str:
return {}
2020-11-21 15:50:42 +01:00
split_codecs = list(filter(None, map(
2021-07-31 12:51:01 +02:00
str.strip, codecs_str.strip().strip(',').split(','))))
vcodec, acodec, tcodec, hdr = None, None, None, None
2020-11-21 15:50:42 +01:00
for full_codec in split_codecs:
2021-11-03 11:56:34 +01:00
parts = full_codec.split('.')
codec = parts[0].replace('0', '')
if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2',
'h263', 'h264', 'mp4v', 'hvc1', 'av1', 'theora', 'dvh1', 'dvhe'):
if not vcodec:
vcodec = '.'.join(parts[:4]) if codec in ('vp9', 'av1', 'hvc1') else full_codec
2021-10-18 15:04:21 +02:00
if codec in ('dvh1', 'dvhe'):
hdr = 'DV'
2021-11-03 11:56:34 +01:00
elif codec == 'av1' and len(parts) > 3 and parts[3] == '10':
hdr = 'HDR10'
elif full_codec.replace('0', '').startswith('vp9.2'):
2021-10-18 15:04:21 +02:00
hdr = 'HDR10'
elif codec in ('flac', 'mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
if not acodec:
acodec = full_codec
elif codec in ('stpp', 'wvtt',):
if not tcodec:
tcodec = full_codec
else:
write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr)
if vcodec or acodec or tcodec:
return {
'vcodec': vcodec or 'none',
'acodec': acodec or 'none',
2021-10-18 15:04:21 +02:00
'dynamic_range': hdr,
**({'tcodec': tcodec} if tcodec is not None else {}),
}
elif len(split_codecs) == 2:
return {
'vcodec': split_codecs[0],
'acodec': split_codecs[1],
}
return {}
2015-01-04 02:20:45 +01:00
def urlhandle_detect_ext(url_handle):
getheader = url_handle.headers.get
2015-01-04 02:20:45 +01:00
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
2015-02-19 00:31:01 +01:00
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 = sanitize_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
2015-01-25 02:38:47 +01:00
def render_table(header_row, data, delim=False, extra_gap=0, hide_empty=False):
""" Render a list of rows, each as a list of values.
Text after a \t will be right aligned """
2021-10-20 18:37:32 +02:00
def width(string):
return len(remove_terminal_sequences(string).replace('\t', ''))
def get_max_lens(table):
2021-10-20 18:37:32 +02:00
return [max(width(str(v)) for v in col) for col in zip(*table)]
def filter_using_list(row, filterArray):
return [col for take, col in itertools.zip_longest(filterArray, row, fillvalue=True) if take]
max_lens = get_max_lens(data) if hide_empty else []
header_row = filter_using_list(header_row, max_lens)
data = [filter_using_list(row, max_lens) for row in data]
2015-01-25 02:38:47 +01:00
table = [header_row] + data
max_lens = get_max_lens(table)
extra_gap += 1
if delim:
table = [header_row, [delim * (ml + extra_gap) for ml in max_lens]] + data
table[1][-1] = table[1][-1][:-extra_gap * len(delim)] # Remove extra_gap from end of delimiter
2021-10-20 18:37:32 +02:00
for row in table:
for pos, text in enumerate(map(str, row)):
if '\t' in text:
row[pos] = text.replace('\t', ' ' * (max_lens[pos] - width(text))) + ' ' * extra_gap
else:
row[pos] = text + ' ' * (max_lens[pos] - width(text) + extra_gap)
ret = '\n'.join(''.join(row).rstrip() for row in table)
2021-10-20 18:37:32 +02:00
return ret
def _match_one(filter_part, dct, incomplete):
# TODO: Generalize code with YoutubeDL._build_format_filter
STRING_OPERATORS = {
'*=': operator.contains,
'^=': lambda attr, value: attr.startswith(value),
'$=': lambda attr, value: attr.endswith(value),
'~=': lambda attr, value: re.search(value, attr),
}
COMPARISON_OPERATORS = {
**STRING_OPERATORS,
'<=': operator.le, # "<=" must be defined above "<"
'<': operator.lt,
'>=': operator.ge,
'>': operator.gt,
'=': operator.eq,
}
operator_rex = re.compile(r'''(?x)\s*
(?P<key>[a-z_]+)
\s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?:
(?P<quote>["\'])(?P<quotedstrval>.+?)(?P=quote)|
(?P<strval>.+?)
