4 from __future__
import unicode_literals
37 import xml
.etree
.ElementTree
41 compat_HTMLParseError
,
46 compat_ctypes_WINFUNCTYPE
,
47 compat_etree_fromstring
,
50 compat_html_entities_html5
,
63 compat_urllib_parse_urlencode
,
64 compat_urllib_parse_urlparse
,
65 compat_urllib_parse_urlunparse
,
66 compat_urllib_parse_quote
,
67 compat_urllib_parse_quote_plus
,
68 compat_urllib_parse_unquote_plus
,
69 compat_urllib_request
,
80 def register_socks_protocols():
81 # "Register" SOCKS protocols
82 # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
83 # URLs with protocols not in urlparse.uses_netloc are not handled correctly
84 for scheme
in ('socks', 'socks4', 'socks4a', 'socks5'):
85 if scheme
not in compat_urlparse
.uses_netloc
:
86 compat_urlparse
.uses_netloc
.append(scheme
)
89 # This is not clearly defined otherwise
90 compiled_regex_type
= type(re
.compile(''))
93 def random_user_agent():
94 _USER_AGENT_TPL
= 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
1673 return _USER_AGENT_TPL
% random
.choice(_CHROME_VERSIONS
)
1677 'User-Agent': random_user_agent(),
1678 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
1679 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
1680 'Accept-Encoding': 'gzip, deflate',
1681 'Accept-Language': 'en-us,en;q=0.5',
1686 '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',
1690 NO_DEFAULT
= object()
1692 ENGLISH_MONTH_NAMES
= [
1693 'January', 'February', 'March', 'April', 'May', 'June',
1694 'July', 'August', 'September', 'October', 'November', 'December']
1697 'en': ENGLISH_MONTH_NAMES
,
1699 'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
1700 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
1703 KNOWN_EXTENSIONS
= (
1704 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
1705 'flv', 'f4v', 'f4a', 'f4b',
1706 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
1707 'mkv', 'mka', 'mk3d',
1710 'asf', 'wmv', 'wma',
1716 'f4f', 'f4m', 'm3u8', 'smil')
1718 REMUX_EXTENSIONS
= ('mp4', 'mkv', 'flv', 'webm', 'mov', 'avi', 'mp3', 'mka', 'm4a', 'ogg', 'opus')
1720 # needed for sanitizing filenames in restricted mode
1721 ACCENT_CHARS
= dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
1722 itertools
.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1723 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1746 '%Y/%m/%d %H:%M:%S',
1748 '%Y-%m-%d %H:%M:%S',
1749 '%Y-%m-%d %H:%M:%S.%f',
1752 '%Y-%m-%dT%H:%M:%SZ',
1753 '%Y-%m-%dT%H:%M:%S.%fZ',
1754 '%Y-%m-%dT%H:%M:%S.%f0Z',
1755 '%Y-%m-%dT%H:%M:%S',
1756 '%Y-%m-%dT%H:%M:%S.%f',
1758 '%b %d %Y at %H:%M',
1759 '%b %d %Y at %H:%M:%S',
1760 '%B %d %Y at %H:%M',
1761 '%B %d %Y at %H:%M:%S',
1764 DATE_FORMATS_DAY_FIRST
= list(DATE_FORMATS
)
1765 DATE_FORMATS_DAY_FIRST
.extend([
1771 '%d/%m/%Y %H:%M:%S',
1774 DATE_FORMATS_MONTH_FIRST
= list(DATE_FORMATS
)
1775 DATE_FORMATS_MONTH_FIRST
.extend([
1780 '%m/%d/%Y %H:%M:%S',
1783 PACKED_CODES_RE
= r
"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1784 JSON_LD_RE
= r
'(?is)<script[^>]+type=(["\']?
)application
/ld\
+json\
1[^
>]*>(?P
<json_ld
>.+?
)</script
>'
1787 def preferredencoding():
1788 """Get preferred encoding.
