4 from __future__
import unicode_literals
39 import xml
.etree
.ElementTree
43 compat_HTMLParseError
,
49 compat_ctypes_WINFUNCTYPE
,
50 compat_etree_fromstring
,
53 compat_html_entities_html5
,
66 compat_urllib_parse_urlencode
,
67 compat_urllib_parse_urlparse
,
68 compat_urllib_parse_urlunparse
,
69 compat_urllib_parse_quote
,
70 compat_urllib_parse_quote_plus
,
71 compat_urllib_parse_unquote_plus
,
72 compat_urllib_request
,
83 def register_socks_protocols():
84 # "Register" SOCKS protocols
85 # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
86 # URLs with protocols not in urlparse.uses_netloc are not handled correctly
87 for scheme
in ('socks', 'socks4', 'socks4a', 'socks5'):
88 if scheme
not in compat_urlparse
.uses_netloc
:
89 compat_urlparse
.uses_netloc
.append(scheme
)
92 # This is not clearly defined otherwise
93 compiled_regex_type
= type(re
.compile(''))
96 def random_user_agent():
97 _USER_AGENT_TPL
= 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
1676 return _USER_AGENT_TPL
% random
.choice(_CHROME_VERSIONS
)
1680 'User-Agent': random_user_agent(),
1681 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
1682 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
1683 'Accept-Encoding': 'gzip, deflate',
1684 'Accept-Language': 'en-us,en;q=0.5',
1689 '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',
1693 NO_DEFAULT
= object()
1695 ENGLISH_MONTH_NAMES
= [
1696 'January', 'February', 'March', 'April', 'May', 'June',
1697 'July', 'August', 'September', 'October', 'November', 'December']
1700 'en': ENGLISH_MONTH_NAMES
,
1702 'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
1703 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
1706 KNOWN_EXTENSIONS
= (
1707 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
1708 'flv', 'f4v', 'f4a', 'f4b',
1709 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
1710 'mkv', 'mka', 'mk3d',
1713 'asf', 'wmv', 'wma',
1719 'f4f', 'f4m', 'm3u8', 'smil')
1721 # needed for sanitizing filenames in restricted mode
1722 ACCENT_CHARS
= dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
1723 itertools
.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1724 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1748 '%Y/%m/%d %H:%M:%S',
1752 '%Y-%m-%d %H:%M:%S',
1753 '%Y-%m-%d %H:%M:%S.%f',
1754 '%Y-%m-%d %H:%M:%S:%f',
1757 '%Y-%m-%dT%H:%M:%SZ',
1758 '%Y-%m-%dT%H:%M:%S.%fZ',
1759 '%Y-%m-%dT%H:%M:%S.%f0Z',
1760 '%Y-%m-%dT%H:%M:%S',
1761 '%Y-%m-%dT%H:%M:%S.%f',
1763 '%b %d %Y at %H:%M',
1764 '%b %d %Y at %H:%M:%S',
1765 '%B %d %Y at %H:%M',
1766 '%B %d %Y at %H:%M:%S',
1770 DATE_FORMATS_DAY_FIRST
= list(DATE_FORMATS
)
1771 DATE_FORMATS_DAY_FIRST
.extend([
1777 '%d/%m/%Y %H:%M:%S',
1780 DATE_FORMATS_MONTH_FIRST
= list(DATE_FORMATS
)
1781 DATE_FORMATS_MONTH_FIRST
.extend([
1786 '%m/%d/%Y %H:%M:%S',
1789 PACKED_CODES_RE
= r
"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1790 JSON_LD_RE
= r
'(?is)<script[^>]+type=(["\']?
)application
/ld\
+json\
1[^
>]*>(?P
<json_ld
>.+?
)</script
>'
1793 def preferredencoding():
1794 """Get preferred encoding.
1796 Returns the best encoding scheme for the system, based on
1797 locale.getpreferredencoding() and some further tweaks.
1800 pref = locale.getpreferredencoding()
1808 def write_json_file(obj, fn):
1809 """ Encode obj as JSON and write it to fn, atomically if possible """
1811 fn = encodeFilename(fn)
1812 if sys.version_info < (3, 0) and sys.platform != 'win32
':
1813 encoding = get_filesystem_encoding()
1814 # os.path.basename returns a bytes object, but NamedTemporaryFile
1815 # will fail if the filename contains non ascii characters unless we
1816 # use a unicode object
1817 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1818 # the same for os.path.dirname
1819 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1821 path_basename = os.path.basename
1822 path_dirname = os.path.dirname
1826 'prefix
': path_basename(fn) + '.',
1827 'dir': path_dirname(fn),
1831 # In Python 2.x, json.dump expects a bytestream.
1832 # In Python 3.x, it writes to a character stream
1833 if sys.version_info < (3, 0):
1838 'encoding
': 'utf
-8',
1841 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1846 if sys.platform == 'win32
':
1847 # Need to remove existing file on Windows, else os.rename raises
1848 # WindowsError or FileExistsError.
1856 os.chmod(tf.name, 0o666 & ~mask)
1859 os.rename(tf.name, fn)
1868 if sys.version_info >= (2, 7):
1869 def find_xpath_attr(node, xpath, key, val=None):
1870 """ Find the xpath xpath[@key=val] """
1871 assert re.match(r'^
[a
-zA
-Z_
-]+$
', key)
1872 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1873 return node.find(expr)
1875 def find_xpath_attr(node, xpath, key, val=None):
1876 for f in node.findall(compat_xpath(xpath)):
1877 if key not in f.attrib:
1879 if val is None or f.attrib.get(key) == val:
1883 # On python2.6 the xml.etree.ElementTree.Element methods don't support
1884 # the namespace parameter
1887 def xpath_with_ns(path
, ns_map
):
1888 components
= [c
.split(':') for c
in path
.split('/')]
1890 for c
in components
:
1892 replaced
.append(c
[0])
1895 replaced
.append('{%s}%s' % (ns_map
[ns
], tag
))
1896 return '/'.join(replaced
)
1899 def xpath_element(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1900 def _find_xpath(xpath
):
1901 return node
.find(compat_xpath(xpath
))
1903 if isinstance(xpath
, (str, compat_str
)):
1904 n
= _find_xpath(xpath
)
1912 if default
is not NO_DEFAULT
:
1915 name
= xpath
if name
is None else name
1916 raise ExtractorError('Could not find XML element %s' % name
)
1922 def xpath_text(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1923 n
= xpath_element(node
, xpath
, name
, fatal
=fatal
, default
=default
)
1924 if n
is None or n
== default
:
1927 if default
is not NO_DEFAULT
:
1930 name
= xpath
if name
is None else name
1931 raise ExtractorError('Could not find XML element\'s text %s' % name
)
1937 def xpath_attr(node
, xpath
, key
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1938 n
= find_xpath_attr(node
, xpath
, key
)
1940 if default
is not NO_DEFAULT
:
1943 name
= '%s[@%s]' % (xpath
, key
) if name
is None else name
1944 raise ExtractorError('Could not find XML attribute %s' % name
)
1947 return n
.attrib
[key
]
1950 def get_element_by_id(id, html
):
1951 """Return the content of the tag with the specified ID in the passed HTML document"""
1952 return get_element_by_attribute('id', id, html
)
1955 def get_element_by_class(class_name
, html
):
1956 """Return the content of the first tag with the specified class in the passed HTML document"""
1957 retval
= get_elements_by_class(class_name
, html
)
1958 return retval
[0] if retval
else None
1961 def get_element_by_attribute(attribute
, value
, html
, escape_value
=True):
1962 retval
= get_elements_by_attribute(attribute
, value
, html
, escape_value
)
1963 return retval
[0] if retval
else None
1966 def get_elements_by_class(class_name
, html
):
1967 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1968 return get_elements_by_attribute(
1969 'class', r
'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1970 html, escape_value=False)
1973 def get_elements_by_attribute(attribute, value, html, escape_value=True):
1974 """Return the content of the tag with the specified attribute in the passed HTML document"""
1976 value = re.escape(value) if escape_value else value
1979 for m in re.finditer(r'''(?xs)
1981 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^
']*'|
))*?
1983 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^
"]*"|
='[^']*'|))*?
1987 ''' % (re.escape(attribute), value), html):
1988 res = m.group('content
')
1990 if res.startswith('"') or res.startswith("'"):
1993 retlist.append(unescapeHTML(res))
1998 class HTMLAttributeParser(compat_HTMLParser):
1999 """Trivial HTML parser to gather the attributes for a single element"""
2003 compat_HTMLParser.__init__(self)
2005 def handle_starttag(self, tag, attrs):
2006 self.attrs = dict(attrs)
2009 def extract_attributes(html_element):
2010 """Given a string for an HTML element such as
2012 a="foo" B="bar" c="&98;az" d=boz
2013 empty= noval entity="&"
2016 Decode and return a dictionary of attributes.
2018 'a
': 'foo
', 'b
': 'bar
', c: 'baz
', d: 'boz
',
2019 'empty
': '', 'noval
': None, 'entity
': '&',
2020 'sq
': '"', 'dq': '\''
2022 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2023 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2025 parser = HTMLAttributeParser()
2027 parser.feed(html_element)
2029 # Older Python may throw HTMLParseError in case of malformed HTML
2030 except compat_HTMLParseError:
2035 def clean_html(html):
2036 """Clean an HTML snippet into a readable string"""
2038 if html is None: # Convenience for sanitizing descriptions etc.
2042 html = html.replace('\n', ' ')
2043 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2044 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2046 html = re.sub('<.*?>', '', html)
2047 # Replace html entities
2048 html = unescapeHTML(html)
2052 def sanitize_open(filename, open_mode):
2053 """Try to open the given filename, and slightly tweak it if this fails.
2055 Attempts to open the given filename. If this fails, it tries to change
2056 the filename slightly, step by step, until it's either able to open it
2057 or it fails and raises a final exception, like the standard open()
2060 It returns the tuple (stream, definitive_file_name).
2064 if sys.platform == 'win32':
2066 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2067 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2068 stream = open(encodeFilename(filename), open_mode)
2069 return (stream, filename)
2070 except (IOError, OSError) as err:
2071 if err.errno in (errno.EACCES,):
2074 # In case of error, try to remove win32 forbidden chars
2075 alt_filename = sanitize_path(filename)
2076 if alt_filename == filename:
2079 # An exception here should be caught in the caller
2080 stream = open(encodeFilename(alt_filename), open_mode)
2081 return (stream, alt_filename)
2084 def timeconvert(timestr):
2085 """Convert RFC 2822 defined time string into system timestamp"""
2087 timetuple = email.utils.parsedate_tz(timestr)
2088 if timetuple is not None:
2089 timestamp = email.utils.mktime_tz(timetuple)
2093 def sanitize_filename(s, restricted=False, is_id=False):
2094 """Sanitizes a string so it could be used as part of a filename.
2095 If restricted is set, use a stricter subset of allowed characters.
2096 Set is_id if this is not an arbitrary string, but an ID that should be kept
2099 def replace_insane(char):
2100 if restricted and char in ACCENT_CHARS:
2101 return ACCENT_CHARS[char]
2102 elif not restricted and char == '\n':
2104 elif char == '?' or ord(char) < 32 or ord(char) == 127:
2107 return '' if restricted else '\''
2109 return '_
-' if restricted else ' -'
2110 elif char in '\\/|
*<>':
2112 if restricted and (char in '!&\'()[]{}$
;`^
,#' or char.isspace()):
2114 if restricted
and ord(char
) > 127:
2121 s
= re
.sub(r
'[0-9]+(?::[0-9]+)+', lambda m
: m
.group(0).replace(':', '_'), s
)
2122 result
= ''.join(map(replace_insane
, s
))
2124 while '__' in result
:
2125 result
= result
.replace('__', '_')
2126 result
= result
.strip('_')
2127 # Common case of "Foreign band name - English song title"
2128 if restricted
and result
.startswith('-_'):
2130 if result
.startswith('-'):
2131 result
= '_' + result
[len('-'):]
2132 result
= result
.lstrip('.')
2138 def sanitize_path(s
, force
=False):
2139 """Sanitizes and normalizes path on Windows"""
2140 if sys
.platform
== 'win32':
2142 drive_or_unc
, _
= os
.path
.splitdrive(s
)
2143 if sys
.version_info
< (2, 7) and not drive_or_unc
:
2144 drive_or_unc
, _
= os
.path
.splitunc(s
)
2150 norm_path
= os
.path
.normpath(remove_start(s
, drive_or_unc
)).split(os
.path
.sep
)
2154 path_part
if path_part
in ['.', '..'] else re
.sub(r
'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part
)
2155 for path_part
in norm_path
]
2157 sanitized_path
.insert(0, drive_or_unc
+ os
.path
.sep
)
2158 elif force
and s
[0] == os
.path
.sep
:
2159 sanitized_path
.insert(0, os
.path
.sep
)
2160 return os
.path
.join(*sanitized_path
)
2163 def sanitize_url(url
):
2164 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2165 # the number of unwanted failures due to missing protocol
2166 if url
.startswith('//'):
2167 return 'http:%s' % url
2168 # Fix some common typos seen so far
2170 # https://github.com/ytdl-org/youtube-dl/issues/15649
2171 (r
'^httpss://', r
'https://'),
2172 # https://bx1.be/lives/direct-tv/
2173 (r
'^rmtp([es]?)://', r
'rtmp\1://'),
2175 for mistake
, fixup
in COMMON_TYPOS
:
2176 if re
.match(mistake
, url
):
2177 return re
.sub(mistake
, fixup
, url
)
2181 def extract_basic_auth(url
):
2182 parts
= compat_urlparse
.urlsplit(url
)
2183 if parts
.username
is None:
2185 url
= compat_urlparse
.urlunsplit(parts
._replace
(netloc
=(
2186 parts
.hostname
if parts
.port
is None
2187 else '%s:%d' % (parts
.hostname
, parts
.port
))))
2188 auth_payload
= base64
.b64encode(
2189 ('%s:%s' % (parts
.username
, parts
.password
or '')).encode('utf-8'))
2190 return url
, 'Basic ' + auth_payload
.decode('utf-8')
2193 def sanitized_Request(url
, *args
, **kwargs
):
2194 url
, auth_header
= extract_basic_auth(escape_url(sanitize_url(url
)))
2195 if auth_header
is not None:
2196 headers
= args
[1] if len(args
) >= 2 else kwargs
.setdefault('headers', {})
2197 headers
['Authorization'] = auth_header
2198 return compat_urllib_request
.Request(url
, *args
, **kwargs
)
2202 """Expand shell variables and ~"""
2203 return os
.path
.expandvars(compat_expanduser(s
))
2206 def orderedSet(iterable
):
2207 """ Remove all duplicates from the input iterable """
2215 def _htmlentity_transform(entity_with_semicolon
):
2216 """Transforms an HTML entity to a character."""
2217 entity
= entity_with_semicolon
[:-1]
2219 # Known non-numeric HTML entity
2220 if entity
in compat_html_entities
.name2codepoint
:
2221 return compat_chr(compat_html_entities
.name2codepoint
[entity
])
2223 # TODO: HTML5 allows entities without a semicolon. For example,
2224 # 'Éric' should be decoded as 'Éric'.