)
\s*$
''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
m = operator_rex.search(filter_part)
if m:
m = m.groupdict()
unnegated_op = COMPARISON_OPERATORS[m['op']]
if m['negation']:
op = lambda attr, value: not unnegated_op(attr, value)
else:
op = unnegated_op
comparison_value = m['quotedstrval'] or m['strval'] or m['intval']
if m['quote']:
comparison_value = comparison_value.replace(r'\%s' % m['quote'], m['quote'])
actual_value = dct.get(m['key'])
numeric_comparison = None
if isinstance(actual_value, compat_numeric_types):
# 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)
try:
numeric_comparison = int(comparison_value)
except ValueError:
numeric_comparison = parse_filesize(comparison_value)
if numeric_comparison is None:
numeric_comparison = parse_filesize(f'{comparison_value}B')
if numeric_comparison is None:
numeric_comparison = parse_duration(comparison_value)
if numeric_comparison is not None and m['op'] in STRING_OPERATORS:
raise ValueError('Operator %s only supports string values!' % m['op'])
if actual_value is None:
return incomplete or m['none_inclusive']
return op(actual_value, comparison_value if numeric_comparison is None else numeric_comparison)
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'))
if incomplete and actual_value is None:
return True
return op(actual_value)
raise ValueError('Invalid filter part %r' % filter_part)
def match_str(filter_str, dct, incomplete=False):
""" Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false
When incomplete, all conditions passes on missing fields
"""
return all(
_match_one(filter_part.replace(r'\&', '&'), dct, incomplete)
for filter_part in re.split(r'(?<!\\)&', filter_str))
def match_filter_func(filter_str):
def _match_func(info_dict, *args, **kwargs):
if match_str(filter_str, info_dict, *args, **kwargs):
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' % timetuple_from_msec(seconds * 1000)
def ass_subtitles_timecode(seconds):
time = timetuple_from_msec(seconds * 1000)
return '%01d:%02d:%02d.%02d' % (*time[:-1], time.milliseconds / 10)
def dfxp2srt(dfxp_data):
'''
@param dfxp_data A bytes-like object containing DFXP data
@returns A unicode object containing converted SRT data
'''
2017-02-23 18:46:20 +01:00
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',
2017-02-23 18:46:20 +01:00
]),
(b'http://www.w3.org/ns/ttml#styling', [
b'http://www.w3.org/ns/ttml#style',
2017-02-23 18:46:20 +01:00
]),
)
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',
2017-02-23 18:46:20 +01:00
'tts': 'http://www.w3.org/ns/ttml#styling',
})
2017-02-23 18:46:20 +01:00
styles = {}
default_style = {}
class TTMLPElementParser(object):
2017-02-23 18:46:20 +01:00
_out = ''
_unclosed_elements = []
_applied_styles = []
def start(self, tag, attrib):
2017-02-23 18:46:20 +01:00
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):
2017-02-23 18:46:20 +01:00
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):
2017-02-23 18:46:20 +01:00
self._out += data
def close(self):
2017-02-23 18:46:20 +01:00
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()
2017-02-23 18:46:20 +01:00
for k, v in LEGACY_NAMESPACES:
for ns in v:
dfxp_data = dfxp_data.replace(ns, k)
dfxp = compat_etree_fromstring(dfxp_data)
out = []
2017-02-23 18:46:20 +01:00
paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
if not paras:
raise ValueError('Invalid dfxp/TTML subtitle')
2017-02-23 18:46:20 +01:00
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
2017-02-23 18:46:20 +01:00
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'))
2015-05-12 06:47:37 +02:00
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