1790 Returns the best encoding scheme for the system, based on
1791 locale.getpreferredencoding() and some further tweaks.
1794 pref = locale.getpreferredencoding()
1802 def write_json_file(obj, fn):
1803 """ Encode obj as JSON and write it to fn, atomically if possible """
1805 fn = encodeFilename(fn)
1806 if sys.version_info < (3, 0) and sys.platform != 'win32
':
1807 encoding = get_filesystem_encoding()
1808 # os.path.basename returns a bytes object, but NamedTemporaryFile
1809 # will fail if the filename contains non ascii characters unless we
1810 # use a unicode object
1811 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1812 # the same for os.path.dirname
1813 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1815 path_basename = os.path.basename
1816 path_dirname = os.path.dirname
1820 'prefix
': path_basename(fn) + '.',
1821 'dir': path_dirname(fn),
1825 # In Python 2.x, json.dump expects a bytestream.
1826 # In Python 3.x, it writes to a character stream
1827 if sys.version_info < (3, 0):
1832 'encoding
': 'utf
-8',
1835 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1840 if sys.platform == 'win32
':
1841 # Need to remove existing file on Windows, else os.rename raises
1842 # WindowsError or FileExistsError.
1850 os.chmod(tf.name, 0o666 & ~mask)
1853 os.rename(tf.name, fn)
1862 if sys.version_info >= (2, 7):
1863 def find_xpath_attr(node, xpath, key, val=None):
1864 """ Find the xpath xpath[@key=val] """
1865 assert re.match(r'^
[a
-zA
-Z_
-]+$
', key)
1866 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1867 return node.find(expr)
1869 def find_xpath_attr(node, xpath, key, val=None):
1870 for f in node.findall(compat_xpath(xpath)):
1871 if key not in f.attrib:
1873 if val is None or f.attrib.get(key) == val:
1877 # On python2.6 the xml.etree.ElementTree.Element methods don't support
1878 # the namespace parameter
1881 def xpath_with_ns(path
, ns_map
):
1882 components
= [c
.split(':') for c
in path
.split('/')]
1884 for c
in components
:
1886 replaced
.append(c
[0])
1889 replaced
.append('{%s}%s' % (ns_map
[ns
], tag
))
1890 return '/'.join(replaced
)
1893 def xpath_element(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1894 def _find_xpath(xpath
):
1895 return node
.find(compat_xpath(xpath
))
1897 if isinstance(xpath
, (str, compat_str
)):
1898 n
= _find_xpath(xpath
)
1906 if default
is not NO_DEFAULT
:
1909 name
= xpath
if name
is None else name
1910 raise ExtractorError('Could not find XML element %s' % name
)
1916 def xpath_text(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1917 n
= xpath_element(node
, xpath
, name
, fatal
=fatal
, default
=default
)
1918 if n
is None or n
== default
:
1921 if default
is not NO_DEFAULT
:
1924 name
= xpath
if name
is None else name
1925 raise ExtractorError('Could not find XML element\'s text %s' % name
)
1931 def xpath_attr(node
, xpath
, key
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1932 n
= find_xpath_attr(node
, xpath
, key
)
1934 if default
is not NO_DEFAULT
:
1937 name
= '%s[@%s]' % (xpath
, key
) if name
is None else name
1938 raise ExtractorError('Could not find XML attribute %s' % name
)
1941 return n
.attrib
[key
]
1944 def get_element_by_id(id, html
):
1945 """Return the content of the tag with the specified ID in the passed HTML document"""
1946 return get_element_by_attribute('id', id, html
)
1949 def get_element_by_class(class_name
, html
):
1950 """Return the content of the first tag with the specified class in the passed HTML document"""
1951 retval
= get_elements_by_class(class_name
, html
)
1952 return retval
[0] if retval
else None
1955 def get_element_by_attribute(attribute
, value
, html
, escape_value
=True):
1956 retval
= get_elements_by_attribute(attribute
, value
, html
, escape_value
)
1957 return retval
[0] if retval
else None
1960 def get_elements_by_class(class_name
, html
):
1961 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1962 return get_elements_by_attribute(
1963 'class', r
'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1964 html, escape_value=False)
1967 def get_elements_by_attribute(attribute, value, html, escape_value=True):
1968 """Return the content of the tag with the specified attribute in the passed HTML document"""
1970 value = re.escape(value) if escape_value else value
1973 for m in re.finditer(r'''(?xs)
1975 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^
']*'|
))*?