2225 if entity_with_semicolon
in compat_html_entities_html5
:
2226 return compat_html_entities_html5
[entity_with_semicolon
]
2228 mobj
= re
.match(r
'#(x[0-9a-fA-F]+|[0-9]+)', entity
)
2229 if mobj
is not None:
2230 numstr
= mobj
.group(1)
2231 if numstr
.startswith('x'):
2233 numstr
= '0%s' % numstr
2236 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2238 return compat_chr(int(numstr
, base
))
2242 # Unknown entity in name, return its literal representation
2243 return '&%s;' % entity
2246 def unescapeHTML(s
):
2249 assert type(s
) == compat_str
2252 r
'&([^&;]+;)', lambda m
: _htmlentity_transform(m
.group(1)), s
)
2255 def escapeHTML(text
):
2258 .replace('&', '&')
2259 .replace('<', '<')
2260 .replace('>', '>')
2261 .replace('"', '"')
2262 .replace("'", ''')
2266 def process_communicate_or_kill(p
, *args
, **kwargs
):
2268 return p
.communicate(*args
, **kwargs
)
2269 except BaseException
: # Including KeyboardInterrupt
2275 def get_subprocess_encoding():
2276 if sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2277 # For subprocess calls, encode with locale encoding
2278 # Refer to http://stackoverflow.com/a/9951851/35070
2279 encoding
= preferredencoding()
2281 encoding
= sys
.getfilesystemencoding()
2282 if encoding
is None:
2287 def encodeFilename(s
, for_subprocess
=False):
2289 @param s The name of the file
2292 assert type(s
) == compat_str
2294 # Python 3 has a Unicode API
2295 if sys
.version_info
>= (3, 0):
2298 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2299 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2300 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2301 if not for_subprocess
and sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2304 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2305 if sys
.platform
.startswith('java'):
2308 return s
.encode(get_subprocess_encoding(), 'ignore')
2311 def decodeFilename(b
, for_subprocess
=False):
2313 if sys
.version_info
>= (3, 0):
2316 if not isinstance(b
, bytes):
2319 return b
.decode(get_subprocess_encoding(), 'ignore')
2322 def encodeArgument(s
):
2323 if not isinstance(s
, compat_str
):
2324 # Legacy code that uses byte strings
2325 # Uncomment the following line after fixing all post processors
2326 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2327 s
= s
.decode('ascii')
2328 return encodeFilename(s
, True)
2331 def decodeArgument(b
):
2332 return decodeFilename(b
, True)
2335 def decodeOption(optval
):
2338 if isinstance(optval
, bytes):
2339 optval
= optval
.decode(preferredencoding())
2341 assert isinstance(optval
, compat_str
)
2345 def formatSeconds(secs
, delim
=':', msec
=False):
2347 ret
= '%d%s%02d%s%02d' % (secs
// 3600, delim
, (secs
% 3600) // 60, delim
, secs
% 60)
2349 ret
= '%d%s%02d' % (secs
// 60, delim
, secs
% 60)
2352 return '%s.%03d' % (ret
, secs
% 1) if msec
else ret
2355 def _ssl_load_windows_store_certs(ssl_context
, storename
):
2356 # Code adapted from _load_windows_store_certs in https://github.com/python/cpython/blob/main/Lib/ssl.py
2358 certs
= [cert
for cert
, encoding
, trust
in ssl
.enum_certificates(storename
)
2359 if encoding
== 'x509_asn' and (
2360 trust
is True or ssl
.Purpose
.SERVER_AUTH
.oid
in trust
)]
2361 except PermissionError
:
2365 ssl_context
.load_verify_locations(cadata
=cert
)
2366 except ssl
.SSLError
:
2370 def make_HTTPS_handler(params
, **kwargs
):
2371 opts_check_certificate
= not params
.get('nocheckcertificate')
2372 context
= ssl
.SSLContext(ssl
.PROTOCOL_TLS_CLIENT
)
2373 context
.check_hostname
= opts_check_certificate
2374 context
.verify_mode
= ssl
.CERT_REQUIRED
if opts_check_certificate
else ssl
.CERT_NONE
2375 if opts_check_certificate
:
2377 context
.load_default_certs()
2378 # Work around the issue in load_default_certs when there are bad certificates. See:
2379 # https://github.com/yt-dlp/yt-dlp/issues/1060,
2380 # https://bugs.python.org/issue35665, https://bugs.python.org/issue45312
2381 except ssl
.SSLError
:
2382 # enum_certificates is not present in mingw python. See https://github.com/yt-dlp/yt-dlp/issues/1151
2383 if sys
.platform
== 'win32' and hasattr(ssl
, 'enum_certificates'):
2384 # Create a new context to discard any certificates that were already loaded
2385 context
= ssl
.SSLContext(ssl
.PROTOCOL_TLS_CLIENT
)
2386 context
.check_hostname
, context
.verify_mode
= True, ssl
.CERT_REQUIRED
2387 for storename
in ('CA', 'ROOT'):
2388 _ssl_load_windows_store_certs(context
, storename
)
2389 context
.set_default_verify_paths()
2390 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2393 def bug_reports_message(before
=';'):
2394 if ytdl_is_updateable():
2395 update_cmd
= 'type yt-dlp -U to update'
2397 update_cmd
= 'see https://github.com/yt-dlp/yt-dlp on how to update'
2398 msg
= 'please report this issue on https://github.com/yt-dlp/yt-dlp .'
2399 msg
+= ' Make sure you are using the latest version; %s.' % update_cmd
2400 msg
+= ' Be sure to call yt-dlp with the --verbose flag and include its complete output.'
2402 before
= before
.rstrip()
2403 if not before
or before
.endswith(('.', '!', '?')):
2404 msg
= msg
[0].title() + msg
[1:]
2406 return (before
+ ' ' if before
else '') + msg
2409 class YoutubeDLError(Exception):
2410 """Base exception for YoutubeDL errors."""
2414 network_exceptions
= [compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
]
2415 if hasattr(ssl
, 'CertificateError'):
2416 network_exceptions
.append(ssl
.CertificateError
)
2417 network_exceptions
= tuple(network_exceptions
)
2420 class ExtractorError(YoutubeDLError
):
2421 """Error during info extraction."""
2423 def __init__(self
, msg
, tb
=None, expected
=False, cause
=None, video_id
=None, ie
=None):
2424 """ tb, if given, is the original traceback (so that it can be printed out).
2425 If expected is set, this is a normal error message and most likely not a bug in yt-dlp.
2427 if sys
.exc_info()[0] in network_exceptions
:
2432 self
.expected
= expected
2434 self
.video_id
= video_id
2436 self
.exc_info
= sys
.exc_info() # preserve original exception
2438 super(ExtractorError
, self
).__init
__(''.join((
2439 format_field(ie
, template
='[%s] '),
2440 format_field(video_id
, template
='%s: '),
2442 format_field(cause
, template
=' (caused by %r)'),
2443 '' if expected
else bug_reports_message())))
2445 def format_traceback(self
):
2446 if self
.traceback
is None:
2448 return ''.join(traceback
.format_tb(self
.traceback
))
2451 class UnsupportedError(ExtractorError
):
2452 def __init__(self
, url
):
2453 super(UnsupportedError
, self
).__init
__(
2454 'Unsupported URL: %s' % url
, expected
=True)
2458 class RegexNotFoundError(ExtractorError
):
2459 """Error when a regex didn't match"""
2463 class GeoRestrictedError(ExtractorError
):
2464 """Geographic restriction Error exception.
2466 This exception may be thrown when a video is not available from your
2467 geographic location due to geographic restrictions imposed by a website.
2470 def __init__(self
, msg
, countries
=None):
2471 super(GeoRestrictedError
, self
).__init
__(msg
, expected
=True)
2473 self
.countries
= countries
2476 class DownloadError(YoutubeDLError
):
2477 """Download Error exception.
2479 This exception may be thrown by FileDownloader objects if they are not
2480 configured to continue on errors. They will contain the appropriate
2484 def __init__(self
, msg
, exc_info
=None):
2485 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2486 super(DownloadError
, self
).__init
__(msg
)
2487 self
.exc_info
= exc_info
2490 class EntryNotInPlaylist(YoutubeDLError
):
2491 """Entry not in playlist exception.
2493 This exception will be thrown by YoutubeDL when a requested entry
2494 is not found in the playlist info_dict
2499 class SameFileError(YoutubeDLError
):
2500 """Same File exception.
2502 This exception will be thrown by FileDownloader objects if they detect
2503 multiple files would have to be downloaded to the same file on disk.
2508 class PostProcessingError(YoutubeDLError
):
2509 """Post Processing exception.
2511 This exception may be raised by PostProcessor's .run() method to
2512 indicate an error in the postprocessing task.
2515 def __init__(self
, msg
):
2516 super(PostProcessingError
, self
).__init
__(msg
)
2520 class ExistingVideoReached(YoutubeDLError
):
2521 """ --max-downloads limit has been reached. """
2525 class RejectedVideoReached(YoutubeDLError
):
2526 """ --max-downloads limit has been reached. """
2530 class ThrottledDownload(YoutubeDLError
):
2531 """ Download speed below --throttled-rate. """
2535 class MaxDownloadsReached(YoutubeDLError
):
2536 """ --max-downloads limit has been reached. """
2540 class UnavailableVideoError(YoutubeDLError
):
2541 """Unavailable Format exception.
2543 This exception will be thrown when a video is requested
2544 in a format that is not available for that video.
2549 class ContentTooShortError(YoutubeDLError
):
2550 """Content Too Short exception.
2552 This exception may be raised by FileDownloader objects when a file they
2553 download is too small for what the server announced first, indicating
2554 the connection was probably interrupted.
2557 def __init__(self
, downloaded
, expected
):
2558 super(ContentTooShortError
, self
).__init
__(
2559 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded
, expected
)
2562 self
.downloaded
= downloaded
2563 self
.expected
= expected
2566 class XAttrMetadataError(YoutubeDLError
):
2567 def __init__(self
, code
=None, msg
='Unknown error'):
2568 super(XAttrMetadataError
, self
).__init
__(msg
)
2572 # Parsing code and msg
2573 if (self
.code
in (errno
.ENOSPC
, errno
.EDQUOT
)
2574 or 'No space left' in self
.msg
or 'Disk quota exceeded' in self
.msg
):
2575 self
.reason
= 'NO_SPACE'
2576 elif self
.code
== errno
.E2BIG
or 'Argument list too long' in self
.msg
:
2577 self
.reason
= 'VALUE_TOO_LONG'
2579 self
.reason
= 'NOT_SUPPORTED'
2582 class XAttrUnavailableError(YoutubeDLError
):
2586 def _create_http_connection(ydl_handler
, http_class
, is_https
, *args
, **kwargs
):
2587 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2588 # expected HTTP responses to meet HTTP/1.0 or later (see also
2589 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2590 if sys
.version_info
< (3, 0):
2591 kwargs
['strict'] = True
2592 hc
= http_class(*args
, **compat_kwargs(kwargs
))
2593 source_address
= ydl_handler
._params
.get('source_address')
2595 if source_address
is not None:
2596 # This is to workaround _create_connection() from socket where it will try all
2597 # address data from getaddrinfo() including IPv6. This filters the result from
2598 # getaddrinfo() based on the source_address value.
2599 # This is based on the cpython socket.create_connection() function.
2600 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2601 def _create_connection(address
, timeout
=socket
._GLOBAL
_DEFAULT
_TIMEOUT
, source_address
=None):
2602 host
, port
= address
2604 addrs
= socket
.getaddrinfo(host
, port
, 0, socket
.SOCK_STREAM
)
2605 af
= socket
.AF_INET
if '.' in source_address
[0] else socket
.AF_INET6
2606 ip_addrs
= [addr
for addr
in addrs
if addr
[0] == af
]
2607 if addrs
and not ip_addrs
:
2608 ip_version
= 'v4' if af
== socket
.AF_INET
else 'v6'
2610 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2611 % (ip_version
, source_address
[0]))
2612 for res
in ip_addrs
:
2613 af
, socktype
, proto
, canonname
, sa
= res
2616 sock
= socket
.socket(af
, socktype
, proto
)
2617 if timeout
is not socket
._GLOBAL
_DEFAULT
_TIMEOUT
:
2618 sock
.settimeout(timeout
)
2619 sock
.bind(source_address
)
2621 err
= None # Explicitly break reference cycle
2623 except socket
.error
as _
:
2625 if sock
is not None:
2630 raise socket
.error('getaddrinfo returns an empty list')
2631 if hasattr(hc
, '_create_connection'):
2632 hc
._create
_connection
= _create_connection
2633 sa
= (source_address
, 0)
2634 if hasattr(hc
, 'source_address'): # Python 2.7+
2635 hc
.source_address
= sa
2637 def _hc_connect(self
, *args
, **kwargs
):
2638 sock
= _create_connection(
2639 (self
.host
, self
.port
), self
.timeout
, sa
)
2641 self
.sock
= ssl
.wrap_socket(
2642 sock
, self
.key_file
, self
.cert_file
,
2643 ssl_version
=ssl
.PROTOCOL_TLSv1
)
2646 hc
.connect
= functools
.partial(_hc_connect
, hc
)
2651 def handle_youtubedl_headers(headers
):
2652 filtered_headers
= headers
2654 if 'Youtubedl-no-compression' in filtered_headers
:
2655 filtered_headers
= dict((k
, v
) for k
, v
in filtered_headers
.items() if k
.lower() != 'accept-encoding')
2656 del filtered_headers
['Youtubedl-no-compression']
2658 return filtered_headers
2661 class YoutubeDLHandler(compat_urllib_request
.HTTPHandler
):
2662 """Handler for HTTP requests and responses.
2664 This class, when installed with an OpenerDirector, automatically adds
2665 the standard headers to every HTTP request and handles gzipped and
2666 deflated responses from web servers. If compression is to be avoided in
2667 a particular request, the original request in the program code only has
2668 to include the HTTP header "Youtubedl-no-compression", which will be
2669 removed before making the real request.
2671 Part of this code was copied from:
2673 http://techknack.net/python-urllib2-handlers/
2675 Andrew Rowls, the author of that code, agreed to release it to the
2679 def __init__(self
, params
, *args
, **kwargs
):
2680 compat_urllib_request
.HTTPHandler
.__init
__(self
, *args
, **kwargs
)
2681 self
._params
= params
2683 def http_open(self
, req
):
2684 conn_class
= compat_http_client
.HTTPConnection
2686 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2688 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2689 del req
.headers
['Ytdl-socks-proxy']
2691 return self
.do_open(functools
.partial(
2692 _create_http_connection
, self
, conn_class
, False),
2700 return zlib
.decompress(data
, -zlib
.MAX_WBITS
)
2702 return zlib
.decompress(data
)
2704 def http_request(self
, req
):
2705 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2706 # always respected by websites, some tend to give out URLs with non percent-encoded
2707 # non-ASCII characters (see telemb.py, ard.py [#3412])
2708 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2709 # To work around aforementioned issue we will replace request's original URL with
2710 # percent-encoded one
2711 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2712 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2713 url
= req
.get_full_url()
2714 url_escaped
= escape_url(url
)
2716 # Substitute URL if any change after escaping
2717 if url
!= url_escaped
:
2718 req
= update_Request(req
, url
=url_escaped
)
2720 for h
, v
in std_headers
.items():
2721 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2722 # The dict keys are capitalized because of this bug by urllib
2723 if h
.capitalize() not in req
.headers
:
2724 req
.add_header(h
, v
)
2726 req
.headers
= handle_youtubedl_headers(req
.headers
)
2728 if sys
.version_info
< (2, 7) and '#' in req
.get_full_url():
2729 # Python 2.6 is brain-dead when it comes to fragments
2730 req
._Request
__original
= req
._Request
__original
.partition('#')[0]
2731 req
._Request
__r
_type
= req
._Request
__r
_type
.partition('#')[0]
2735 def http_response(self
, req
, resp
):
2738 if resp
.headers
.get('Content-encoding', '') == 'gzip':
2739 content
= resp
.read()
2740 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
), mode
='rb')
2742 uncompressed
= io
.BytesIO(gz
.read())
2743 except IOError as original_ioerror
:
2744 # There may be junk add the end of the file
2745 # See http://stackoverflow.com/q/4928560/35070 for details
2746 for i
in range(1, 1024):
2748 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
[:-i
]), mode
='rb')
2749 uncompressed
= io
.BytesIO(gz
.read())
2754 raise original_ioerror
2755 resp
= compat_urllib_request
.addinfourl(uncompressed
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2756 resp
.msg
= old_resp
.msg
2757 del resp
.headers
['Content-encoding']
2759 if resp
.headers
.get('Content-encoding', '') == 'deflate':
2760 gz
= io
.BytesIO(self
.deflate(resp
.read()))
2761 resp
= compat_urllib_request
.addinfourl(gz
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2762 resp
.msg
= old_resp
.msg
2763 del resp
.headers
['Content-encoding']
2764 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2765 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2766 if 300 <= resp
.code
< 400:
2767 location
= resp
.headers
.get('Location')
2769 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2770 if sys
.version_info
>= (3, 0):
2771 location
= location
.encode('iso-8859-1').decode('utf-8')
2773 location
= location
.decode('utf-8')
2774 location_escaped
= escape_url(location
)
2775 if location
!= location_escaped
:
2776 del resp
.headers
['Location']
2777 if sys
.version_info
< (3, 0):
2778 location_escaped
= location_escaped
.encode('utf-8')
2779 resp
.headers
['Location'] = location_escaped
2782 https_request
= http_request
2783 https_response
= http_response
2786 def make_socks_conn_class(base_class
, socks_proxy
):
2787 assert issubclass(base_class
, (
2788 compat_http_client
.HTTPConnection
, compat_http_client
.HTTPSConnection
))
2790 url_components
= compat_urlparse
.urlparse(socks_proxy
)
2791 if url_components
.scheme
.lower() == 'socks5':
2792 socks_type
= ProxyType
.SOCKS5
2793 elif url_components
.scheme
.lower() in ('socks', 'socks4'):
2794 socks_type
= ProxyType
.SOCKS4
2795 elif url_components
.scheme
.lower() == 'socks4a':
2796 socks_type
= ProxyType
.SOCKS4A
2798 def unquote_if_non_empty(s
):
2801 return compat_urllib_parse_unquote_plus(s
)
2805 url_components
.hostname
, url_components
.port
or 1080,
2807 unquote_if_non_empty(url_components
.username
),
2808 unquote_if_non_empty(url_components
.password
),
2811 class SocksConnection(base_class
):
2813 self
.sock
= sockssocket()
2814 self
.sock
.setproxy(*proxy_args
)
2815 if type(self
.timeout
) in (int, float):
2816 self
.sock
.settimeout(self
.timeout
)
2817 self
.sock
.connect((self
.host
, self
.port
))
2819 if isinstance(self
, compat_http_client
.HTTPSConnection
):
2820 if hasattr(self
, '_context'): # Python > 2.6
2821 self
.sock
= self
._context
.wrap_socket(
2822 self
.sock
, server_hostname
=self
.host
)
2824 self
.sock
= ssl
.wrap_socket(self
.sock
)
2826 return SocksConnection
2829 class YoutubeDLHTTPSHandler(compat_urllib_request
.HTTPSHandler
):
2830 def __init__(self
, params
, https_conn_class
=None, *args
, **kwargs
):
2831 compat_urllib_request
.HTTPSHandler
.__init
__(self
, *args
, **kwargs
)
2832 self
._https
_conn
_class
= https_conn_class
or compat_http_client
.HTTPSConnection
2833 self
._params
= params
2835 def https_open(self
, req
):
2837 conn_class
= self
._https
_conn
_class
2839 if hasattr(self
, '_context'): # python > 2.6
2840 kwargs
['context'] = self
._context
2841 if hasattr(self
, '_check_hostname'): # python 3.x
2842 kwargs
['check_hostname'] = self
._check
_hostname
2844 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2846 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2847 del req
.headers
['Ytdl-socks-proxy']
2849 return self
.do_open(functools
.partial(
2850 _create_http_connection
, self
, conn_class
, True),
2854 class YoutubeDLCookieJar(compat_cookiejar
.MozillaCookieJar
):
2856 See [1] for cookie file format.
2858 1. https://curl.haxx.se/docs/http-cookies.html
2860 _HTTPONLY_PREFIX
= '#HttpOnly_'
2862 _HEADER
= '''# Netscape HTTP Cookie File
2863 # This file is generated by yt-dlp. Do not edit.