2015-05-12 06:47:37 +02:00
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(argdict, keys, default=[], use_compat=True):
if isinstance(argdict, (list, tuple)): # for backward compatibility
if use_compat:
return argdict
else:
argdict = None
if argdict is None:
return default
assert isinstance(argdict, dict)
assert isinstance(keys, (list, tuple))
for key_list in keys:
arg_list = list(filter(
lambda x: x is not None,
2021-07-10 23:59:44 +02:00
[argdict.get(key.lower()) for key in variadic(key_list)]))
if arg_list:
return [arg for args in arg_list for arg in args]
return default
def _configuration_args(main_key, argdict, exe, keys=None, default=[], use_compat=True):
main_key, exe = main_key.lower(), exe.lower()
root_key = exe if main_key == exe else f'{main_key}+{exe}'
keys = [f'{root_key}{k}' for k in (keys or [''])]
if root_key in keys:
if main_key != exe:
keys.append((main_key, exe))
keys.append('default')
else:
use_compat = False
return cli_configuration_args(argdict, keys, default, use_compat)
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',
2019-01-06 19:02:34 +01:00
'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
2015-06-27 07:13:57 +02:00
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))
2018-07-29 00:52:42 +02:00
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_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
2016-04-23 15:30:06 +02:00
req.add_header('Ytdl-socks-proxy', proxy)
# yt-dlp's http/https handlers do wrapping the socket with socks
2016-04-23 15:30:06 +02:00
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
2016-02-24 15:08:40 +01:00
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):
2016-02-26 07:37:20 +01:00
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]
2016-02-24 15:08:40 +01:00
ret = ''
while num:
ret = table[num % n] + ret
num = num // n
return ret
def decode_packed_codes(code):
2016-10-19 18:28:49 +02:00
mobj = re.search(PACKED_CODES_RE, code)
2020-11-21 15:50:42 +01:00
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)],
2020-11-21 15:50:42 +01:00
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
2019-03-09 13:14:41 +01:00
# 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
2019-03-09 13:14:41 +01:00
# 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. '
'yt-dlp 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 = Popen(
cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
except EnvironmentError as e:
raise XAttrMetadataError(e.errno, e.strerror)
stdout, stderr = p.communicate_or_kill()
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),
}
2021-01-07 07:41:05 +01:00
# Templates for internet shortcut files, which are plain text files.
DOT_URL_LINK_TEMPLATE = '''
[InternetShortcut]
URL=%(url)s
'''.lstrip()
DOT_WEBLOC_LINK_TEMPLATE = '''
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
\t<key>URL</key>
\t<string>%(url)s</string>
</dict>
</plist>
'''.lstrip()
DOT_DESKTOP_LINK_TEMPLATE = '''
[Desktop Entry]
Encoding=UTF-8
Name=%(filename)s
Type=Link
URL=%(url)s
Icon=text-html
'''.lstrip()
LINK_TEMPLATES = {
'url': DOT_URL_LINK_TEMPLATE,
'desktop': DOT_DESKTOP_LINK_TEMPLATE,
'webloc': DOT_WEBLOC_LINK_TEMPLATE,
}
def iri_to_uri(iri):
"""
Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
The function doesn't add an additional layer of escaping; e.g., it doesn't escape `%3C` as `%253C`. Instead, it percent-escapes characters with an underlying UTF-8 encoding *besides* those already escaped, leaving the URI intact.
"""
iri_parts = compat_urllib_parse_urlparse(iri)
if '[' in iri_parts.netloc:
raise ValueError('IPv6 URIs are not, yet, supported.')