1977 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^
"]*"|
='[^']*'|))*?
1981 ''' % (re.escape(attribute), value), html):
1982 res = m.group('content
')
1984 if res.startswith('"') or res.startswith("'"):
1987 retlist.append(unescapeHTML(res))
1992 class HTMLAttributeParser(compat_HTMLParser):
1993 """Trivial HTML parser to gather the attributes for a single element"""
1997 compat_HTMLParser.__init__(self)
1999 def handle_starttag(self, tag, attrs):
2000 self.attrs = dict(attrs)
2003 def extract_attributes(html_element):
2004 """Given a string for an HTML element such as
2006 a="foo" B="bar" c="&98;az" d=boz
2007 empty= noval entity="&"
2010 Decode and return a dictionary of attributes.
2012 'a
': 'foo
', 'b
': 'bar
', c: 'baz
', d: 'boz
',
2013 'empty
': '', 'noval
': None, 'entity
': '&',
2014 'sq
': '"', 'dq': '\''
2016 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2017 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2019 parser = HTMLAttributeParser()
2021 parser.feed(html_element)
2023 # Older Python may throw HTMLParseError in case of malformed HTML
2024 except compat_HTMLParseError:
2029 def clean_html(html):
2030 """Clean an HTML snippet into a readable string"""
2032 if html is None: # Convenience for sanitizing descriptions etc.
2036 html = html.replace('\n', ' ')
2037 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2038 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2040 html = re.sub('<.*?>', '', html)
2041 # Replace html entities
2042 html = unescapeHTML(html)
2046 def sanitize_open(filename, open_mode):
2047 """Try to open the given filename, and slightly tweak it if this fails.
2049 Attempts to open the given filename. If this fails, it tries to change
2050 the filename slightly, step by step, until it's either able to open it
2051 or it fails and raises a final exception, like the standard open()
2054 It returns the tuple (stream, definitive_file_name).
2058 if sys.platform == 'win32':
2060 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2061 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2062 stream = open(encodeFilename(filename), open_mode)
2063 return (stream, filename)
2064 except (IOError, OSError) as err:
2065 if err.errno in (errno.EACCES,):
2068 # In case of error, try to remove win32 forbidden chars
2069 alt_filename = sanitize_path(filename)
2070 if alt_filename == filename:
2073 # An exception here should be caught in the caller
2074 stream = open(encodeFilename(alt_filename), open_mode)
2075 return (stream, alt_filename)
2078 def timeconvert(timestr):
2079 """Convert RFC 2822 defined time string into system timestamp"""
2081 timetuple = email.utils.parsedate_tz(timestr)
2082 if timetuple is not None:
2083 timestamp = email.utils.mktime_tz(timetuple)
2087 def sanitize_filename(s, restricted=False, is_id=False):
2088 """Sanitizes a string so it could be used as part of a filename.