2866 _CookieFileEntry
= collections
.namedtuple(
2868 ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
2870 def save(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2872 Save cookies to a file.
2874 Most of the code is taken from CPython 3.8 and slightly adapted
2875 to support cookie files with UTF-8 in both python 2 and 3.
2877 if filename
is None:
2878 if self
.filename
is not None:
2879 filename
= self
.filename
2881 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2883 # Store session cookies with `expires` set to 0 instead of an empty
2886 if cookie
.expires
is None:
2889 with io
.open(filename
, 'w', encoding
='utf-8') as f
:
2890 f
.write(self
._HEADER
)
2893 if not ignore_discard
and cookie
.discard
:
2895 if not ignore_expires
and cookie
.is_expired(now
):
2901 if cookie
.domain
.startswith('.'):
2902 initial_dot
= 'TRUE'
2904 initial_dot
= 'FALSE'
2905 if cookie
.expires
is not None:
2906 expires
= compat_str(cookie
.expires
)
2909 if cookie
.value
is None:
2910 # cookies.txt regards 'Set-Cookie: foo' as a cookie
2911 # with no name, whereas http.cookiejar regards it as a
2912 # cookie with no value.
2917 value
= cookie
.value
2919 '\t'.join([cookie
.domain
, initial_dot
, cookie
.path
,
2920 secure
, expires
, name
, value
]) + '\n')
2922 def load(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2923 """Load cookies from a file."""
2924 if filename
is None:
2925 if self
.filename
is not None:
2926 filename
= self
.filename
2928 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2930 def prepare_line(line
):
2931 if line
.startswith(self
._HTTPONLY
_PREFIX
):
2932 line
= line
[len(self
._HTTPONLY
_PREFIX
):]
2933 # comments and empty lines are fine
2934 if line
.startswith('#') or not line
.strip():
2936 cookie_list
= line
.split('\t')
2937 if len(cookie_list
) != self
._ENTRY
_LEN
:
2938 raise compat_cookiejar
.LoadError('invalid length %d' % len(cookie_list
))
2939 cookie
= self
._CookieFileEntry
(*cookie_list
)
2940 if cookie
.expires_at
and not cookie
.expires_at
.isdigit():
2941 raise compat_cookiejar
.LoadError('invalid expires at %s' % cookie
.expires_at
)
2945 with io
.open(filename
, encoding
='utf-8') as f
:
2948 cf
.write(prepare_line(line
))
2949 except compat_cookiejar
.LoadError
as e
:
2951 'WARNING: skipping cookie file entry due to %s: %r\n'
2952 % (e
, line
), sys
.stderr
)
2955 self
._really
_load
(cf
, filename
, ignore_discard
, ignore_expires
)
2956 # Session cookies are denoted by either `expires` field set to
2957 # an empty string or 0. MozillaCookieJar only recognizes the former
2958 # (see [1]). So we need force the latter to be recognized as session
2959 # cookies on our own.
2960 # Session cookies may be important for cookies-based authentication,
2961 # e.g. usually, when user does not check 'Remember me' check box while
2962 # logging in on a site, some important cookies are stored as session
2963 # cookies so that not recognizing them will result in failed login.
2964 # 1. https://bugs.python.org/issue17164
2966 # Treat `expires=0` cookies as session cookies
2967 if cookie
.expires
== 0:
2968 cookie
.expires
= None
2969 cookie
.discard
= True
2972 class YoutubeDLCookieProcessor(compat_urllib_request
.HTTPCookieProcessor
):
2973 def __init__(self
, cookiejar
=None):
2974 compat_urllib_request
.HTTPCookieProcessor
.__init
__(self
, cookiejar
)
2976 def http_response(self
, request
, response
):
2977 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2978 # characters in Set-Cookie HTTP header of last response (see
2979 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2980 # In order to at least prevent crashing we will percent encode Set-Cookie
2981 # header before HTTPCookieProcessor starts processing it.
2982 # if sys.version_info < (3, 0) and response.headers:
2983 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2984 # set_cookie = response.headers.get(set_cookie_header)
2986 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2987 # if set_cookie != set_cookie_escaped:
2988 # del response.headers[set_cookie_header]
2989 # response.headers[set_cookie_header] = set_cookie_escaped
2990 return compat_urllib_request
.HTTPCookieProcessor
.http_response(self
, request
, response
)
2992 https_request
= compat_urllib_request
.HTTPCookieProcessor
.http_request
2993 https_response
= http_response
2996 class YoutubeDLRedirectHandler(compat_urllib_request
.HTTPRedirectHandler
):
2997 """YoutubeDL redirect handler
2999 The code is based on HTTPRedirectHandler implementation from CPython [1].
3001 This redirect handler solves two issues:
3002 - ensures redirect URL is always unicode under python 2
3003 - introduces support for experimental HTTP response status code
3004 308 Permanent Redirect [2] used by some sites [3]
3006 1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py
3007 2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308
3008 3. https://github.com/ytdl-org/youtube-dl/issues/28768
3011 http_error_301
= http_error_303
= http_error_307
= http_error_308
= compat_urllib_request
.HTTPRedirectHandler
.http_error_302
3013 def redirect_request(self
, req
, fp
, code
, msg
, headers
, newurl
):
3014 """Return a Request or None in response to a redirect.
3016 This is called by the http_error_30x methods when a
3017 redirection response is received. If a redirection should
3018 take place, return a new Request to allow http_error_30x to
3019 perform the redirect. Otherwise, raise HTTPError if no-one
3020 else should try to handle this url. Return None if you can't
3021 but another Handler might.
3023 m
= req
.get_method()
3024 if (not (code
in (301, 302, 303, 307, 308) and m
in ("GET", "HEAD")
3025 or code
in (301, 302, 303) and m
== "POST")):
3026 raise compat_HTTPError(req
.full_url
, code
, msg
, headers
, fp
)
3027 # Strictly (according to RFC 2616), 301 or 302 in response to
3028 # a POST MUST NOT cause a redirection without confirmation
3029 # from the user (of urllib.request, in this case). In practice,
3030 # essentially all clients do redirect in this case, so we do
3033 # On python 2 urlh.geturl() may sometimes return redirect URL
3034 # as byte string instead of unicode. This workaround allows
3035 # to force it always return unicode.
3036 if sys
.version_info
[0] < 3:
3037 newurl
= compat_str(newurl
)
3039 # Be conciliant with URIs containing a space. This is mainly
3040 # redundant with the more complete encoding done in http_error_302(),
3041 # but it is kept for compatibility with other callers.
3042 newurl
= newurl
.replace(' ', '%20')
3044 CONTENT_HEADERS
= ("content-length", "content-type")
3045 # NB: don't use dict comprehension for python 2.6 compatibility
3046 newheaders
= dict((k
, v
) for k
, v
in req
.headers
.items()
3047 if k
.lower() not in CONTENT_HEADERS
)
3048 return compat_urllib_request
.Request(
3049 newurl
, headers
=newheaders
, origin_req_host
=req
.origin_req_host
,
3053 def extract_timezone(date_str
):
3056 ^.{8,}? # >=8 char non-TZ prefix, if present
3057 (?P<tz>Z| # just the UTC Z, or
3058 (?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or
3059 (?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d)) # not preceded by 3 alpha word or >= 4 alpha or 2 digits
3060 [ ]? # optional space
3061 (?P<sign>\+|-) # +/-
3062 (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2}) # hh[:]mm
3066 timezone
= datetime
.timedelta()
3068 date_str
= date_str
[:-len(m
.group('tz'))]
3069 if not m
.group('sign'):
3070 timezone
= datetime
.timedelta()
3072 sign
= 1 if m
.group('sign') == '+' else -1
3073 timezone
= datetime
.timedelta(
3074 hours
=sign
* int(m
.group('hours')),
3075 minutes
=sign
* int(m
.group('minutes')))
3076 return timezone
, date_str
3079 def parse_iso8601(date_str
, delimiter
='T', timezone
=None):
3080 """ Return a UNIX timestamp from the given date """
3082 if date_str
is None:
3085 date_str
= re
.sub(r
'\.[0-9]+', '', date_str
)
3087 if timezone
is None:
3088 timezone
, date_str
= extract_timezone(date_str
)
3091 date_format
= '%Y-%m-%d{0}%H:%M:%S'.format(delimiter
)
3092 dt
= datetime
.datetime
.strptime(date_str
, date_format
) - timezone
3093 return calendar
.timegm(dt
.timetuple())
3098 def date_formats(day_first
=True):
3099 return DATE_FORMATS_DAY_FIRST
if day_first
else DATE_FORMATS_MONTH_FIRST
3102 def unified_strdate(date_str
, day_first
=True):
3103 """Return a string with the date in the format YYYYMMDD"""
3105 if date_str
is None:
3109 date_str
= date_str
.replace(',', ' ')
3110 # Remove AM/PM + timezone
3111 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
3112 _
, date_str
= extract_timezone(date_str
)
3114 for expression
in date_formats(day_first
):
3116 upload_date
= datetime
.datetime
.strptime(date_str
, expression
).strftime('%Y%m%d')
3119 if upload_date
is None:
3120 timetuple
= email
.utils
.parsedate_tz(date_str
)
3123 upload_date
= datetime
.datetime(*timetuple
[:6]).strftime('%Y%m%d')
3126 if upload_date
is not None:
3127 return compat_str(upload_date
)
3130 def unified_timestamp(date_str
, day_first
=True):
3131 if date_str
is None:
3134 date_str
= re
.sub(r
'[,|]', '', date_str
)
3136 pm_delta
= 12 if re
.search(r
'(?i)PM', date_str
) else 0
3137 timezone
, date_str
= extract_timezone(date_str
)
3139 # Remove AM/PM + timezone
3140 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
3142 # Remove unrecognized timezones from ISO 8601 alike timestamps
3143 m
= re
.search(r
'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str
)
3145 date_str
= date_str
[:-len(m
.group('tz'))]
3147 # Python only supports microseconds, so remove nanoseconds
3148 m
= re
.search(r
'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str
)
3150 date_str
= m
.group(1)
3152 for expression
in date_formats(day_first
):
3154 dt
= datetime
.datetime
.strptime(date_str
, expression
) - timezone
+ datetime
.timedelta(hours
=pm_delta
)
3155 return calendar
.timegm(dt
.timetuple())
3158 timetuple
= email
.utils
.parsedate_tz(date_str
)
3160 return calendar
.timegm(timetuple
) + pm_delta
* 3600
3163 def determine_ext(url
, default_ext
='unknown_video'):
3164 if url
is None or '.' not in url
:
3166 guess
= url
.partition('?')[0].rpartition('.')[2]
3167 if re
.match(r
'^[A-Za-z0-9]+$', guess
):
3169 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
3170 elif guess
.rstrip('/') in KNOWN_EXTENSIONS
:
3171 return guess
.rstrip('/')
3176 def subtitles_filename(filename
, sub_lang
, sub_format
, expected_real_ext
=None):
3177 return replace_extension(filename
, sub_lang
+ '.' + sub_format
, expected_real_ext
)
3180 def datetime_from_str(date_str
, precision
='auto', format
='%Y%m%d'):
3182 Return a datetime object from a string in the format YYYYMMDD or
3183 (now|today|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
3185 format: string date format used to return datetime object from
3186 precision: round the time portion of a datetime object.
3187 auto|microsecond|second|minute|hour|day.
3188 auto: round to the unit provided in date_str (if applicable).
3190 auto_precision
= False
3191 if precision
== 'auto':
3192 auto_precision
= True
3193 precision
= 'microsecond'
3194 today
= datetime_round(datetime
.datetime
.now(), precision
)
3195 if date_str
in ('now', 'today'):
3197 if date_str
== 'yesterday':
3198 return today
- datetime
.timedelta(days
=1)
3200 r
'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)(s)?',
3202 if match
is not None:
3203 start_time
= datetime_from_str(match
.group('start'), precision
, format
)
3204 time
= int(match
.group('time')) * (-1 if match
.group('sign') == '-' else 1)
3205 unit
= match
.group('unit')
3206 if unit
== 'month' or unit
== 'year':
3207 new_date
= datetime_add_months(start_time
, time
* 12 if unit
== 'year' else time
)
3213 delta
= datetime
.timedelta(**{unit + 's': time}
)
3214 new_date
= start_time
+ delta
3216 return datetime_round(new_date
, unit
)
3219 return datetime_round(datetime
.datetime
.strptime(date_str
, format
), precision
)
3222 def date_from_str(date_str
, format
='%Y%m%d'):
3224 Return a datetime object from a string in the format YYYYMMDD or
3225 (now|today|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
3227 format: string date format used to return datetime object from
3229 return datetime_from_str(date_str
, precision
='microsecond', format
=format
).date()
3232 def datetime_add_months(dt
, months
):
3233 """Increment/Decrement a datetime object by months."""
3234 month
= dt
.month
+ months
- 1
3235 year
= dt
.year
+ month
// 12
3236 month
= month
% 12 + 1
3237 day
= min(dt
.day
, calendar
.monthrange(year
, month
)[1])
3238 return dt
.replace(year
, month
, day
)
3241 def datetime_round(dt
, precision
='day'):
3243 Round a datetime object's time to a specific precision
3245 if precision
== 'microsecond':
3254 roundto
= lambda x
, n
: ((x
+ n
/ 2) // n
) * n
3255 timestamp
= calendar
.timegm(dt
.timetuple())
3256 return datetime
.datetime
.utcfromtimestamp(roundto(timestamp
, unit_seconds
[precision
]))
3259 def hyphenate_date(date_str
):
3261 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
3262 match
= re
.match(r
'^(\d\d\d\d)(\d\d)(\d\d)$', date_str
)
3263 if match
is not None:
3264 return '-'.join(match
.groups())
3269 class DateRange(object):
3270 """Represents a time interval between two dates"""
3272 def __init__(self
, start
=None, end
=None):
3273 """start and end must be strings in the format accepted by date"""
3274 if start
is not None:
3275 self
.start
= date_from_str(start
)
3277 self
.start
= datetime
.datetime
.min.date()
3279 self
.end
= date_from_str(end
)
3281 self
.end
= datetime
.datetime
.max.date()
3282 if self
.start
> self
.end
:
3283 raise ValueError('Date range: "%s" , the start date must be before the end date' % self
)
3287 """Returns a range that only contains the given day"""
3288 return cls(day
, day
)
3290 def __contains__(self
, date
):
3291 """Check if the date is in the range"""
3292 if not isinstance(date
, datetime
.date
):
3293 date
= date_from_str(date
)
3294 return self
.start
<= date
<= self
.end
3297 return '%s - %s' % (self
.start
.isoformat(), self
.end
.isoformat())
3300 def platform_name():
3301 """ Returns the platform name as a compat_str """
3302 res
= platform
.platform()
3303 if isinstance(res
, bytes):
3304 res
= res
.decode(preferredencoding())
3306 assert isinstance(res
, compat_str
)
3310 def get_windows_version():
3311 ''' Get Windows version. None if it's not running on Windows '''
3312 if compat_os_name
== 'nt':
3313 return version_tuple(platform
.win32_ver()[1])
3318 def _windows_write_string(s
, out
):
3319 """ Returns True if the string was written using special methods,
3320 False if it has yet to be written out."""
3321 # Adapted from http://stackoverflow.com/a/3259271/35070
3324 import ctypes
.wintypes
3332 fileno
= out
.fileno()
3333 except AttributeError:
3334 # If the output stream doesn't have a fileno, it's virtual
3336 except io
.UnsupportedOperation
:
3337 # Some strange Windows pseudo files?