# Querying `.netloc`, when there's only one bracket, also raises a ValueError.
# The `safe` argument values, that the following code uses, contain the characters that should not be percent-encoded. Everything else but letters, digits and '_.-' will be percent-encoded with an underlying UTF-8 encoding. Everything already percent-encoded will be left as is.
net_location = ''
if iri_parts.username:
net_location += compat_urllib_parse_quote(iri_parts.username, safe=r"!$%&'()*+,~")
if iri_parts.password is not None:
net_location += ':' + compat_urllib_parse_quote(iri_parts.password, safe=r"!$%&'()*+,~")
net_location += '@'
net_location += iri_parts.hostname.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
# The 'idna' encoding produces ASCII text.
if iri_parts.port is not None and iri_parts.port != 80:
net_location += ':' + str(iri_parts.port)
return compat_urllib_parse_urlunparse(
(iri_parts.scheme,
net_location,
compat_urllib_parse_quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"),
# Unsure about the `safe` argument, since this is a legacy way of handling parameters.
compat_urllib_parse_quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"),
# Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
compat_urllib_parse_quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"),
compat_urllib_parse_quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~")))
# Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
def to_high_limit_path(path):
if sys.platform in ['win32', 'cygwin']:
# Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
return r'\\?\ '.rstrip() + os.path.abspath(path)
return path
2021-01-07 07:41:05 +01:00
2021-08-07 13:20:46 +02:00
def format_field(obj, field=None, template='%s', ignore=(None, ''), default='', func=None):
val = traverse_obj(obj, *variadic(field))
if val in ignore:
return default
return template % (func(val) if func else val)
2021-01-08 17:14:50 +01:00
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)
_HEX_TABLE = '0123456789abcdef'
def random_uuidv4():
return re.sub(r'[xy]', lambda x: _HEX_TABLE[random.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
def make_dir(path, to_screen=None):
try:
dn = os.path.dirname(path)
if dn and not os.path.exists(dn):
os.makedirs(dn)
return True
except (OSError, IOError) as err:
if callable(to_screen) is not None:
to_screen('unable to create directory ' + error_to_compat_str(err))
return False
def get_executable_path():
from zipimport import zipimporter
if hasattr(sys, 'frozen'): # Running from PyInstaller
path = os.path.dirname(sys.executable)
elif isinstance(globals().get('__loader__'), zipimporter): # Running from ZIP
path = os.path.join(os.path.dirname(__file__), '../..')
else:
path = os.path.join(os.path.dirname(__file__), '..')
return os.path.abspath(path)
def load_plugins(name, suffix, namespace):
classes = {}
try:
plugins_spec = importlib.util.spec_from_file_location(
name, os.path.join(get_executable_path(), 'ytdlp_plugins', name, '__init__.py'))
plugins = importlib.util.module_from_spec(plugins_spec)
sys.modules[plugins_spec.name] = plugins
plugins_spec.loader.exec_module(plugins)
for name in dir(plugins):
if name in namespace:
continue
if not name.endswith(suffix):
continue
klass = getattr(plugins, name)
classes[name] = namespace[name] = klass
except FileNotFoundError:
pass
return classes
2021-07-11 00:14:39 +02:00
def traverse_obj(
2021-07-21 07:47:27 +02:00
obj, *path_list, default=None, expected_type=None, get_all=True,
2021-07-11 00:14:39 +02:00
casesense=True, is_user_input=False, traverse_string=False):
''' Traverse nested list/dict/tuple
@param path_list A list of paths which are checked one by one.
Each path is a list of keys where each key is a string,
a function, a tuple of strings/None or "...".
When a fuction is given, it takes the key as argument and
returns whether the key matches or not. When a tuple is given,
all the keys given in the tuple are traversed, and
"..." traverses all the keys in the object
"None" returns the object without traversal
2021-07-11 00:14:39 +02:00
@param default Default value to return
2021-07-21 07:47:27 +02:00
@param expected_type Only accept final value of this type (Can also be any callable)
@param get_all Return all the values obtained from a path or only the first one
@param casesense Whether to consider dictionary keys as case sensitive
@param is_user_input Whether the keys are generated from user input. If True,
strings are converted to int/slice if necessary
@param traverse_string Whether to traverse inside strings. If True, any
non-compatible object will also be converted into a string
# TODO: Write tests
'''
2021-07-11 00:14:39 +02:00
if not casesense:
2021-07-31 12:51:01 +02:00
_lower = lambda k: (k.lower() if isinstance(k, str) else k)
path_list = (map(_lower, variadic(path)) for path in path_list)
def _traverse_obj(obj, path, _current_depth=0):
nonlocal depth
path = tuple(variadic(path))
for i, key in enumerate(path):
if None in (key, obj):
return obj
if isinstance(key, (list, tuple)):
obj = [_traverse_obj(obj, sub_key, _current_depth) for sub_key in key]
key = ...