2089 If restricted is set, use a stricter subset of allowed characters.
2090 Set is_id if this is not an arbitrary string, but an ID that should be kept
2093 def replace_insane(char):
2094 if restricted and char in ACCENT_CHARS:
2095 return ACCENT_CHARS[char]
2096 if char == '?' or ord(char) < 32 or ord(char) == 127:
2099 return '' if restricted else '\''
2101 return '_
-' if restricted else ' -'
2102 elif char in '\\/|
*<>':
2104 if restricted and (char in '!&\'()[]{}$
;`^
,#' or char.isspace()):
2106 if restricted
and ord(char
) > 127:
2111 s
= re
.sub(r
'[0-9]+(?::[0-9]+)+', lambda m
: m
.group(0).replace(':', '_'), s
)
2112 result
= ''.join(map(replace_insane
, s
))
2114 while '__' in result
:
2115 result
= result
.replace('__', '_')
2116 result
= result
.strip('_')
2117 # Common case of "Foreign band name - English song title"
2118 if restricted
and result
.startswith('-_'):
2120 if result
.startswith('-'):
2121 result
= '_' + result
[len('-'):]
2122 result
= result
.lstrip('.')
2128 def sanitize_path(s
):
2129 """Sanitizes and normalizes path on Windows"""
2130 if sys
.platform
!= 'win32':
2132 drive_or_unc
, _
= os
.path
.splitdrive(s
)
2133 if sys
.version_info
< (2, 7) and not drive_or_unc
:
2134 drive_or_unc
, _
= os
.path
.splitunc(s
)
2135 norm_path
= os
.path
.normpath(remove_start(s
, drive_or_unc
)).split(os
.path
.sep
)
2139 path_part
if path_part
in ['.', '..'] else re
.sub(r
'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part
)
2140 for path_part
in norm_path
]
2142 sanitized_path
.insert(0, drive_or_unc
+ os
.path
.sep
)
2143 return os
.path
.join(*sanitized_path
)
2146 def sanitize_url(url
):
2147 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2148 # the number of unwanted failures due to missing protocol
2149 if url
.startswith('//'):
2150 return 'http:%s' % url
2151 # Fix some common typos seen so far
2153 # https://github.com/ytdl-org/youtube-dl/issues/15649
2154 (r
'^httpss://', r
'https://'),
2155 # https://bx1.be/lives/direct-tv/
2156 (r
'^rmtp([es]?)://', r
'rtmp\1://'),
2158 for mistake
, fixup
in COMMON_TYPOS
:
2159 if re
.match(mistake
, url
):
2160 return re
.sub(mistake
, fixup
, url
)
2164 def sanitized_Request(url
, *args
, **kwargs
):
2165 return compat_urllib_request
.Request(sanitize_url(url
), *args
, **kwargs
)
2169 """Expand shell variables and ~"""
2170 return os
.path
.expandvars(compat_expanduser(s
))
2173 def orderedSet(iterable
):
2174 """ Remove all duplicates from the input iterable """
2182 def _htmlentity_transform(entity_with_semicolon
):
2183 """Transforms an HTML entity to a character."""
2184 entity
= entity_with_semicolon
[:-1]
2186 # Known non-numeric HTML entity
2187 if entity
in compat_html_entities
.name2codepoint
:
2188 return compat_chr(compat_html_entities
.name2codepoint
[entity
])
2190 # TODO: HTML5 allows entities without a semicolon. For example,
2191 # 'Éric' should be decoded as 'Éric'.