3339 if fileno
not in WIN_OUTPUT_IDS
:
3342 GetStdHandle
= compat_ctypes_WINFUNCTYPE(
3343 ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.DWORD
)(
3344 ('GetStdHandle', ctypes
.windll
.kernel32
))
3345 h
= GetStdHandle(WIN_OUTPUT_IDS
[fileno
])
3347 WriteConsoleW
= compat_ctypes_WINFUNCTYPE(
3348 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.LPWSTR
,
3349 ctypes
.wintypes
.DWORD
, ctypes
.POINTER(ctypes
.wintypes
.DWORD
),
3350 ctypes
.wintypes
.LPVOID
)(('WriteConsoleW', ctypes
.windll
.kernel32
))
3351 written
= ctypes
.wintypes
.DWORD(0)
3353 GetFileType
= compat_ctypes_WINFUNCTYPE(ctypes
.wintypes
.DWORD
, ctypes
.wintypes
.DWORD
)(('GetFileType', ctypes
.windll
.kernel32
))
3354 FILE_TYPE_CHAR
= 0x0002
3355 FILE_TYPE_REMOTE
= 0x8000
3356 GetConsoleMode
= compat_ctypes_WINFUNCTYPE(
3357 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
,
3358 ctypes
.POINTER(ctypes
.wintypes
.DWORD
))(
3359 ('GetConsoleMode', ctypes
.windll
.kernel32
))
3360 INVALID_HANDLE_VALUE
= ctypes
.wintypes
.DWORD(-1).value
3362 def not_a_console(handle
):
3363 if handle
== INVALID_HANDLE_VALUE
or handle
is None:
3365 return ((GetFileType(handle
) & ~FILE_TYPE_REMOTE
) != FILE_TYPE_CHAR
3366 or GetConsoleMode(handle
, ctypes
.byref(ctypes
.wintypes
.DWORD())) == 0)
3368 if not_a_console(h
):
3371 def next_nonbmp_pos(s
):
3373 return next(i
for i
, c
in enumerate(s
) if ord(c
) > 0xffff)
3374 except StopIteration:
3378 count
= min(next_nonbmp_pos(s
), 1024)
3380 ret
= WriteConsoleW(
3381 h
, s
, count
if count
else 2, ctypes
.byref(written
), None)
3383 raise OSError('Failed to write string')
3384 if not count
: # We just wrote a non-BMP character
3385 assert written
.value
== 2
3388 assert written
.value
> 0
3389 s
= s
[written
.value
:]
3393 def write_string(s
, out
=None, encoding
=None):
3396 assert type(s
) == compat_str
3398 if sys
.platform
== 'win32' and encoding
is None and hasattr(out
, 'fileno'):
3399 if _windows_write_string(s
, out
):
3402 if ('b' in getattr(out
, 'mode', '')
3403 or sys
.version_info
[0] < 3): # Python 2 lies about mode of sys.stderr
3404 byt
= s
.encode(encoding
or preferredencoding(), 'ignore')
3406 elif hasattr(out
, 'buffer'):
3407 enc
= encoding
or getattr(out
, 'encoding', None) or preferredencoding()
3408 byt
= s
.encode(enc
, 'ignore')
3409 out
.buffer.write(byt
)
3415 def bytes_to_intlist(bs
):
3418 if isinstance(bs
[0], int): # Python 3
3421 return [ord(c
) for c
in bs
]
3424 def intlist_to_bytes(xs
):
3427 return compat_struct_pack('%dB' % len(xs
), *xs
)
3430 # Cross-platform file locking
3431 if sys
.platform
== 'win32':
3432 import ctypes
.wintypes
3435 class OVERLAPPED(ctypes
.Structure
):
3437 ('Internal', ctypes
.wintypes
.LPVOID
),
3438 ('InternalHigh', ctypes
.wintypes
.LPVOID
),
3439 ('Offset', ctypes
.wintypes
.DWORD
),
3440 ('OffsetHigh', ctypes
.wintypes
.DWORD
),
3441 ('hEvent', ctypes
.wintypes
.HANDLE
),
3444 kernel32
= ctypes
.windll
.kernel32
3445 LockFileEx
= kernel32
.LockFileEx
3446 LockFileEx
.argtypes
= [
3447 ctypes
.wintypes
.HANDLE
, # hFile
3448 ctypes
.wintypes
.DWORD
, # dwFlags
3449 ctypes
.wintypes
.DWORD
, # dwReserved
3450 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3451 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3452 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3454 LockFileEx
.restype
= ctypes
.wintypes
.BOOL
3455 UnlockFileEx
= kernel32
.UnlockFileEx
3456 UnlockFileEx
.argtypes
= [
3457 ctypes
.wintypes
.HANDLE
, # hFile
3458 ctypes
.wintypes
.DWORD
, # dwReserved
3459 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3460 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3461 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3463 UnlockFileEx
.restype
= ctypes
.wintypes
.BOOL
3464 whole_low
= 0xffffffff
3465 whole_high
= 0x7fffffff
3467 def _lock_file(f
, exclusive
):
3468 overlapped
= OVERLAPPED()
3469 overlapped
.Offset
= 0
3470 overlapped
.OffsetHigh
= 0
3471 overlapped
.hEvent
= 0
3472 f
._lock
_file
_overlapped
_p
= ctypes
.pointer(overlapped
)
3473 handle
= msvcrt
.get_osfhandle(f
.fileno())
3474 if not LockFileEx(handle
, 0x2 if exclusive
else 0x0, 0,
3475 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3476 raise OSError('Locking file failed: %r' % ctypes
.FormatError())
3478 def _unlock_file(f
):
3479 assert f
._lock
_file
_overlapped
_p
3480 handle
= msvcrt
.get_osfhandle(f
.fileno())
3481 if not UnlockFileEx(handle
, 0,
3482 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3483 raise OSError('Unlocking file failed: %r' % ctypes
.FormatError())
3486 # Some platforms, such as Jython, is missing fcntl
3490 def _lock_file(f
, exclusive
):
3491 fcntl
.flock(f
, fcntl
.LOCK_EX
if exclusive
else fcntl
.LOCK_SH
)
3493 def _unlock_file(f
):
3494 fcntl
.flock(f
, fcntl
.LOCK_UN
)
3496 UNSUPPORTED_MSG
= 'file locking is not supported on this platform'
3498 def _lock_file(f
, exclusive
):
3499 raise IOError(UNSUPPORTED_MSG
)
3501 def _unlock_file(f
):
3502 raise IOError(UNSUPPORTED_MSG
)
3505 class locked_file(object):
3506 def __init__(self
, filename
, mode
, encoding
=None):
3507 assert mode
in ['r', 'a', 'w']
3508 self
.f
= io
.open(filename
, mode
, encoding
=encoding
)
3511 def __enter__(self
):
3512 exclusive
= self
.mode
!= 'r'
3514 _lock_file(self
.f
, exclusive
)
3520 def __exit__(self
, etype
, value
, traceback
):
3522 _unlock_file(self
.f
)
3529 def write(self
, *args
):
3530 return self
.f
.write(*args
)
3532 def read(self
, *args
):
3533 return self
.f
.read(*args
)
3536 def get_filesystem_encoding():
3537 encoding
= sys
.getfilesystemencoding()
3538 return encoding
if encoding
is not None else 'utf-8'
3541 def shell_quote(args
):
3543 encoding
= get_filesystem_encoding()
3545 if isinstance(a
, bytes):
3546 # We may get a filename encoded with 'encodeFilename'
3547 a
= a
.decode(encoding
)
3548 quoted_args
.append(compat_shlex_quote(a
))
3549 return ' '.join(quoted_args
)
3552 def smuggle_url(url
, data
):
3553 """ Pass additional data in a URL for internal use. """
3555 url
, idata
= unsmuggle_url(url
, {})
3557 sdata
= compat_urllib_parse_urlencode(
3558 {'__youtubedl_smuggle': json.dumps(data)}
)
3559 return url
+ '#' + sdata
3562 def unsmuggle_url(smug_url
, default
=None):
3563 if '#__youtubedl_smuggle' not in smug_url
:
3564 return smug_url
, default
3565 url
, _
, sdata
= smug_url
.rpartition('#')
3566 jsond
= compat_parse_qs(sdata
)['__youtubedl_smuggle'][0]
3567 data
= json
.loads(jsond
)
3571 def format_bytes(bytes):
3574 if type(bytes) is str:
3575 bytes = float(bytes)
3579 exponent
= int(math
.log(bytes, 1024.0))
3580 suffix
= ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent
]
3581 converted
= float(bytes) / float(1024 ** exponent
)
3582 return '%.2f%s' % (converted
, suffix
)
3585 def lookup_unit_table(unit_table
, s
):
3586 units_re
= '|'.join(re
.escape(u
) for u
in unit_table
)
3588 r
'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re
, s
)
3591 num_str
= m
.group('num').replace(',', '.')
3592 mult
= unit_table
[m
.group('unit')]
3593 return int(float(num_str
) * mult
)
3596 def parse_filesize(s
):
3600 # The lower-case forms are of course incorrect and unofficial,
3601 # but we support those too
3618 'megabytes': 1000 ** 2,
3619 'mebibytes': 1024 ** 2,
3625 'gigabytes': 1000 ** 3,
3626 'gibibytes': 1024 ** 3,
3632 'terabytes': 1000 ** 4,
3633 'tebibytes': 1024 ** 4,
3639 'petabytes': 1000 ** 5,
3640 'pebibytes': 1024 ** 5,
3646 'exabytes': 1000 ** 6,
3647 'exbibytes': 1024 ** 6,
3653 'zettabytes': 1000 ** 7,
3654 'zebibytes': 1024 ** 7,
3660 'yottabytes': 1000 ** 8,
3661 'yobibytes': 1024 ** 8,
3664 return lookup_unit_table(_UNIT_TABLE
, s
)
3673 if re
.match(r
'^[\d,.]+$', s
):
3674 return str_to_int(s
)
3685 return lookup_unit_table(_UNIT_TABLE
, s
)
3688 def parse_resolution(s
):
3692 mobj
= re
.search(r
'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s
)
3695 'width': int(mobj
.group('w')),
3696 'height': int(mobj
.group('h')),
3699 mobj
= re
.search(r
'\b(\d+)[pPiI]\b', s
)
3701 return {'height': int(mobj.group(1))}
3703 mobj
= re
.search(r
'\b([48])[kK]\b', s
)
3705 return {'height': int(mobj.group(1)) * 540}
3710 def parse_bitrate(s
):
3711 if not isinstance(s
, compat_str
):
3713 mobj
= re
.search(r
'\b(\d+)\s*kbps', s
)
3715 return int(mobj
.group(1))
3718 def month_by_name(name
, lang
='en'):
3719 """ Return the number of a month by (locale-independently) English name """
3721 month_names
= MONTH_NAMES
.get(lang
, MONTH_NAMES
['en'])
3724 return month_names
.index(name
) + 1
3729 def month_by_abbreviation(abbrev
):
3730 """ Return the number of a month by (locale-independently) English
3734 return [s
[:3] for s
in ENGLISH_MONTH_NAMES
].index(abbrev
) + 1
3739 def fix_xml_ampersands(xml_str
):
3740 """Replace all the '&' by '&' in XML"""
3742 r
'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3747 def setproctitle(title
):
3748 assert isinstance(title
, compat_str
)
3750 # ctypes in Jython is not complete
3751 # http://bugs.jython.org/issue2148
3752 if sys
.platform
.startswith('java'):
3756 libc
= ctypes
.cdll
.LoadLibrary('libc.so.6')
3760 # LoadLibrary in Windows Python 2.7.13 only expects
3761 # a bytestring, but since unicode_literals turns
3762 # every string into a unicode string, it fails.
3764 title_bytes
= title
.encode('utf-8')
3765 buf
= ctypes
.create_string_buffer(len(title_bytes
))
3766 buf
.value
= title_bytes
3768 libc
.prctl(15, buf
, 0, 0, 0)
3769 except AttributeError:
3770 return # Strange libc, just skip this
3773 def remove_start(s
, start
):
3774 return s
[len(start
):] if s
is not None and s
.startswith(start
) else s
3777 def remove_end(s
, end
):
3778 return s
[:-len(end
)] if s
is not None and s
.endswith(end
) else s
3781 def remove_quotes(s
):
3782 if s
is None or len(s
) < 2:
3784 for quote
in ('"', "'", ):
3785 if s
[0] == quote
and s
[-1] == quote
:
3790 def get_domain(url
):
3791 domain
= re
.match(r
'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url
)
3792 return domain
.group('domain') if domain
else None
3795 def url_basename(url
):
3796 path
= compat_urlparse
.urlparse(url
).path
3797 return path
.strip('/').split('/')[-1]
3801 return re
.match(r
'https?://[^?#&]+/', url
).group()
3804 def urljoin(base
, path
):
3805 if isinstance(path
, bytes):
3806 path
= path
.decode('utf-8')
3807 if not isinstance(path
, compat_str
) or not path
:
3809 if re
.match(r
'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path
):
3811 if isinstance(base
, bytes):
3812 base
= base
.decode('utf-8')
3813 if not isinstance(base
, compat_str
) or not re
.match(
3814 r
'^(?:https?:)?//', base
):
3816 return compat_urlparse
.urljoin(base
, path
)
3819 class HEADRequest(compat_urllib_request
.Request
):
3820 def get_method(self
):
3824 class PUTRequest(compat_urllib_request
.Request
):
3825 def get_method(self
):
3829 def int_or_none(v
, scale
=1, default
=None, get_attr
=None, invscale
=1):
3832 v
= getattr(v
, get_attr
, None)
3838 return int(v
) * invscale
// scale
3839 except (ValueError, TypeError):
3843 def str_or_none(v
, default
=None):
3844 return default
if v
is None else compat_str(v
)
3847 def str_to_int(int_str
):
3848 """ A more relaxed version of int_or_none """
3849 if isinstance(int_str
, compat_integer_types
):
3851 elif isinstance(int_str
, compat_str
):
3852 int_str
= re
.sub(r
'[,\.\+]', '', int_str
)
3853 return int_or_none(int_str
)
3856 def float_or_none(v
, scale
=1, invscale
=1, default
=None):
3860 return float(v
) * invscale
/ scale
3861 except (ValueError, TypeError):
3865 def bool_or_none(v
, default
=None):
3866 return v
if isinstance(v
, bool) else default
3869 def strip_or_none(v
, default
=None):
3870 return v
.strip() if isinstance(v
, compat_str
) else default
3873 def url_or_none(url
):
3874 if not url
or not isinstance(url
, compat_str
):
3877 return url
if re
.match(r
'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url
) else None
3880 def strftime_or_none(timestamp
, date_format
, default
=None):
3881 datetime_object
= None
3883 if isinstance(timestamp
, compat_numeric_types
): # unix timestamp
3884 datetime_object
= datetime
.datetime
.utcfromtimestamp(timestamp
)
3885 elif isinstance(timestamp
, compat_str
): # assume YYYYMMDD
3886 datetime_object
= datetime
.datetime
.strptime(timestamp
, '%Y%m%d')
3887 return datetime_object
.strftime(date_format
)
3888 except (ValueError, TypeError, AttributeError):
3892 def parse_duration(s
):
3893 if not isinstance(s
, compat_basestring
):
3898 days
, hours
, mins
, secs
, ms
= [None] * 5
3899 m
= re
.match(r
'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s
)
3901 days
, hours
, mins
, secs
, ms
= m
.groups()
3906 [0-9]+\s*y(?:ears?)?\s*
3909 [0-9]+\s*m(?:onths?)?\s*
3912 [0-9]+\s*w(?:eeks?)?\s*
3915 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3919 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3922 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3925 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3928 days
, hours
, mins
, secs
, ms
= m
.groups()
3930 m
= re
.match(r
'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s
)
3932 hours
, mins
= m
.groups()
3938 duration
+= float(secs
)
3940 duration
+= float(mins
) * 60
3942 duration
+= float(hours
) * 60 * 60
3944 duration
+= float(days
) * 24 * 60 * 60
3946 duration
+= float(ms
)
3950 def prepend_extension(filename
, ext
, expected_real_ext
=None):
3951 name
, real_ext
= os
.path
.splitext(filename
)
3953 '{0}.{1}{2}'.format(name
, ext
, real_ext
)
3954 if not expected_real_ext
or real_ext
[1:] == expected_real_ext
3955 else '{0}.{1}'.format(filename
, ext
))
3958 def replace_extension(filename
, ext
, expected_real_ext
=None):
3959 name
, real_ext
= os
.path
.splitext(filename
)
3960 return '{0}.{1}'.format(
3961 name
if not expected_real_ext
or real_ext
[1:] == expected_real_ext
else filename
,
3965 def check_executable(exe
, args
=[]):
3966 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3967 args can be a list of arguments for a short output (like -version) """
3969 process_communicate_or_kill(subprocess
.Popen(
3970 [exe
] + args
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
))
3976 def get_exe_version(exe
, args
=['--version'],
3977 version_re
=None, unrecognized
='present'):
3978 """ Returns the version of the specified executable,
3979 or False if the executable is not present """
3981 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3982 # SIGTTOU if yt-dlp is run in the background.