if key is ...:
obj = (obj.values() if isinstance(obj, dict)
else obj if isinstance(obj, (list, tuple, LazyList))
else str(obj) if traverse_string else [])
_current_depth += 1
depth = max(depth, _current_depth)
return [_traverse_obj(inner_obj, path[i + 1:], _current_depth) for inner_obj in obj]
elif callable(key):
if isinstance(obj, (list, tuple, LazyList)):
obj = enumerate(obj)
elif isinstance(obj, dict):
obj = obj.items()
else:
if not traverse_string:
return None
obj = str(obj)
_current_depth += 1
depth = max(depth, _current_depth)
return [_traverse_obj(v, path[i + 1:], _current_depth) for k, v in obj if key(k)]
elif isinstance(obj, dict) and not (is_user_input and key == ':'):
2021-07-11 00:14:39 +02:00
obj = (obj.get(key) if casesense or (key in obj)
else next((v for k, v in obj.items() if _lower(k) == key), None))
else:
if is_user_input:
key = (int_or_none(key) if ':' not in key
else slice(*map(int_or_none, key.split(':'))))
if key == slice(None):
return _traverse_obj(obj, (..., *path[i + 1:]), _current_depth)
2021-07-11 00:14:39 +02:00
if not isinstance(key, (int, slice)):
return None
if not isinstance(obj, (list, tuple, LazyList)):
2021-07-11 00:14:39 +02:00
if not traverse_string:
return None
obj = str(obj)
try:
obj = obj[key]
except IndexError:
return None
2021-07-11 00:14:39 +02:00
return obj
2021-07-21 07:47:27 +02:00
if isinstance(expected_type, type):
type_test = lambda val: val if isinstance(val, expected_type) else None
elif expected_type is not None:
type_test = expected_type
else:
type_test = lambda val: val
for path in path_list:
depth = 0
val = _traverse_obj(obj, path)
2021-07-11 00:14:39 +02:00
if val is not None:
if depth:
for _ in range(depth - 1):
2021-07-20 19:46:38 +02:00
val = itertools.chain.from_iterable(v for v in val if v is not None)
2021-07-21 07:47:27 +02:00
val = [v for v in map(type_test, val) if v is not None]
if val:
2021-07-21 07:47:27 +02:00
return val if get_all else val[0]
else:
val = type_test(val)
if val is not None:
return val
2021-07-11 00:14:39 +02:00
return default
def traverse_dict(dictn, keys, casesense=True):
2021-11-29 18:46:06 +01:00
write_string('DeprecationWarning: yt_dlp.utils.traverse_dict is deprecated '
'and may be removed in a future version. Use yt_dlp.utils.traverse_obj instead')
return traverse_obj(dictn, keys, casesense=casesense, is_user_input=True, traverse_string=True)
2021-07-10 23:59:44 +02:00
def variadic(x, allowed_types=(str, bytes, dict)):
return x if isinstance(x, collections.abc.Iterable) and not isinstance(x, allowed_types) else (x,)
def decode_base(value, digits):
# This will convert given base-x string to scalar (long or int)
table = {char: index for index, char in enumerate(digits)}
result = 0
base = len(digits)
for chr in value:
result *= base
result += table[chr]
return result
def time_seconds(**kwargs):
t = datetime.datetime.now(datetime.timezone(datetime.timedelta(**kwargs)))
return t.timestamp()
# create a JSON Web Signature (jws) with HS256 algorithm
# the resulting format is in JWS Compact Serialization
# implemented following JWT https://www.rfc-editor.org/rfc/rfc7519.html
# implemented following JWS https://www.rfc-editor.org/rfc/rfc7515.html
def jwt_encode_hs256(payload_data, key, headers={}):
header_data = {
'alg': 'HS256',
'typ': 'JWT',
}
if headers:
header_data.update(headers)
header_b64 = base64.b64encode(json.dumps(header_data).encode('utf-8'))
payload_b64 = base64.b64encode(json.dumps(payload_data).encode('utf-8'))
h = hmac.new(key.encode('utf-8'), header_b64 + b'.' + payload_b64, hashlib.sha256)
signature_b64 = base64.b64encode(h.digest())
token = header_b64 + b'.' + payload_b64 + b'.' + signature_b64
return token
# can be extended in future to verify the signature and parse header and return the algorithm used if it's not HS256
def jwt_decode_hs256(jwt):
header_b64, payload_b64, signature_b64 = jwt.split('.')