2192 if entity_with_semicolon
in compat_html_entities_html5
:
2193 return compat_html_entities_html5
[entity_with_semicolon
]
2195 mobj
= re
.match(r
'#(x[0-9a-fA-F]+|[0-9]+)', entity
)
2196 if mobj
is not None:
2197 numstr
= mobj
.group(1)
2198 if numstr
.startswith('x'):
2200 numstr
= '0%s' % numstr
2203 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2205 return compat_chr(int(numstr
, base
))
2209 # Unknown entity in name, return its literal representation
2210 return '&%s;' % entity
2213 def unescapeHTML(s
):
2216 assert type(s
) == compat_str
2219 r
'&([^&;]+;)', lambda m
: _htmlentity_transform(m
.group(1)), s
)
2222 def process_communicate_or_kill(p
, *args
, **kwargs
):
2224 return p
.communicate(*args
, **kwargs
)
2225 except BaseException
: # Including KeyboardInterrupt
2231 def get_subprocess_encoding():
2232 if sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2233 # For subprocess calls, encode with locale encoding
2234 # Refer to http://stackoverflow.com/a/9951851/35070
2235 encoding
= preferredencoding()
2237 encoding
= sys
.getfilesystemencoding()
2238 if encoding
is None:
2243 def encodeFilename(s
, for_subprocess
=False):
2245 @param s The name of the file
2248 assert type(s
) == compat_str
2250 # Python 3 has a Unicode API
2251 if sys
.version_info
>= (3, 0):
2254 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2255 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2256 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2257 if not for_subprocess
and sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2260 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2261 if sys
.platform
.startswith('java'):
2264 return s
.encode(get_subprocess_encoding(), 'ignore')
2267 def decodeFilename(b
, for_subprocess
=False):
2269 if sys
.version_info
>= (3, 0):
2272 if not isinstance(b
, bytes):
2275 return b
.decode(get_subprocess_encoding(), 'ignore')
2278 def encodeArgument(s
):
2279 if not isinstance(s
, compat_str
):
2280 # Legacy code that uses byte strings
2281 # Uncomment the following line after fixing all post processors
2282 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2283 s
= s
.decode('ascii')
2284 return encodeFilename(s
, True)
2287 def decodeArgument(b
):
2288 return decodeFilename(b
, True)
2291 def decodeOption(optval
):
2294 if isinstance(optval
, bytes):
2295 optval
= optval
.decode(preferredencoding())
2297 assert isinstance(optval
, compat_str
)
2301 def formatSeconds(secs
, delim
=':'):
2303 return '%d%s%02d%s%02d' % (secs
// 3600, delim
, (secs
% 3600) // 60, delim
, secs
% 60)
2305 return '%d%s%02d' % (secs
// 60, delim
, secs
% 60)
2310 def make_HTTPS_handler(params
, **kwargs
):
2311 opts_no_check_certificate
= params
.get('nocheckcertificate', False)
2312 if hasattr(ssl
, 'create_default_context'): # Python >= 3.4 or 2.7.9
2313 context
= ssl
.create_default_context(ssl
.Purpose
.SERVER_AUTH
)
2314 if opts_no_check_certificate
:
2315 context
.check_hostname
= False
2316 context
.verify_mode
= ssl
.CERT_NONE
2318 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2321 # (create_default_context present but HTTPSHandler has no context=)
2324 if sys
.version_info
< (3, 2):
2325 return YoutubeDLHTTPSHandler(params
, **kwargs
)
2326 else: # Python < 3.4
2327 context
= ssl
.SSLContext(ssl
.PROTOCOL_TLSv1
)
2328 context
.verify_mode
= (ssl
.CERT_NONE
2329 if opts_no_check_certificate
2330 else ssl
.CERT_REQUIRED
)
2331 context
.set_default_verify_paths()
2332 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2335 def bug_reports_message():
2336 if ytdl_is_updateable():
2337 update_cmd
= 'type youtube-dlc -U to update'
2339 update_cmd
= 'see https://github.com/pukkandan/yt-dlp on how to update'
2340 msg
= '; please report this issue on https://github.com/pukkandan/yt-dlp .'
2341 msg
+= ' Make sure you are using the latest version; %s.' % update_cmd
2342 msg
+= ' Be sure to call youtube-dlc with the --verbose flag and include its complete output.'
2346 class YoutubeDLError(Exception):
2347 """Base exception for YoutubeDL errors."""
2351 class ExtractorError(YoutubeDLError
):
2352 """Error during info extraction."""
2354 def __init__(self
, msg
, tb
=None, expected
=False, cause
=None, video_id
=None):
2355 """ tb, if given, is the original traceback (so that it can be printed out).