3983 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3984 out
, _
= process_communicate_or_kill(subprocess
.Popen(
3985 [encodeArgument(exe
)] + args
,
3986 stdin
=subprocess
.PIPE
,
3987 stdout
=subprocess
.PIPE
, stderr
=subprocess
.STDOUT
))
3990 if isinstance(out
, bytes): # Python 2.x
3991 out
= out
.decode('ascii', 'ignore')
3992 return detect_exe_version(out
, version_re
, unrecognized
)
3995 def detect_exe_version(output
, version_re
=None, unrecognized
='present'):
3996 assert isinstance(output
, compat_str
)
3997 if version_re
is None:
3998 version_re
= r
'version\s+([-0-9._a-zA-Z]+)'
3999 m
= re
.search(version_re
, output
)
4006 class LazyList(collections
.abc
.Sequence
):
4007 ''' Lazy immutable list from an iterable
4008 Note that slices of a LazyList are lists and not LazyList'''
4010 class IndexError(IndexError):
4013 def __init__(self
, iterable
):
4014 self
.__iterable
= iter(iterable
)
4016 self
.__reversed
= False
4020 # We need to consume the entire iterable to iterate in reverse
4021 yield from self
.exhaust()
4023 yield from self
.__cache
4024 for item
in self
.__iterable
:
4025 self
.__cache
.append(item
)
4028 def __exhaust(self
):
4029 self
.__cache
.extend(self
.__iterable
)
4033 ''' Evaluate the entire iterable '''
4034 return self
.__exhaust
()[::-1 if self
.__reversed
else 1]
4037 def __reverse_index(x
):
4038 return None if x
is None else -(x
+ 1)
4040 def __getitem__(self
, idx
):
4041 if isinstance(idx
, slice):
4043 idx
= slice(self
.__reverse
_index
(idx
.start
), self
.__reverse
_index
(idx
.stop
), -(idx
.step
or 1))
4044 start
, stop
, step
= idx
.start
, idx
.stop
, idx
.step
or 1
4045 elif isinstance(idx
, int):
4047 idx
= self
.__reverse
_index
(idx
)
4048 start
, stop
, step
= idx
, idx
, 0
4050 raise TypeError('indices must be integers or slices')
4051 if ((start
or 0) < 0 or (stop
or 0) < 0
4052 or (start
is None and step
< 0)
4053 or (stop
is None and step
> 0)):
4054 # We need to consume the entire iterable to be able to slice from the end
4055 # Obviously, never use this with infinite iterables
4058 return self
.__cache
[idx
]
4059 except IndexError as e
:
4060 raise self
.IndexError(e
) from e
4061 n
= max(start
or 0, stop
or 0) - len(self
.__cache
) + 1
4063 self
.__cache
.extend(itertools
.islice(self
.__iterable
, n
))
4065 return self
.__cache
[idx
]
4066 except IndexError as e
:
4067 raise self
.IndexError(e
) from e
4071 self
[-1] if self
.__reversed
else self
[0]
4072 except self
.IndexError:
4078 return len(self
.__cache
)
4081 self
.__reversed
= not self
.__reversed
4085 # repr and str should mimic a list. So we exhaust the iterable
4086 return repr(self
.exhaust())
4089 return repr(self
.exhaust())
4094 # This is only useful for tests
4095 return len(self
.getslice())
4097 def __init__(self
, pagefunc
, pagesize
, use_cache
=True):
4098 self
._pagefunc
= pagefunc
4099 self
._pagesize
= pagesize
4100 self
._use
_cache
= use_cache
4103 def getpage(self
, pagenum
):
4104 page_results
= self
._cache
.get(pagenum
) or list(self
._pagefunc
(pagenum
))
4106 self
._cache
[pagenum
] = page_results
4109 def getslice(self
, start
=0, end
=None):
4110 return list(self
._getslice
(start
, end
))
4112 def _getslice(self
, start
, end
):
4113 raise NotImplementedError('This method must be implemented by subclasses')
4115 def __getitem__(self
, idx
):
4116 # NOTE: cache must be enabled if this is used
4117 if not isinstance(idx
, int) or idx
< 0:
4118 raise TypeError('indices must be non-negative integers')
4119 entries
= self
.getslice(idx
, idx
+ 1)
4120 return entries
[0] if entries
else None
4123 class OnDemandPagedList(PagedList
):
4124 def _getslice(self
, start
, end
):
4125 for pagenum
in itertools
.count(start
// self
._pagesize
):
4126 firstid
= pagenum
* self
._pagesize
4127 nextfirstid
= pagenum
* self
._pagesize
+ self
._pagesize
4128 if start
>= nextfirstid
:
4132 start
% self
._pagesize
4133 if firstid
<= start
< nextfirstid
4136 ((end
- 1) % self
._pagesize
) + 1
4137 if (end
is not None and firstid
<= end
<= nextfirstid
)
4140 page_results
= self
.getpage(pagenum
)
4141 if startv
!= 0 or endv
is not None:
4142 page_results
= page_results
[startv
:endv
]
4143 yield from page_results
4145 # A little optimization - if current page is not "full", ie. does
4146 # not contain page_size videos then we can assume that this page
4147 # is the last one - there are no more ids on further pages -
4148 # i.e. no need to query again.
4149 if len(page_results
) + startv
< self
._pagesize
:
4152 # If we got the whole page, but the next page is not interesting,
4153 # break out early as well
4154 if end
== nextfirstid
:
4158 class InAdvancePagedList(PagedList
):
4159 def __init__(self
, pagefunc
, pagecount
, pagesize
):
4160 self
._pagecount
= pagecount
4161 PagedList
.__init
__(self
, pagefunc
, pagesize
, True)
4163 def _getslice(self
, start
, end
):
4164 start_page
= start
// self
._pagesize
4166 self
._pagecount
if end
is None else (end
// self
._pagesize
+ 1))
4167 skip_elems
= start
- start_page
* self
._pagesize
4168 only_more
= None if end
is None else end
- start
4169 for pagenum
in range(start_page
, end_page
):
4170 page_results
= self
.getpage(pagenum
)
4172 page_results
= page_results
[skip_elems
:]
4174 if only_more
is not None:
4175 if len(page_results
) < only_more
:
4176 only_more
-= len(page_results
)
4178 yield from page_results
[:only_more
]
4180 yield from page_results
4183 def uppercase_escape(s
):
4184 unicode_escape
= codecs
.getdecoder('unicode_escape')
4186 r
'\\U[0-9a-fA-F]{8}',
4187 lambda m
: unicode_escape(m
.group(0))[0],
4191 def lowercase_escape(s
):
4192 unicode_escape
= codecs
.getdecoder('unicode_escape')
4194 r
'\\u[0-9a-fA-F]{4}',
4195 lambda m
: unicode_escape(m
.group(0))[0],
4199 def escape_rfc3986(s
):
4200 """Escape non-ASCII characters as suggested by RFC 3986"""
4201 if sys
.version_info
< (3, 0) and isinstance(s
, compat_str
):
4202 s
= s
.encode('utf-8')
4203 return compat_urllib_parse
.quote(s
, b
"%/;:@&=+$,!~*'()?#[]")
4206 def escape_url(url
):
4207 """Escape URL as suggested by RFC 3986"""
4208 url_parsed
= compat_urllib_parse_urlparse(url
)
4209 return url_parsed
._replace
(
4210 netloc
=url_parsed
.netloc
.encode('idna').decode('ascii'),
4211 path
=escape_rfc3986(url_parsed
.path
),
4212 params
=escape_rfc3986(url_parsed
.params
),
4213 query
=escape_rfc3986(url_parsed
.query
),
4214 fragment
=escape_rfc3986(url_parsed
.fragment
)
4219 return compat_parse_qs(compat_urllib_parse_urlparse(url
).query
)
4222 def read_batch_urls(batch_fd
):
4224 if not isinstance(url
, compat_str
):
4225 url
= url
.decode('utf-8', 'replace')
4226 BOM_UTF8
= ('\xef\xbb\xbf', '\ufeff')
4227 for bom
in BOM_UTF8
:
4228 if url
.startswith(bom
):
4229 url
= url
[len(bom
):]
4231 if not url
or url
.startswith(('#', ';', ']')):
4233 # "#" cannot be stripped out since it is part of the URI
4234 # However, it can be safely stipped out if follwing a whitespace
4235 return re
.split(r
'\s#', url
, 1)[0].rstrip()
4237 with contextlib
.closing(batch_fd
) as fd
:
4238 return [url
for url
in map(fixup
, fd
) if url
]
4241 def urlencode_postdata(*args
, **kargs
):
4242 return compat_urllib_parse_urlencode(*args
, **kargs
).encode('ascii')
4245 def update_url_query(url
, query
):
4248 parsed_url
= compat_urlparse
.urlparse(url
)
4249 qs
= compat_parse_qs(parsed_url
.query
)
4251 return compat_urlparse
.urlunparse(parsed_url
._replace
(
4252 query
=compat_urllib_parse_urlencode(qs
, True)))
4255 def update_Request(req
, url
=None, data
=None, headers
={}, query={}
):
4256 req_headers
= req
.headers
.copy()
4257 req_headers
.update(headers
)
4258 req_data
= data
or req
.data
4259 req_url
= update_url_query(url
or req
.get_full_url(), query
)
4260 req_get_method
= req
.get_method()
4261 if req_get_method
== 'HEAD':
4262 req_type
= HEADRequest
4263 elif req_get_method
== 'PUT':
4264 req_type
= PUTRequest
4266 req_type
= compat_urllib_request
.Request
4268 req_url
, data
=req_data
, headers
=req_headers
,
4269 origin_req_host
=req
.origin_req_host
, unverifiable
=req
.unverifiable
)
4270 if hasattr(req
, 'timeout'):
4271 new_req
.timeout
= req
.timeout
4275 def _multipart_encode_impl(data
, boundary
):
4276 content_type
= 'multipart/form-data; boundary=%s' % boundary
4279 for k
, v
in data
.items():
4280 out
+= b
'--' + boundary
.encode('ascii') + b
'\r\n'
4281 if isinstance(k
, compat_str
):
4282 k
= k
.encode('utf-8')
4283 if isinstance(v
, compat_str
):
4284 v
= v
.encode('utf-8')
4285 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
4286 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
4287 content
= b
'Content-Disposition: form-data; name="' + k
+ b
'"\r\n\r\n' + v
+ b
'\r\n'
4288 if boundary
.encode('ascii') in content
:
4289 raise ValueError('Boundary overlaps with data')
4292 out
+= b
'--' + boundary
.encode('ascii') + b
'--\r\n'
4294 return out
, content_type
4297 def multipart_encode(data
, boundary
=None):
4299 Encode a dict to RFC 7578-compliant form-data
4302 A dict where keys and values can be either Unicode or bytes-like
4305 If specified a Unicode object, it's used as the boundary. Otherwise
4306 a random boundary is generated.
4308 Reference: https://tools.ietf.org/html/rfc7578
4310 has_specified_boundary
= boundary
is not None
4313 if boundary
is None:
4314 boundary
= '---------------' + str(random
.randrange(0x0fffffff, 0xffffffff))
4317 out
, content_type
= _multipart_encode_impl(data
, boundary
)
4320 if has_specified_boundary
:
4324 return out
, content_type
4327 def dict_get(d
, key_or_keys
, default
=None, skip_false_values
=True):
4328 if isinstance(key_or_keys
, (list, tuple)):
4329 for key
in key_or_keys
:
4330 if key
not in d
or d
[key
] is None or skip_false_values
and not d
[key
]:
4334 return d
.get(key_or_keys
, default
)
4337 def try_get(src
, getter
, expected_type
=None):
4338 for get
in variadic(getter
):
4341 except (AttributeError, KeyError, TypeError, IndexError):
4344 if expected_type
is None or isinstance(v
, expected_type
):
4348 def merge_dicts(*dicts
):
4350 for a_dict
in dicts
:
4351 for k
, v
in a_dict
.items():
4355 or (isinstance(v
, compat_str
) and v
4356 and isinstance(merged
[k
], compat_str
)
4357 and not merged
[k
])):
4362 def encode_compat_str(string
, encoding
=preferredencoding(), errors
='strict'):
4363 return string
if isinstance(string
, compat_str
) else compat_str(string
, encoding
, errors
)
4375 TV_PARENTAL_GUIDELINES
= {
4385 def parse_age_limit(s
):
4387 return s
if 0 <= s
<= 21 else None
4388 if not isinstance(s
, compat_basestring
):
4390 m
= re
.match(r
'^(?P<age>\d{1,2})\+?$', s
)
4392 return int(m
.group('age'))
4395 return US_RATINGS
[s
]
4396 m
= re
.match(r
'^TV[_-]?(%s)$' % '|'.join(k
[3:] for k
in TV_PARENTAL_GUIDELINES
), s
)
4398 return TV_PARENTAL_GUIDELINES
['TV-' + m
.group(1)]
4402 def strip_jsonp(code
):
4405 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
4406 (?:\s*&&\s*(?P=func_name))?
4407 \s*\(\s*(?P<callback_data>.*)\);?
4408 \s*?(?://[^\n]*)*$''',
4409 r
'\g<callback_data>', code
)
4412 def js_to_json(code
, vars={}):
4413 # vars is a dict of var, val pairs to substitute
4414 COMMENT_RE
= r
'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n'
4415 SKIP_RE
= r
'\s*(?:{comment})?\s*'.format(comment
=COMMENT_RE
)
4417 (r
'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip
=SKIP_RE
), 16),
4418 (r
'(?s)^(0+[0-7]+){skip}:?$'.format(skip
=SKIP_RE
), 8),
4423 if v
in ('true', 'false', 'null'):
4425 elif v
in ('undefined', 'void 0'):
4427 elif v
.startswith('/*') or v
.startswith('//') or v
.startswith('!') or v
== ',':
4430 if v
[0] in ("'", '"'):
4431 v
= re
.sub(r
'(?s)\\.|"', lambda m
: {
4436 }.get(m
.group(0), m
.group(0)), v
[1:-1])
4438 for regex
, base
in INTEGER_TABLE
:
4439 im
= re
.match(regex
, v
)
4441 i
= int(im
.group(1), base
)
4442 return '"%d":' % i
if v
.endswith(':') else '%d' % i
4449 return re
.sub(r
'''(?sx)
4450 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4451 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4452 {comment}|,(?={skip}[\]}}])|
4453 void\s0|(?:(?<![0-9])[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*|
4454 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4457 '''.format(comment
=COMMENT_RE
, skip
=SKIP_RE
), fix_kv
, code
)
4460 def qualities(quality_ids
):
4461 """ Get a numeric quality value out of a list of possible values """
4464 return quality_ids
.index(qid
)
4471 'default': '%(title)s [%(id)s].%(ext)s',
4472 'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s',
4478 'description': 'description',
4479 'annotation': 'annotations.xml',
4480 'infojson': 'info.json',
4481 'pl_thumbnail': None,
4482 'pl_description': 'description',
4483 'pl_infojson': 'info.json',
4486 # As of [1] format syntax is:
4487 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
4488 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
4489 STR_FORMAT_RE_TMPL
= r
'''(?x)
4490 (?<!%)(?P<prefix>(?:%%)*)
4492 (?P<has_key>\((?P<key>{0})\))?
4494 (?P<conversion>[#0\-+ ]+)?
4496 (?P<precision>\.\d+)?