payload_data = json.loads(base64.urlsafe_b64decode(payload_b64))
return payload_data
def supports_terminal_sequences(stream):
if compat_os_name == 'nt':
from .compat import WINDOWS_VT_MODE # Must be imported locally
if not WINDOWS_VT_MODE or get_windows_version() < (10, 0, 10586):
return False
elif not os.getenv('TERM'):
return False
try:
return stream.isatty()
except BaseException:
return False
2021-10-20 18:37:32 +02:00
_terminal_sequences_re = re.compile('\033\\[[^m]+m')
def remove_terminal_sequences(string):
return _terminal_sequences_re.sub('', string)
def number_of_digits(number):
return len('%d' % number)
2021-11-06 02:05:24 +01:00
def join_nonempty(*values, delim='-', from_dict=None):
if from_dict is not None:
2021-11-09 23:44:42 +01:00
values = map(from_dict.get, values)
2021-11-06 02:05:24 +01:00
return delim.join(map(str, filter(None, values)))
def scale_thumbnails_to_max_format_width(formats, thumbnails, url_width_re):
"""
Find the largest format dimensions in terms of video width and, for each thumbnail:
* Modify the URL: Match the width with the provided regex and replace with the former width
* Update dimensions
This function is useful with video services that scale the provided thumbnails on demand
"""
_keys = ('width', 'height')
max_dimensions = max(
[tuple(format.get(k) or 0 for k in _keys) for format in formats],
default=(0, 0))
if not max_dimensions[0]:
return thumbnails
return [
merge_dicts(
{'url': re.sub(url_width_re, str(max_dimensions[0]), thumbnail['url'])},
dict(zip(_keys, max_dimensions)), thumbnail)
for thumbnail in thumbnails
]
def parse_http_range(range):
""" Parse value of "Range" or "Content-Range" HTTP header into tuple. """
if not range:
return None, None, None
crg = re.search(r'bytes[ =](\d+)-(\d+)?(?:/(\d+))?', range)
if not crg:
return None, None, None
return int(crg.group(1)), int_or_none(crg.group(2)), int_or_none(crg.group(3))
class Config:
own_args = None
filename = None
__initialized = False
def __init__(self, parser, label=None):
self._parser, self.label = parser, label
self._loaded_paths, self.configs = set(), []
def init(self, args=None, filename=None):
assert not self.__initialized
directory = ''
if filename:
location = os.path.realpath(filename)
directory = os.path.dirname(location)
if location in self._loaded_paths:
return False
self._loaded_paths.add(location)
self.__initialized = True
self.own_args, self.filename = args, filename
for location in self._parser.parse_args(args)[0].config_locations or []:
location = os.path.join(directory, expand_path(location))
if os.path.isdir(location):
location = os.path.join(location, 'yt-dlp.conf')
if not os.path.exists(location):
self._parser.error(f'config location {location} does not exist')
self.append_config(self.read_file(location), location)
return True
def __str__(self):
label = join_nonempty(
self.label, 'config', f'"{self.filename}"' if self.filename else '',
delim=' ')
return join_nonempty(
self.own_args is not None and f'{label[0].upper()}{label[1:]}: {self.hide_login_info(self.own_args)}',
*(f'\n{c}'.replace('\n', '\n| ')[1:] for c in self.configs),
delim='\n')
@staticmethod
def read_file(filename, default=[]):
try:
optionf = open(filename)
except IOError:
return default # silently skip if file is not present