2356 If expected is set, this is a normal error message and most likely not a bug in youtube-dlc.
2359 if sys
.exc_info()[0] in (compat_urllib_error
.URLError
, socket
.timeout
, UnavailableVideoError
):
2361 if video_id
is not None:
2362 msg
= video_id
+ ': ' + msg
2364 msg
+= ' (caused by %r)' % cause
2366 msg
+= bug_reports_message()
2367 super(ExtractorError
, self
).__init
__(msg
)
2370 self
.exc_info
= sys
.exc_info() # preserve original exception
2372 self
.video_id
= video_id
2374 def format_traceback(self
):
2375 if self
.traceback
is None:
2377 return ''.join(traceback
.format_tb(self
.traceback
))
2380 class UnsupportedError(ExtractorError
):
2381 def __init__(self
, url
):
2382 super(UnsupportedError
, self
).__init
__(
2383 'Unsupported URL: %s' % url
, expected
=True)
2387 class RegexNotFoundError(ExtractorError
):
2388 """Error when a regex didn't match"""
2392 class GeoRestrictedError(ExtractorError
):
2393 """Geographic restriction Error exception.
2395 This exception may be thrown when a video is not available from your
2396 geographic location due to geographic restrictions imposed by a website.
2399 def __init__(self
, msg
, countries
=None):
2400 super(GeoRestrictedError
, self
).__init
__(msg
, expected
=True)
2402 self
.countries
= countries
2405 class DownloadError(YoutubeDLError
):
2406 """Download Error exception.
2408 This exception may be thrown by FileDownloader objects if they are not
2409 configured to continue on errors. They will contain the appropriate
2413 def __init__(self
, msg
, exc_info
=None):
2414 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2415 super(DownloadError
, self
).__init
__(msg
)
2416 self
.exc_info
= exc_info
2419 class SameFileError(YoutubeDLError
):
2420 """Same File exception.
2422 This exception will be thrown by FileDownloader objects if they detect
2423 multiple files would have to be downloaded to the same file on disk.
2428 class PostProcessingError(YoutubeDLError
):
2429 """Post Processing exception.
2431 This exception may be raised by PostProcessor's .run() method to
2432 indicate an error in the postprocessing task.
2435 def __init__(self
, msg
):
2436 super(PostProcessingError
, self
).__init
__(msg
)
2440 class ExistingVideoReached(YoutubeDLError
):
2441 """ --max-downloads limit has been reached. """
2445 class RejectedVideoReached(YoutubeDLError
):
2446 """ --max-downloads limit has been reached. """
2450 class MaxDownloadsReached(YoutubeDLError
):
2451 """ --max-downloads limit has been reached. """
2455 class UnavailableVideoError(YoutubeDLError
):
2456 """Unavailable Format exception.
2458 This exception will be thrown when a video is requested
2459 in a format that is not available for that video.
2464 class ContentTooShortError(YoutubeDLError
):
2465 """Content Too Short exception.
2467 This exception may be raised by FileDownloader objects when a file they
2468 download is too small for what the server announced first, indicating
2469 the connection was probably interrupted.