4497 (?P<len_mod>[hlL])? # unused in python
4498 {1} # conversion type
4503 STR_FORMAT_TYPES
= 'diouxXeEfFgGcrs'
4506 def limit_length(s
, length
):
4507 """ Add ellipses to overly long strings """
4512 return s
[:length
- len(ELLIPSES
)] + ELLIPSES
4516 def version_tuple(v
):
4517 return tuple(int(e
) for e
in re
.split(r
'[-.]', v
))
4520 def is_outdated_version(version
, limit
, assume_new
=True):
4522 return not assume_new
4524 return version_tuple(version
) < version_tuple(limit
)
4526 return not assume_new
4529 def ytdl_is_updateable():
4530 """ Returns if yt-dlp can be updated with -U """
4532 from .update
import is_non_updateable
4534 return not is_non_updateable()
4537 def args_to_str(args
):
4538 # Get a short string representation for a subprocess command
4539 return ' '.join(compat_shlex_quote(a
) for a
in args
)
4542 def error_to_compat_str(err
):
4544 # On python 2 error byte string must be decoded with proper
4545 # encoding rather than ascii
4546 if sys
.version_info
[0] < 3:
4547 err_str
= err_str
.decode(preferredencoding())
4551 def mimetype2ext(mt
):
4555 mt
, _
, params
= mt
.partition(';')
4560 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4561 # it's the most popular one
4562 'audio/mpeg': 'mp3',
4563 'audio/x-wav': 'wav',
4565 'audio/wave': 'wav',
4568 ext
= FULL_MAP
.get(mt
)
4574 'smptett+xml': 'tt',
4578 'x-mp4-fragmented': 'mp4',
4579 'x-ms-sami': 'sami',
4582 'x-mpegurl': 'm3u8',
4583 'vnd.apple.mpegurl': 'm3u8',
4587 'vnd.ms-sstr+xml': 'ism',
4591 'filmstrip+json': 'fs',
4595 _
, _
, subtype
= mt
.rpartition('/')
4596 ext
= SUBTYPE_MAP
.get(subtype
.lower())
4607 _
, _
, suffix
= subtype
.partition('+')
4608 ext
= SUFFIX_MAP
.get(suffix
)
4612 return subtype
.replace('+', '.')
4615 def parse_codecs(codecs_str
):
4616 # http://tools.ietf.org/html/rfc6381
4619 split_codecs
= list(filter(None, map(
4620 str.strip
, codecs_str
.strip().strip(',').split(','))))
4621 vcodec
, acodec
= None, None
4622 for full_codec
in split_codecs
:
4623 codec
= full_codec
.split('.')[0]
4624 if codec
in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4627 elif codec
in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4631 write_string('WARNING: Unknown codec %s\n' % full_codec
, sys
.stderr
)
4632 if not vcodec
and not acodec
:
4633 if len(split_codecs
) == 2:
4635 'vcodec': split_codecs
[0],
4636 'acodec': split_codecs
[1],
4640 'vcodec': vcodec
or 'none',
4641 'acodec': acodec
or 'none',
4646 def urlhandle_detect_ext(url_handle
):
4647 getheader
= url_handle
.headers
.get
4649 cd
= getheader('Content-Disposition')
4651 m
= re
.match(r
'attachment;\s*filename="(?P<filename>[^"]+)"', cd
)
4653 e
= determine_ext(m
.group('filename'), default_ext
=None)
4657 return mimetype2ext(getheader('Content-Type'))
4660 def encode_data_uri(data
, mime_type
):
4661 return 'data:%s;base64,%s' % (mime_type
, base64
.b64encode(data
).decode('ascii'))
4664 def age_restricted(content_limit
, age_limit
):
4665 """ Returns True iff the content should be blocked """
4667 if age_limit
is None: # No limit set
4669 if content_limit
is None:
4670 return False # Content available for everyone
4671 return age_limit
< content_limit
4674 def is_html(first_bytes
):
4675 """ Detect whether a file contains HTML by examining its first bytes. """
4678 (b
'\xef\xbb\xbf', 'utf-8'),
4679 (b
'\x00\x00\xfe\xff', 'utf-32-be'),
4680 (b
'\xff\xfe\x00\x00', 'utf-32-le'),
4681 (b
'\xff\xfe', 'utf-16-le'),
4682 (b
'\xfe\xff', 'utf-16-be'),
4684 for bom
, enc
in BOMS
:
4685 if first_bytes
.startswith(bom
):
4686 s
= first_bytes
[len(bom
):].decode(enc
, 'replace')
4689 s
= first_bytes
.decode('utf-8', 'replace')
4691 return re
.match(r
'^\s*<', s
)
4694 def determine_protocol(info_dict
):
4695 protocol
= info_dict
.get('protocol')
4696 if protocol
is not None:
4699 url
= info_dict
['url']
4700 if url
.startswith('rtmp'):
4702 elif url
.startswith('mms'):
4704 elif url
.startswith('rtsp'):
4707 ext
= determine_ext(url
)
4713 return compat_urllib_parse_urlparse(url
).scheme
4716 def render_table(header_row
, data
, delim
=False, extraGap
=0, hideEmpty
=False):
4717 """ Render a list of rows, each as a list of values """
4719 def get_max_lens(table
):
4720 return [max(len(compat_str(v
)) for v
in col
) for col
in zip(*table
)]
4722 def filter_using_list(row
, filterArray
):
4723 return [col
for (take
, col
) in zip(filterArray
, row
) if take
]
4726 max_lens
= get_max_lens(data
)
4727 header_row
= filter_using_list(header_row
, max_lens
)
4728 data
= [filter_using_list(row
, max_lens
) for row
in data
]
4730 table
= [header_row
] + data
4731 max_lens
= get_max_lens(table
)
4733 table
= [header_row
] + [['-' * ml
for ml
in max_lens
]] + data
4734 format_str
= ' '.join('%-' + compat_str(ml
+ extraGap
) + 's' for ml
in max_lens
[:-1]) + ' %s'
4735 return '\n'.join(format_str
% tuple(row
) for row
in table
)
4738 def _match_one(filter_part
, dct
, incomplete
):
4739 # TODO: Generalize code with YoutubeDL._build_format_filter
4740 STRING_OPERATORS
= {
4741 '*=': operator
.contains
,
4742 '^=': lambda attr
, value
: attr
.startswith(value
),
4743 '$=': lambda attr
, value
: attr
.endswith(value
),
4744 '~=': lambda attr
, value
: re
.search(value
, attr
),
4746 COMPARISON_OPERATORS
= {
4748 '<=': operator
.le
, # "<=" must be defined above "<"
4755 operator_rex
= re
.compile(r
'''(?x)\s*
4757 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4759 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4760 (?P<quote>["\'])(?P
<quotedstrval
>.+?
)(?P
=quote
)|
4764 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4765 m = operator_rex.search(filter_part)
4767 unnegated_op = COMPARISON_OPERATORS[m.group('op')]
4768 if m.group('negation'):
4769 op = lambda attr, value: not unnegated_op(attr, value)
4772 actual_value = dct.get(m.group('key'))
4773 if (m.group('quotedstrval') is not None
4774 or m.group('strval') is not None
4775 # If the original field is a string and matching comparisonvalue is
4776 # a number we should respect the origin of the original field
4777 # and process comparison value as a string (see
4778 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4779 or actual_value is not None and m.group('intval') is not None
4780 and isinstance(actual_value, compat_str)):
4781 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4782 quote = m.group('quote')
4783 if quote is not None:
4784 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4786 if m.group('op') in STRING_OPERATORS:
4787 raise ValueError('Operator %s only supports string values!' % m.group('op'))
4789 comparison_value = int(m.group('intval'))
4791 comparison_value = parse_filesize(m.group('intval'))
4792 if comparison_value is None:
4793 comparison_value = parse_filesize(m.group('intval') + 'B')
4794 if comparison_value is None:
4796 'Invalid integer value %r in filter part %r' % (
4797 m.group('intval'), filter_part))
4798 if actual_value is None:
4799 return incomplete or m.group('none_inclusive')
4800 return op(actual_value, comparison_value)
4803 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4804 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4806 operator_rex = re.compile(r'''(?x
)\s
*
4807 (?P
<op
>%s)\s
*(?P
<key
>[a
-z_
]+)
4809 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4810 m = operator_rex.search(filter_part)
4812 op = UNARY_OPERATORS[m.group('op')]
4813 actual_value = dct.get(m.group('key'))
4814 if incomplete and actual_value is None:
4816 return op(actual_value)
4818 raise ValueError('Invalid filter part %r' % filter_part)
4821 def match_str(filter_str, dct, incomplete=False):
4822 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false
4823 When incomplete, all conditions passes on missing fields
4826 _match_one(filter_part.replace(r'\&', '&'), dct, incomplete)
4827 for filter_part in re.split(r'(?<!\\)&', filter_str))
4830 def match_filter_func(filter_str):
4831 def _match_func(info_dict, *args, **kwargs):
4832 if match_str(filter_str, info_dict, *args, **kwargs):
4835 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4836 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4840 def parse_dfxp_time_expr(time_expr):
4844 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4846 return float(mobj.group('time_offset'))
4848 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4850 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4853 def srt_subtitles_timecode(seconds):
4854 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4857 def dfxp2srt(dfxp_data):
4859 @param dfxp_data A
bytes-like
object containing DFXP data
4860 @returns A
unicode object containing converted SRT data
4862 LEGACY_NAMESPACES = (
4863 (b'http://www.w3.org/ns/ttml', [
4864 b'http://www.w3.org/2004/11/ttaf1',
4865 b'http://www.w3.org/2006/04/ttaf1',
4866 b'http://www.w3.org/2006/10/ttaf1',
4868 (b'http://www.w3.org/ns/ttml#styling', [
4869 b'http://www.w3.org/ns/ttml#style',
4873 SUPPORTED_STYLING = [
4882 _x = functools.partial(xpath_with_ns, ns_map={
4883 'xml': 'http://www.w3.org/XML/1998/namespace',
4884 'ttml': 'http://www.w3.org/ns/ttml',
4885 'tts': 'http://www.w3.org/ns/ttml#styling',
4891 class TTMLPElementParser(object):
4893 _unclosed_elements = []
4894 _applied_styles = []
4896 def start(self, tag, attrib):
4897 if tag in (_x('ttml:br'), 'br'):
4900 unclosed_elements = []
4902 element_style_id = attrib.get('style')
4904 style.update(default_style)
4905 if element_style_id:
4906 style.update(styles.get(element_style_id, {}))
4907 for prop in SUPPORTED_STYLING:
4908 prop_val = attrib.get(_x('tts:' + prop))
4910 style[prop] = prop_val
4913 for k, v in sorted(style.items()):
4914 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4917 font += ' color="%s"' % v
4918 elif k == 'fontSize':
4919 font += ' size="%s"' % v
4920 elif k == 'fontFamily':
4921 font += ' face="%s"' % v
4922 elif k == 'fontWeight' and v == 'bold':
4924 unclosed_elements.append('b')
4925 elif k == 'fontStyle' and v == 'italic':
4927 unclosed_elements.append('i')
4928 elif k == 'textDecoration' and v == 'underline':
4930 unclosed_elements.append('u')
4932 self._out += '<font' + font + '>'
4933 unclosed_elements.append('font')
4935 if self._applied_styles:
4936 applied_style.update(self._applied_styles[-1])
4937 applied_style.update(style)
4938 self._applied_styles.append(applied_style)
4939 self._unclosed_elements.append(unclosed_elements)
4942 if tag not in (_x('ttml:br'), 'br'):
4943 unclosed_elements = self._unclosed_elements.pop()
4944 for element in reversed(unclosed_elements):
4945 self._out += '</%s>' % element
4946 if unclosed_elements and self._applied_styles:
4947 self._applied_styles.pop()
4949 def data(self, data):
4953 return self._out.strip()
4955 def parse_node(node):
4956 target = TTMLPElementParser()
4957 parser = xml.etree.ElementTree.XMLParser(target=target)
4958 parser.feed(xml.etree.ElementTree.tostring(node))
4959 return parser.close()
4961 for k, v in LEGACY_NAMESPACES:
4963 dfxp_data = dfxp_data.replace(ns, k)
4965 dfxp = compat_etree_fromstring(dfxp_data)
4967 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4970 raise ValueError('Invalid dfxp/TTML subtitle')
4974 for style in dfxp.findall(_x('.//ttml:style')):
4975 style_id = style.get('id') or style.get(_x('xml:id'))
4978 parent_style_id = style.get('style')
4980 if parent_style_id not in styles:
4983 styles[style_id] = styles[parent_style_id].copy()
4984 for prop in SUPPORTED_STYLING:
4985 prop_val = style.get(_x('tts:' + prop))
4987 styles.setdefault(style_id, {})[prop] = prop_val
4993 for p in ('body', 'div'):
4994 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4997 style = styles.get(ele.get('style'))
5000 default_style.update(style)
5002 for para, index in zip(paras, itertools.count(1)):
5003 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
5004 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
5005 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
5006 if begin_time is None:
5011 end_time = begin_time + dur
5012 out.append('%d\n%s --> %s\n%s\n\n' % (
5014 srt_subtitles_timecode(begin_time),
5015 srt_subtitles_timecode(end_time),
5021 def cli_option(params, command_option, param):
5022 param = params.get(param)
5024 param = compat_str(param)
5025 return [command_option, param] if param is not None else []
5028 def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
5029 param = params.get(param)
5032 assert isinstance(param, bool)
5034 return [command_option + separator + (true_value if param else false_value)]
5035 return [command_option, true_value if param else false_value]
5038 def cli_valueless_option(params, command_option, param, expected_value=True):
5039 param = params.get(param)
5040 return [command_option] if param == expected_value else []
5043 def cli_configuration_args(argdict, keys, default=[], use_compat=True):
5044 if isinstance(argdict, (list, tuple)): # for backward compatibility
5051 assert isinstance(argdict, dict)
5053 assert isinstance(keys, (list, tuple))
5054 for key_list in keys:
5055 arg_list = list(filter(
5056 lambda x: x is not None,
5057 [argdict.get(key.lower()) for key in variadic(key_list)]))
5059 return [arg for args in arg_list for arg in args]
5063 def _configuration_args(main_key, argdict, exe, keys=None, default=[], use_compat=True):
5064 main_key, exe = main_key.lower(), exe.lower()
5065 root_key = exe if main_key == exe else f'{main_key}+{exe}'
5066 keys = [f'{root_key}{k}' for k in (keys or [''])]
5067 if root_key in keys:
5069 keys.append((main_key, exe))
5070 keys.append('default')
5073 return cli_configuration_args(argdict, keys, default, use_compat)
5076 class ISO639Utils(object):
5077 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
5136 'iw': 'heb', # Replaced by he in 1989 revision
5146 'in': 'ind', # Replaced by id in 1989 revision
5261 'ji': 'yid', # Replaced by yi in 1989 revision
5269 def short2long(cls, code):
5270 """Convert language code from ISO 639-1 to ISO 639-2/T"""
5271 return cls._lang_map.get(code[:2])
5274 def long2short(cls, code):
5275 """Convert language code from ISO 639-2/T to ISO 639-1"""
5276 for short_name, long_name in cls._lang_map.items():
5277 if long_name == code:
5281 class ISO3166Utils(object):
5282 # From http://data.okfn.org/data/core/country-list
5284 'AF': 'Afghanistan',
5285 'AX': 'Åland Islands',
5288 'AS': 'American Samoa',
5293 'AG': 'Antigua and Barbuda',
5310 'BO': 'Bolivia, Plurinational State of',
5311 'BQ': 'Bonaire, Sint Eustatius and Saba',