try:
# FIXME: https://github.com/ytdl-org/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56
contents = optionf.read()
if sys.version_info < (3,):
contents = contents.decode(preferredencoding())
res = compat_shlex_split(contents, comments=True)
finally:
optionf.close()
return res
@staticmethod
def hide_login_info(opts):
PRIVATE_OPTS = set(['-p', '--password', '-u', '--username', '--video-password', '--ap-password', '--ap-username'])
eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$')
def _scrub_eq(o):
m = eqre.match(o)
if m:
return m.group('key') + '=PRIVATE'
else:
return o
opts = list(map(_scrub_eq, opts))
for idx, opt in enumerate(opts):
if opt in PRIVATE_OPTS and idx + 1 < len(opts):
opts[idx + 1] = 'PRIVATE'
return opts
def append_config(self, *args, label=None):
config = type(self)(self._parser, label)
config._loaded_paths = self._loaded_paths
if config.init(*args):
self.configs.append(config)
@property
def all_args(self):
for config in reversed(self.configs):
yield from config.all_args
yield from self.own_args or []
def parse_args(self):
return self._parser.parse_args(list(self.all_args))
class WebSocketsWrapper():
"""Wraps websockets module to use in non-async scopes"""
def __init__(self, url, headers=None):
self.loop = asyncio.events.new_event_loop()
self.conn = compat_websockets.connect(
url, extra_headers=headers, ping_interval=None,
close_timeout=float('inf'), loop=self.loop, ping_timeout=float('inf'))
atexit.register(self.__exit__, None, None, None)
def __enter__(self):
self.pool = self.run_with_loop(self.conn.__aenter__(), self.loop)
return self
def send(self, *args):
self.run_with_loop(self.pool.send(*args), self.loop)
def recv(self, *args):
return self.run_with_loop(self.pool.recv(*args), self.loop)
def __exit__(self, type, value, traceback):
try:
return self.run_with_loop(self.conn.__aexit__(type, value, traceback), self.loop)
finally:
self.loop.close()
self._cancel_all_tasks(self.loop)
# taken from https://github.com/python/cpython/blob/3.9/Lib/asyncio/runners.py with modifications
# for contributors: If there's any new library using asyncio needs to be run in non-async, move these function out of this class
@staticmethod
def run_with_loop(main, loop):
if not asyncio.coroutines.iscoroutine(main):
raise ValueError(f'a coroutine was expected, got {main!r}')
try:
return loop.run_until_complete(main)
finally:
loop.run_until_complete(loop.shutdown_asyncgens())
if hasattr(loop, 'shutdown_default_executor'):
loop.run_until_complete(loop.shutdown_default_executor())
@staticmethod
def _cancel_all_tasks(loop):
to_cancel = asyncio.tasks.all_tasks(loop)
if not to_cancel:
return
for task in to_cancel:
task.cancel()
loop.run_until_complete(
asyncio.tasks.gather(*to_cancel, loop=loop, return_exceptions=True))
for task in to_cancel:
if task.cancelled():
continue
if task.exception() is not None:
loop.call_exception_handler({
'message': 'unhandled exception during asyncio.run() shutdown',
'exception': task.exception(),
'task': task,
})
has_websockets = bool(compat_websockets)
def merge_headers(*dicts):
"""Merge dicts of network headers case insensitively, prioritizing the latter ones"""
return {k.capitalize(): v for k, v in itertools.chain.from_iterable(map(dict.items, dicts))}