2472 def __init__(self
, downloaded
, expected
):
2473 super(ContentTooShortError
, self
).__init
__(
2474 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded
, expected
)
2477 self
.downloaded
= downloaded
2478 self
.expected
= expected
2481 class XAttrMetadataError(YoutubeDLError
):
2482 def __init__(self
, code
=None, msg
='Unknown error'):
2483 super(XAttrMetadataError
, self
).__init
__(msg
)
2487 # Parsing code and msg
2488 if (self
.code
in (errno
.ENOSPC
, errno
.EDQUOT
)
2489 or 'No space left' in self
.msg
or 'Disk quota exceeded' in self
.msg
):
2490 self
.reason
= 'NO_SPACE'
2491 elif self
.code
== errno
.E2BIG
or 'Argument list too long' in self
.msg
:
2492 self
.reason
= 'VALUE_TOO_LONG'
2494 self
.reason
= 'NOT_SUPPORTED'
2497 class XAttrUnavailableError(YoutubeDLError
):
2501 def _create_http_connection(ydl_handler
, http_class
, is_https
, *args
, **kwargs
):
2502 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2503 # expected HTTP responses to meet HTTP/1.0 or later (see also
2504 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2505 if sys
.version_info
< (3, 0):
2506 kwargs
['strict'] = True
2507 hc
= http_class(*args
, **compat_kwargs(kwargs
))
2508 source_address
= ydl_handler
._params
.get('source_address')
2510 if source_address
is not None:
2511 # This is to workaround _create_connection() from socket where it will try all
2512 # address data from getaddrinfo() including IPv6. This filters the result from
2513 # getaddrinfo() based on the source_address value.
2514 # This is based on the cpython socket.create_connection() function.
2515 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2516 def _create_connection(address
, timeout
=socket
._GLOBAL
_DEFAULT
_TIMEOUT
, source_address
=None):
2517 host
, port
= address
2519 addrs
= socket
.getaddrinfo(host
, port
, 0, socket
.SOCK_STREAM
)
2520 af
= socket
.AF_INET
if '.' in source_address
[0] else socket
.AF_INET6
2521 ip_addrs
= [addr
for addr
in addrs
if addr
[0] == af
]
2522 if addrs
and not ip_addrs
:
2523 ip_version
= 'v4' if af
== socket
.AF_INET
else 'v6'
2525 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2526 % (ip_version
, source_address
[0]))
2527 for res
in ip_addrs
:
2528 af
, socktype
, proto
, canonname
, sa
= res
2531 sock
= socket
.socket(af
, socktype
, proto
)
2532 if timeout
is not socket
._GLOBAL
_DEFAULT
_TIMEOUT
:
2533 sock
.settimeout(timeout
)
2534 sock
.bind(source_address
)
2536 err
= None # Explicitly break reference cycle
2538 except socket
.error
as _
:
2540 if sock
is not None:
2545 raise socket
.error('getaddrinfo returns an empty list')
2546 if hasattr(hc
, '_create_connection'):
2547 hc
._create
_connection
= _create_connection
2548 sa
= (source_address
, 0)
2549 if hasattr(hc
, 'source_address'): # Python 2.7+
2550 hc
.source_address
= sa
2552 def _hc_connect(self
, *args
, **kwargs
):
2553 sock
= _create_connection(
2554 (self
.host
, self
.port
), self
.timeout
, sa
)
2556 self
.sock
= ssl
.wrap_socket(
2557 sock
, self
.key_file
, self
.cert_file
,
2558 ssl_version
=ssl
.PROTOCOL_TLSv1
)
2561 hc
.connect
= functools
.partial(_hc_connect
, hc
)
2566 def handle_youtubedl_headers(headers
):
2567 filtered_headers
= headers
2569 if 'Youtubedl-no-compression' in filtered_headers
:
2570 filtered_headers
= dict((k
, v
) for k
, v
in filtered_headers
.items() if k
.lower() != 'accept-encoding')
2571 del filtered_headers
['Youtubedl-no-compression']
2573 return filtered_headers
2576 class YoutubeDLHandler(compat_urllib_request
.HTTPHandler
):
2577 """Handler for HTTP requests and responses.
2579 This class, when installed with an OpenerDirector, automatically adds
2580 the standard headers to every HTTP request and handles gzipped and
2581 deflated responses from web servers. If compression is to be avoided in
2582 a particular request, the original request in the program code only has
2583 to include the HTTP header "Youtubedl-no-compression", which will be
2584 removed before making the real request.