5312 'BA': 'Bosnia and Herzegovina',
5314 'BV': 'Bouvet Island',
5316 'IO': 'British Indian Ocean Territory',
5317 'BN': 'Brunei Darussalam',
5319 'BF': 'Burkina Faso',
5325 'KY': 'Cayman Islands',
5326 'CF': 'Central African Republic',
5330 'CX': 'Christmas Island',
5331 'CC': 'Cocos (Keeling) Islands',
5335 'CD': 'Congo, the Democratic Republic of the',
5336 'CK': 'Cook Islands',
5338 'CI': 'Côte d\'Ivoire',
5343 'CZ': 'Czech Republic',
5347 'DO': 'Dominican Republic',
5350 'SV': 'El Salvador',
5351 'GQ': 'Equatorial Guinea',
5355 'FK': 'Falkland Islands (Malvinas)',
5356 'FO': 'Faroe Islands',
5360 'GF': 'French Guiana',
5361 'PF': 'French Polynesia',
5362 'TF': 'French Southern Territories',
5377 'GW': 'Guinea-Bissau',
5380 'HM': 'Heard Island and McDonald Islands',
5381 'VA': 'Holy See (Vatican City State)',
5388 'IR': 'Iran, Islamic Republic of',
5391 'IM': 'Isle of Man',
5401 'KP': 'Korea, Democratic People\'s Republic of',
5402 'KR': 'Korea, Republic of',
5405 'LA': 'Lao People\'s Democratic Republic',
5411 'LI': 'Liechtenstein',
5415 'MK': 'Macedonia, the Former Yugoslav Republic of',
5422 'MH': 'Marshall Islands',
5428 'FM': 'Micronesia, Federated States of',
5429 'MD': 'Moldova, Republic of',
5440 'NL': 'Netherlands',
5441 'NC': 'New Caledonia',
5442 'NZ': 'New Zealand',
5447 'NF': 'Norfolk Island',
5448 'MP': 'Northern Mariana Islands',
5453 'PS': 'Palestine, State of',
5455 'PG': 'Papua New Guinea',
5458 'PH': 'Philippines',
5462 'PR': 'Puerto Rico',
5466 'RU': 'Russian Federation',
5468 'BL': 'Saint Barthélemy',
5469 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
5470 'KN': 'Saint Kitts and Nevis',
5471 'LC': 'Saint Lucia',
5472 'MF': 'Saint Martin (French part)',
5473 'PM': 'Saint Pierre and Miquelon',
5474 'VC': 'Saint Vincent and the Grenadines',
5477 'ST': 'Sao Tome and Principe',
5478 'SA': 'Saudi Arabia',
5482 'SL': 'Sierra Leone',
5484 'SX': 'Sint Maarten (Dutch part)',
5487 'SB': 'Solomon Islands',
5489 'ZA': 'South Africa',
5490 'GS': 'South Georgia and the South Sandwich Islands',
5491 'SS': 'South Sudan',
5496 'SJ': 'Svalbard and Jan Mayen',
5499 'CH': 'Switzerland',
5500 'SY': 'Syrian Arab Republic',
5501 'TW': 'Taiwan, Province of China',
5503 'TZ': 'Tanzania, United Republic of',
5505 'TL': 'Timor-Leste',
5509 'TT': 'Trinidad and Tobago',
5512 'TM': 'Turkmenistan',
5513 'TC': 'Turks and Caicos Islands',
5517 'AE': 'United Arab Emirates',
5518 'GB': 'United Kingdom',
5519 'US': 'United States',
5520 'UM': 'United States Minor Outlying Islands',
5524 'VE': 'Venezuela, Bolivarian Republic of',
5526 'VG': 'Virgin Islands, British',
5527 'VI': 'Virgin Islands, U.S.',
5528 'WF': 'Wallis and Futuna',
5529 'EH': 'Western Sahara',
5536 def short2full(cls, code):
5537 """Convert an ISO 3166-2 country code to the corresponding full name"""
5538 return cls._country_map.get(code.upper())
5541 class GeoUtils(object):
5542 # Major IPv4 address blocks per country
5544 'AD': '46.172.224.0/19',
5545 'AE': '94.200.0.0/13',
5546 'AF': '149.54.0.0/17',
5547 'AG': '209.59.64.0/18',
5548 'AI': '204.14.248.0/21',
5549 'AL': '46.99.0.0/16',
5550 'AM': '46.70.0.0/15',
5551 'AO': '105.168.0.0/13',
5552 'AP': '182.50.184.0/21',
5553 'AQ': '23.154.160.0/24',
5554 'AR': '181.0.0.0/12',
5555 'AS': '202.70.112.0/20',
5556 'AT': '77.116.0.0/14',
5557 'AU': '1.128.0.0/11',
5558 'AW': '181.41.0.0/18',
5559 'AX': '185.217.4.0/22',
5560 'AZ': '5.197.0.0/16',
5561 'BA': '31.176.128.0/17',
5562 'BB': '65.48.128.0/17',
5563 'BD': '114.130.0.0/16',
5565 'BF': '102.178.0.0/15',
5566 'BG': '95.42.0.0/15',
5567 'BH': '37.131.0.0/17',
5568 'BI': '154.117.192.0/18',
5569 'BJ': '137.255.0.0/16',
5570 'BL': '185.212.72.0/23',
5571 'BM': '196.12.64.0/18',
5572 'BN': '156.31.0.0/16',
5573 'BO': '161.56.0.0/16',
5574 'BQ': '161.0.80.0/20',
5575 'BR': '191.128.0.0/12',
5576 'BS': '24.51.64.0/18',
5577 'BT': '119.2.96.0/19',
5578 'BW': '168.167.0.0/16',
5579 'BY': '178.120.0.0/13',
5580 'BZ': '179.42.192.0/18',
5581 'CA': '99.224.0.0/11',
5582 'CD': '41.243.0.0/16',
5583 'CF': '197.242.176.0/21',
5584 'CG': '160.113.0.0/16',
5585 'CH': '85.0.0.0/13',
5586 'CI': '102.136.0.0/14',
5587 'CK': '202.65.32.0/19',
5588 'CL': '152.172.0.0/14',
5589 'CM': '102.244.0.0/14',
5590 'CN': '36.128.0.0/10',
5591 'CO': '181.240.0.0/12',
5592 'CR': '201.192.0.0/12',
5593 'CU': '152.206.0.0/15',
5594 'CV': '165.90.96.0/19',
5595 'CW': '190.88.128.0/17',
5596 'CY': '31.153.0.0/16',
5597 'CZ': '88.100.0.0/14',
5599 'DJ': '197.241.0.0/17',
5600 'DK': '87.48.0.0/12',
5601 'DM': '192.243.48.0/20',
5602 'DO': '152.166.0.0/15',
5603 'DZ': '41.96.0.0/12',
5604 'EC': '186.68.0.0/15',
5605 'EE': '90.190.0.0/15',
5606 'EG': '156.160.0.0/11',
5607 'ER': '196.200.96.0/20',
5608 'ES': '88.0.0.0/11',
5609 'ET': '196.188.0.0/14',
5610 'EU': '2.16.0.0/13',
5611 'FI': '91.152.0.0/13',
5612 'FJ': '144.120.0.0/16',
5613 'FK': '80.73.208.0/21',
5614 'FM': '119.252.112.0/20',
5615 'FO': '88.85.32.0/19',
5617 'GA': '41.158.0.0/15',
5619 'GD': '74.122.88.0/21',
5620 'GE': '31.146.0.0/16',
5621 'GF': '161.22.64.0/18',
5622 'GG': '62.68.160.0/19',
5623 'GH': '154.160.0.0/12',
5624 'GI': '95.164.0.0/16',
5625 'GL': '88.83.0.0/19',
5626 'GM': '160.182.0.0/15',
5627 'GN': '197.149.192.0/18',
5628 'GP': '104.250.0.0/19',
5629 'GQ': '105.235.224.0/20',
5630 'GR': '94.64.0.0/13',
5631 'GT': '168.234.0.0/16',
5632 'GU': '168.123.0.0/16',
5633 'GW': '197.214.80.0/20',
5634 'GY': '181.41.64.0/18',
5635 'HK': '113.252.0.0/14',
5636 'HN': '181.210.0.0/16',
5637 'HR': '93.136.0.0/13',
5638 'HT': '148.102.128.0/17',
5639 'HU': '84.0.0.0/14',
5640 'ID': '39.192.0.0/10',
5641 'IE': '87.32.0.0/12',
5642 'IL': '79.176.0.0/13',
5643 'IM': '5.62.80.0/20',
5644 'IN': '117.192.0.0/10',
5645 'IO': '203.83.48.0/21',
5646 'IQ': '37.236.0.0/14',
5647 'IR': '2.176.0.0/12',
5648 'IS': '82.221.0.0/16',
5649 'IT': '79.0.0.0/10',
5650 'JE': '87.244.64.0/18',
5651 'JM': '72.27.0.0/17',
5652 'JO': '176.29.0.0/16',
5653 'JP': '133.0.0.0/8',
5654 'KE': '105.48.0.0/12',
5655 'KG': '158.181.128.0/17',
5656 'KH': '36.37.128.0/17',
5657 'KI': '103.25.140.0/22',
5658 'KM': '197.255.224.0/20',
5659 'KN': '198.167.192.0/19',
5660 'KP': '175.45.176.0/22',
5661 'KR': '175.192.0.0/10',
5662 'KW': '37.36.0.0/14',
5663 'KY': '64.96.0.0/15',
5664 'KZ': '2.72.0.0/13',
5665 'LA': '115.84.64.0/18',
5666 'LB': '178.135.0.0/16',
5667 'LC': '24.92.144.0/20',
5668 'LI': '82.117.0.0/19',
5669 'LK': '112.134.0.0/15',
5670 'LR': '102.183.0.0/16',
5671 'LS': '129.232.0.0/17',
5672 'LT': '78.56.0.0/13',
5673 'LU': '188.42.0.0/16',
5674 'LV': '46.109.0.0/16',
5675 'LY': '41.252.0.0/14',
5676 'MA': '105.128.0.0/11',
5677 'MC': '88.209.64.0/18',
5678 'MD': '37.246.0.0/16',
5679 'ME': '178.175.0.0/17',
5680 'MF': '74.112.232.0/21',
5681 'MG': '154.126.0.0/17',
5682 'MH': '117.103.88.0/21',
5683 'MK': '77.28.0.0/15',
5684 'ML': '154.118.128.0/18',
5685 'MM': '37.111.0.0/17',
5686 'MN': '49.0.128.0/17',
5687 'MO': '60.246.0.0/16',
5688 'MP': '202.88.64.0/20',
5689 'MQ': '109.203.224.0/19',
5690 'MR': '41.188.64.0/18',
5691 'MS': '208.90.112.0/22',
5692 'MT': '46.11.0.0/16',
5693 'MU': '105.16.0.0/12',
5694 'MV': '27.114.128.0/18',
5695 'MW': '102.70.0.0/15',
5696 'MX': '187.192.0.0/11',
5697 'MY': '175.136.0.0/13',
5698 'MZ': '197.218.0.0/15',
5699 'NA': '41.182.0.0/16',
5700 'NC': '101.101.0.0/18',
5701 'NE': '197.214.0.0/18',
5702 'NF': '203.17.240.0/22',
5703 'NG': '105.112.0.0/12',
5704 'NI': '186.76.0.0/15',
5705 'NL': '145.96.0.0/11',
5706 'NO': '84.208.0.0/13',
5707 'NP': '36.252.0.0/15',
5708 'NR': '203.98.224.0/19',
5709 'NU': '49.156.48.0/22',
5710 'NZ': '49.224.0.0/14',
5711 'OM': '5.36.0.0/15',
5712 'PA': '186.72.0.0/15',
5713 'PE': '186.160.0.0/14',
5714 'PF': '123.50.64.0/18',
5715 'PG': '124.240.192.0/19',
5716 'PH': '49.144.0.0/13',
5717 'PK': '39.32.0.0/11',
5718 'PL': '83.0.0.0/11',
5719 'PM': '70.36.0.0/20',
5720 'PR': '66.50.0.0/16',
5721 'PS': '188.161.0.0/16',
5722 'PT': '85.240.0.0/13',
5723 'PW': '202.124.224.0/20',
5724 'PY': '181.120.0.0/14',
5725 'QA': '37.210.0.0/15',
5726 'RE': '102.35.0.0/16',
5727 'RO': '79.112.0.0/13',
5728 'RS': '93.86.0.0/15',
5729 'RU': '5.136.0.0/13',
5730 'RW': '41.186.0.0/16',
5731 'SA': '188.48.0.0/13',
5732 'SB': '202.1.160.0/19',
5733 'SC': '154.192.0.0/11',
5734 'SD': '102.120.0.0/13',
5735 'SE': '78.64.0.0/12',
5736 'SG': '8.128.0.0/10',
5737 'SI': '188.196.0.0/14',
5738 'SK': '78.98.0.0/15',
5739 'SL': '102.143.0.0/17',
5740 'SM': '89.186.32.0/19',
5741 'SN': '41.82.0.0/15',
5742 'SO': '154.115.192.0/18',
5743 'SR': '186.179.128.0/17',
5744 'SS': '105.235.208.0/21',
5745 'ST': '197.159.160.0/19',
5746 'SV': '168.243.0.0/16',
5747 'SX': '190.102.0.0/20',
5749 'SZ': '41.84.224.0/19',
5750 'TC': '65.255.48.0/20',
5751 'TD': '154.68.128.0/19',
5752 'TG': '196.168.0.0/14',
5753 'TH': '171.96.0.0/13',
5754 'TJ': '85.9.128.0/18',
5755 'TK': '27.96.24.0/21',
5756 'TL': '180.189.160.0/20',
5757 'TM': '95.85.96.0/19',
5758 'TN': '197.0.0.0/11',
5759 'TO': '175.176.144.0/21',
5760 'TR': '78.160.0.0/11',
5761 'TT': '186.44.0.0/15',
5762 'TV': '202.2.96.0/19',
5763 'TW': '120.96.0.0/11',
5764 'TZ': '156.156.0.0/14',
5765 'UA': '37.52.0.0/14',
5766 'UG': '102.80.0.0/13',
5768 'UY': '167.56.0.0/13',
5769 'UZ': '84.54.64.0/18',
5770 'VA': '212.77.0.0/19',
5771 'VC': '207.191.240.0/21',
5772 'VE': '186.88.0.0/13',
5773 'VG': '66.81.192.0/20',
5774 'VI': '146.226.0.0/16',
5775 'VN': '14.160.0.0/11',
5776 'VU': '202.80.32.0/20',
5777 'WF': '117.20.32.0/21',
5778 'WS': '202.4.32.0/19',
5779 'YE': '134.35.0.0/16',
5780 'YT': '41.242.116.0/22',
5781 'ZA': '41.0.0.0/11',
5782 'ZM': '102.144.0.0/13',
5783 'ZW': '102.177.192.0/18',
5787 def random_ipv4(cls, code_or_block):
5788 if len(code_or_block) == 2:
5789 block = cls._country_ip_map.get(code_or_block.upper())
5793 block = code_or_block
5794 addr, preflen = block.split('/')
5795 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5796 addr_max = addr_min | (0xffffffff >> int(preflen))
5797 return compat_str(socket.inet_ntoa(
5798 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5801 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5802 def __init__(self, proxies=None):
5803 # Set default handlers
5804 for type in ('http', 'https'):
5805 setattr(self, '%s_open' % type,
5806 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5807 meth(r, proxy, type))
5808 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5810 def proxy_open(self, req, proxy, type):
5811 req_proxy = req.headers.get('Ytdl-request-proxy')
5812 if req_proxy is not None:
5814 del req.headers['Ytdl-request-proxy']
5816 if proxy == '__noproxy__':
5817 return None # No Proxy
5818 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5819 req.add_header('Ytdl-socks-proxy', proxy)
5820 # yt-dlp's http/https handlers do wrapping the socket with socks
5822 return compat_urllib_request.ProxyHandler.proxy_open(
5823 self, req, proxy, type)
5826 # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5827 # released into Public Domain
5828 # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5830 def long_to_bytes(n, blocksize=0):
5831 """long_to_bytes(n:long, blocksize:int) : string
5832 Convert a long integer to a byte string.
5834 If optional blocksize is given and greater than zero, pad the front of the
5835 byte string with binary zeros so that the length is a multiple of
5838 # after much testing, this algorithm was deemed to be the fastest
5842 s = compat_struct_pack('>I', n & 0xffffffff) + s
5844 # strip off leading zeros
5845 for i in range(len(s)):
5846 if s[i] != b'\000'[0]:
5849 # only happens when n == 0
5853 # add back some pad bytes. this could be done more efficiently w.r.t. the
5854 # de-padding being done above, but sigh...
5855 if blocksize > 0 and len(s) % blocksize:
5856 s = (blocksize - len(s) % blocksize) * b'\000' + s
5860 def bytes_to_long(s):
5861 """bytes_to_long(string) : long
5862 Convert a byte string to a long integer.
5864 This is (essentially) the inverse of long_to_bytes().
5869 extra = (4 - length % 4)
5870 s = b'\000' * extra + s
5871 length = length + extra
5872 for i in range(0, length, 4):
5873 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5877 def ohdave_rsa_encrypt(data, exponent, modulus):
5879 Implement OHDave
's RSA algorithm. See http://www.ohdave.com/rsa/
5882 data: data to encrypt, bytes-like object
5883 exponent, modulus: parameter e and N of RSA algorithm, both integer
5884 Output: hex string of encrypted data
5886 Limitation: supports one block encryption only
5889 payload = int(binascii.hexlify(data[::-1]), 16)
5890 encrypted = pow(payload, exponent, modulus)
5891 return '%x' % encrypted
5894 def pkcs1pad(data, length):
5896 Padding input data with PKCS#1 scheme
5898 @param {int[]} data input data
5899 @param {int} length target length
5900 @returns {int[]} padded data
5902 if len(data) > length - 11:
5903 raise ValueError('Input data too
long for PKCS
#1 padding')
5905 pseudo_random
= [random
.randint(0, 254) for _
in range(length
- len(data
) - 3)]
5906 return [0, 2] + pseudo_random
+ [0] + data
5909 def encode_base_n(num
, n
, table
=None):
5910 FULL_TABLE
= '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5912 table
= FULL_TABLE
[:n
]
5915 raise ValueError('base %d exceeds table length %d' % (n
, len(table
)))
5922 ret
= table
[num
% n
] + ret
5927 def decode_packed_codes(code
):
5928 mobj
= re
.search(PACKED_CODES_RE
, code
)
5929 obfuscated_code
, base
, count
, symbols
= mobj
.groups()
5932 symbols
= symbols
.split('|')
5937 base_n_count
= encode_base_n(count
, base
)
5938 symbol_table
[base_n_count
] = symbols
[count
] or base_n_count
5941 r
'\b(\w+)\b', lambda mobj
: symbol_table
[mobj
.group(0)],
5945 def caesar(s
, alphabet
, shift
):
5950 alphabet
[(alphabet
.index(c
) + shift
) % l
] if c
in alphabet
else c
5955 return caesar(s
, r
'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5958 def parse_m3u8_attributes(attrib
):
5960 for (key
, val
) in re
.findall(r
'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib
):
5961 if val
.startswith('"'):
5967 def urshift(val
, n
):
5968 return val
>> n
if val
>= 0 else (val
+ 0x100000000) >> n
5971 # Based on png2str() written by @gdkchan and improved by @yokrysty
5972 # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5973 def decode_png(png_data
):
5974 # Reference: https://www.w3.org/TR/PNG/
5975 header
= png_data
[8:]
5977 if png_data
[:8] != b
'\x89PNG\x0d\x0a\x1a\x0a' or header
[4:8] != b
'IHDR':
5978 raise IOError('Not a valid PNG file.')
5980 int_map
= {1: '>B', 2: '>H', 4: '>I'}
5981 unpack_integer
= lambda x
: compat_struct_unpack(int_map
[len(x
)], x
)[0]
5986 length
= unpack_integer(header
[:4])
5989 chunk_type
= header
[:4]
5992 chunk_data
= header
[:length
]
5993 header
= header
[length
:]
5995 header
= header
[4:] # Skip CRC
6003 ihdr
= chunks
[0]['data']
6005 width
= unpack_integer(ihdr
[:4])
6006 height
= unpack_integer(ihdr
[4:8])
6010 for chunk
in chunks
:
6011 if chunk
['type'] == b
'IDAT':
6012 idat
+= chunk
['data']
6015 raise IOError('Unable to read PNG data.')
6017 decompressed_data
= bytearray(zlib
.decompress(idat
))
6022 def _get_pixel(idx
):
6027 for y
in range(height
):
6028 basePos
= y
* (1 + stride
)
6029 filter_type
= decompressed_data
[basePos
]
6033 pixels
.append(current_row
)
6035 for x
in range(stride
):
6036 color
= decompressed_data
[1 + basePos
+ x
]
6037 basex
= y
* stride
+ x
6042 left
= _get_pixel(basex
- 3)
6044 up
= _get_pixel(basex
- stride
)
6046 if filter_type
== 1: # Sub
6047 color
= (color
+ left
) & 0xff
6048 elif filter_type
== 2: # Up
6049 color
= (color
+ up
) & 0xff
6050 elif filter_type
== 3: # Average
6051 color
= (color
+ ((left
+ up
) >> 1)) & 0xff
6052 elif filter_type
== 4: # Paeth
6058 c
= _get_pixel(basex
- stride
- 3)
6066 if pa
<= pb
and pa
<= pc
:
6067 color
= (color
+ a
) & 0xff
6069 color
= (color
+ b
) & 0xff
6071 color
= (color
+ c
) & 0xff
6073 current_row
.append(color
)
6075 return width
, height
, pixels
6078 def write_xattr(path
, key
, value
):
6079 # This mess below finds the best xattr tool for the job
6081 # try the pyxattr module...
6084 if hasattr(xattr
, 'set'): # pyxattr
6085 # Unicode arguments are not supported in python-pyxattr until
6087 # See https://github.com/ytdl-org/youtube-dl/issues/5498
6088 pyxattr_required_version
= '0.5.0'
6089 if version_tuple(xattr
.__version
__) < version_tuple(pyxattr_required_version
):
6090 # TODO: fallback to CLI tools
6091 raise XAttrUnavailableError(
6092 'python-pyxattr is detected but is too old. '
6093 'yt-dlp requires %s or above while your version is %s. '
6094 'Falling back to other xattr implementations' % (
6095 pyxattr_required_version
, xattr
.__version
__))
6097 setxattr
= xattr
.set
6099 setxattr
= xattr
.setxattr
6102 setxattr(path
, key
, value
)
6103 except EnvironmentError as e
:
6104 raise XAttrMetadataError(e
.errno
, e
.strerror
)
6107 if compat_os_name
== 'nt':
6108 # Write xattrs to NTFS Alternate Data Streams:
6109 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
6110 assert ':' not in key
6111 assert os
.path
.exists(path
)
6113 ads_fn
= path
+ ':' + key
6115 with open(ads_fn
, 'wb') as f
:
6117 except EnvironmentError as e
:
6118 raise XAttrMetadataError(e
.errno
, e
.strerror
)
6120 user_has_setfattr
= check_executable('setfattr', ['--version'])
6121 user_has_xattr
= check_executable('xattr', ['-h'])
6123 if user_has_setfattr
or user_has_xattr
:
6125 value
= value
.decode('utf-8')
6126 if user_has_setfattr
:
6127 executable
= 'setfattr'
6128 opts
= ['-n', key
, '-v', value
]
6129 elif user_has_xattr
:
6130 executable
= 'xattr'
6131 opts
= ['-w', key
, value
]
6133 cmd
= ([encodeFilename(executable
, True)]
6134 + [encodeArgument(o
) for o
in opts
]
6135 + [encodeFilename(path
, True)])
6138 p
= subprocess
.Popen(
6139 cmd
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
, stdin
=subprocess
.PIPE
)
6140 except EnvironmentError as e
:
6141 raise XAttrMetadataError(e
.errno
, e
.strerror
)
6142 stdout
, stderr
= process_communicate_or_kill(p
)
6143 stderr
= stderr
.decode('utf-8', 'replace')
6144 if p
.returncode
!= 0:
6145 raise XAttrMetadataError(p
.returncode
, stderr
)
6148 # On Unix, and can't find pyxattr, setfattr, or xattr.
6149 if sys
.platform
.startswith('linux'):
6150 raise XAttrUnavailableError(
6151 "Couldn't find a tool to set the xattrs. "
6152 "Install either the python 'pyxattr' or 'xattr' "
6153 "modules, or the GNU 'attr' package "
6154 "(which contains the 'setfattr' tool).")
6156 raise XAttrUnavailableError(
6157 "Couldn't find a tool to set the xattrs. "
6158 "Install either the python 'xattr' module, "
6159 "or the 'xattr' binary.")
6162 def random_birthday(year_field
, month_field
, day_field
):
6163 start_date
= datetime
.date(1950, 1, 1)
6164 end_date
= datetime
.date(1995, 12, 31)
6165 offset
= random
.randint(0, (end_date
- start_date
).days
)
6166 random_date
= start_date
+ datetime
.timedelta(offset
)
6168 year_field
: str(random_date
.year
),
6169 month_field
: str(random_date
.month
),
6170 day_field
: str(random_date
.day
),
6174 # Templates for internet shortcut files, which are plain text files.
6175 DOT_URL_LINK_TEMPLATE
= '''
6180 DOT_WEBLOC_LINK_TEMPLATE
= '''
6181 <?xml version="1.0" encoding="UTF-8"?>
6182 <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
6183 <plist version="1.0">
6186 \t<string>%(url)s</string>
6191 DOT_DESKTOP_LINK_TEMPLATE
= '''
6201 def iri_to_uri(iri
):
6203 Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
6205 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.
6208 iri_parts
= compat_urllib_parse_urlparse(iri
)
6210 if '[' in iri_parts
.netloc
:
6211 raise ValueError('IPv6 URIs are not, yet, supported.')
6212 # Querying `.netloc`, when there's only one bracket, also raises a ValueError.
6214 # 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.
6217 if iri_parts
.username
:
6218 net_location
+= compat_urllib_parse_quote(iri_parts
.username
, safe
=r
"!$%&'()*+,~")
6219 if iri_parts
.password
is not None:
6220 net_location
+= ':' + compat_urllib_parse_quote(iri_parts
.password
, safe
=r
"!$%&'()*+,~")
6223 net_location
+= iri_parts
.hostname
.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
6224 # The 'idna' encoding produces ASCII text.
6225 if iri_parts
.port
is not None and iri_parts
.port
!= 80:
6226 net_location
+= ':' + str(iri_parts
.port
)
6228 return compat_urllib_parse_urlunparse(
6232 compat_urllib_parse_quote_plus(iri_parts
.path
, safe
=r
"!$%&'()*+,/:;=@|~"),
6234 # Unsure about the `safe` argument, since this is a legacy way of handling parameters.
6235 compat_urllib_parse_quote_plus(iri_parts
.params
, safe
=r
"!$%&'()*+,/:;=@|~"),
6237 # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
6238 compat_urllib_parse_quote_plus(iri_parts
.query
, safe
=r
"!$%&'()*+,/:;=?@{|}~"),
6240 compat_urllib_parse_quote_plus(iri_parts
.fragment
, safe
=r
"!#$%&'()*+,/:;=?@{|}~")))
6242 # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
6245 def to_high_limit_path(path
):
6246 if sys
.platform
in ['win32', 'cygwin']:
6247 # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
6248 return r
'\\?\ '.rstrip() + os
.path
.abspath(path
)
6253 def format_field(obj
, field
=None, template
='%s', ignore
=(None, ''), default
='', func
=None):
6255 val
= obj
if obj
is not None else default
6257 val
= obj
.get(field
, default
)
6258 if func
and val
not in ignore
:
6260 return template
% val
if val
not in ignore
else default
6263 def clean_podcast_url(url
):
6264 return re
.sub(r
'''(?x)
6268 media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
6271 (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
6274 cn\.co| # https://podcorn.com/analytics-prefix/
6275 st\.fm # https://podsights.com/docs/
6280 _HEX_TABLE
= '0123456789abcdef'
6283 def random_uuidv4():
6284 return re
.sub(r
'[xy]', lambda x
: _HEX_TABLE
[random
.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
6287 def make_dir(path
, to_screen
=None):
6289 dn
= os
.path
.dirname(path
)
6290 if dn
and not os
.path
.exists(dn
):
6293 except (OSError, IOError) as err
:
6294 if callable(to_screen
) is not None:
6295 to_screen('unable to create directory ' + error_to_compat_str(err
))
6299 def get_executable_path():
6300 from zipimport
import zipimporter
6301 if hasattr(sys
, 'frozen'): # Running from PyInstaller
6302 path
= os
.path
.dirname(sys
.executable
)
6303 elif isinstance(globals().get('__loader__'), zipimporter
): # Running from ZIP
6304 path
= os
.path
.join(os
.path
.dirname(__file__
), '../..')
6306 path
= os
.path
.join(os
.path
.dirname(__file__
), '..')
6307 return os
.path
.abspath(path
)
6310 def load_plugins(name
, suffix
, namespace
):
6311 plugin_info
= [None]
6314 plugin_info
= imp
.find_module(
6315 name
, [os
.path
.join(get_executable_path(), 'ytdlp_plugins')])
6316 plugins
= imp
.load_module(name
, *plugin_info
)
6317 for name
in dir(plugins
):
6318 if name
in namespace
:
6320 if not name
.endswith(suffix
):
6322 klass
= getattr(plugins
, name
)
6323 classes
[name
] = namespace
[name
] = klass
6327 if plugin_info
[0] is not None:
6328 plugin_info
[0].close()
6333 obj
, *path_list
, default
=None, expected_type
=None, get_all
=True,
6334 casesense
=True, is_user_input
=False, traverse_string
=False):
6335 ''' Traverse nested list/dict/tuple
6336 @param path_list A list of paths which are checked one by one.
6337 Each path is a list of keys where each key is a string,
6338 a function, a tuple of strings or "...".
6339 When a fuction is given, it takes the key as argument and
6340 returns whether the key matches or not. When a tuple is given,
6341 all the keys given in the tuple are traversed, and
6342 "..." traverses all the keys in the object
6343 @param default Default value to return
6344 @param expected_type Only accept final value of this type (Can also be any callable)
6345 @param get_all Return all the values obtained from a path or only the first one
6346 @param casesense Whether to consider dictionary keys as case sensitive
6347 @param is_user_input Whether the keys are generated from user input. If True,
6348 strings are converted to int/slice if necessary
6349 @param traverse_string Whether to traverse inside strings. If True, any
6350 non-compatible object will also be converted into a string
6354 _lower
= lambda k
: (k
.lower() if isinstance(k
, str) else k
)
6355 path_list
= (map(_lower
, variadic(path
)) for path
in path_list
)
6357 def _traverse_obj(obj
, path
, _current_depth
=0):
6361 path
= tuple(variadic(path
))
6362 for i
, key
in enumerate(path
):
6363 if isinstance(key
, (list, tuple)):
6364 obj
= [_traverse_obj(obj
, sub_key
, _current_depth
) for sub_key
in key
]
6367 obj
= (obj
.values() if isinstance(obj
, dict)
6368 else obj
if isinstance(obj
, (list, tuple, LazyList
))
6369 else str(obj
) if traverse_string
else [])
6371 depth
= max(depth
, _current_depth
)
6372 return [_traverse_obj(inner_obj
, path
[i
+ 1:], _current_depth
) for inner_obj
in obj
]
6374 if isinstance(obj
, (list, tuple, LazyList
)):
6375 obj
= enumerate(obj
)
6376 elif isinstance(obj
, dict):
6379 if not traverse_string
:
6383 depth
= max(depth
, _current_depth
)
6384 return [_traverse_obj(v
, path
[i
+ 1:], _current_depth
) for k
, v
in obj
if key(k
)]
6385 elif isinstance(obj
, dict) and not (is_user_input
and key
== ':'):
6386 obj
= (obj
.get(key
) if casesense
or (key
in obj
)
6387 else next((v
for k
, v
in obj
.items() if _lower(k
) == key
), None))
6390 key
= (int_or_none(key
) if ':' not in key
6391 else slice(*map(int_or_none
, key
.split(':'))))
6392 if key
== slice(None):
6393 return _traverse_obj(obj
, (..., *path
[i
+ 1:]), _current_depth
)
6394 if not isinstance(key
, (int, slice)):
6396 if not isinstance(obj
, (list, tuple, LazyList
)):
6397 if not traverse_string
:
6406 if isinstance(expected_type
, type):
6407 type_test
= lambda val
: val
if isinstance(val
, expected_type
) else None
6408 elif expected_type
is not None:
6409 type_test
= expected_type
6411 type_test
= lambda val
: val
6413 for path
in path_list
:
6415 val
= _traverse_obj(obj
, path
)
6418 for _
in range(depth
- 1):
6419 val
= itertools
.chain
.from_iterable(v
for v
in val
if v
is not None)
6420 val
= [v
for v
in map(type_test
, val
) if v
is not None]
6422 return val
if get_all
else val
[0]
6424 val
= type_test(val
)
6430 def traverse_dict(dictn
, keys
, casesense
=True):
6431 ''' For backward compatibility. Do not use '''
6432 return traverse_obj(dictn
, keys
, casesense
=casesense
,
6433 is_user_input
=True, traverse_string
=True)
6436 def variadic(x
, allowed_types
=(str, bytes)):
6437 return x
if isinstance(x
, collections
.abc
.Iterable
) and not isinstance(x
, allowed_types
) else (x
,)
6440 # create a JSON Web Signature (jws) with HS256 algorithm
6441 # the resulting format is in JWS Compact Serialization
6442 # implemented following JWT https://www.rfc-editor.org/rfc/rfc7519.html
6443 # implemented following JWS https://www.rfc-editor.org/rfc/rfc7515.html
6444 def jwt_encode_hs256(payload_data
, key
, headers
={}):
6450 header_data
.update(headers
)
6451 header_b64
= base64
.b64encode(json
.dumps(header_data
).encode('utf-8'))
6452 payload_b64
= base64
.b64encode(json
.dumps(payload_data
).encode('utf-8'))
6453 h
= hmac
.new(key
.encode('utf-8'), header_b64
+ b
'.' + payload_b64
, hashlib
.sha256
)
6454 signature_b64
= base64
.b64encode(h
.digest())
6455 token
= header_b64
+ b
'.' + payload_b64
+ b
'.' + signature_b64
6459 def supports_terminal_sequences(stream
):
6460 if compat_os_name
== 'nt':
6461 if get_windows_version() < (10, ):
6463 elif not os
.getenv('TERM'):
6466 return stream
.isatty()
6467 except BaseException
:
6471 TERMINAL_SEQUENCES
= {
6474 'ERASE_LINE': '\x1b[K',
6475 'RED': '\033[0;31m',
6476 'YELLOW': '\033[0;33m',
6477 'BLUE': '\033[0;34m',
6478 'RESET_STYLE': '\033[0m',