2586 Part of this code was copied from:
2588 http://techknack.net/python-urllib2-handlers/
2590 Andrew Rowls, the author of that code, agreed to release it to the
2594 def __init__(self
, params
, *args
, **kwargs
):
2595 compat_urllib_request
.HTTPHandler
.__init
__(self
, *args
, **kwargs
)
2596 self
._params
= params
2598 def http_open(self
, req
):
2599 conn_class
= compat_http_client
.HTTPConnection
2601 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2603 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2604 del req
.headers
['Ytdl-socks-proxy']
2606 return self
.do_open(functools
.partial(
2607 _create_http_connection
, self
, conn_class
, False),
2615 return zlib
.decompress(data
, -zlib
.MAX_WBITS
)
2617 return zlib
.decompress(data
)
2619 def http_request(self
, req
):
2620 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2621 # always respected by websites, some tend to give out URLs with non percent-encoded
2622 # non-ASCII characters (see telemb.py, ard.py [#3412])
2623 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2624 # To work around aforementioned issue we will replace request's original URL with
2625 # percent-encoded one
2626 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2627 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2628 url
= req
.get_full_url()
2629 url_escaped
= escape_url(url
)
2631 # Substitute URL if any change after escaping
2632 if url
!= url_escaped
:
2633 req
= update_Request(req
, url
=url_escaped
)
2635 for h
, v
in std_headers
.items():
2636 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2637 # The dict keys are capitalized because of this bug by urllib
2638 if h
.capitalize() not in req
.headers
:
2639 req
.add_header(h
, v
)
2641 req
.headers
= handle_youtubedl_headers(req
.headers
)
2643 if sys
.version_info
< (2, 7) and '#' in req
.get_full_url():
2644 # Python 2.6 is brain-dead when it comes to fragments
2645 req
._Request
__original
= req
._Request
__original
.partition('#')[0]
2646 req
._Request
__r
_type
= req
._Request
__r
_type
.partition('#')[0]
2650 def http_response(self
, req
, resp
):
2653 if resp
.headers
.get('Content-encoding', '') == 'gzip':
2654 content
= resp
.read()
2655 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
), mode
='rb')
2657 uncompressed
= io
.BytesIO(gz
.read())
2658 except IOError as original_ioerror
:
2659 # There may be junk add the end of the file
2660 # See http://stackoverflow.com/q/4928560/35070 for details
2661 for i
in range(1, 1024):
2663 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
[:-i
]), mode
='rb')
2664 uncompressed
= io
.BytesIO(gz
.read())
2669 raise original_ioerror
2670 resp
= compat_urllib_request
.addinfourl(uncompressed
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2671 resp
.msg
= old_resp
.msg
2672 del resp
.headers
['Content-encoding']
2674 if resp
.headers
.get('Content-encoding', '') == 'deflate':
2675 gz
= io
.BytesIO(self
.deflate(resp
.read()))
2676 resp
= compat_urllib_request
.addinfourl(gz
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2677 resp
.msg
= old_resp
.msg
2678 del resp
.headers
['Content-encoding']
2679 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2680 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2681 if 300 <= resp
.code
< 400:
2682 location
= resp
.headers
.get('Location')
2684 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2685 if sys
.version_info
>= (3, 0):
2686 location
= location
.encode('iso-8859-1').decode('utf-8')
2688 location
= location
.decode('utf-8')
2689 location_escaped
= escape_url(location
)
2690 if location
!= location_escaped
:
2691 del resp
.headers
['Location']
2692 if sys
.version_info
< (3, 0):
2693 location_escaped
= location_escaped
.encode('utf-8')
2694 resp
.headers
['Location'] = location_escaped
2697 https_request
= http_request
2698 https_response
= http_response
2701 def make_socks_conn_class(base_class
, socks_proxy
):
2702 assert issubclass(base_class
, (
2703 compat_http_client
.HTTPConnection
, compat_http_client
.HTTPSConnection
))
2705 url_components
= compat_urlparse
.urlparse(socks_proxy
)
2706 if url_components
.scheme
.lower() == 'socks5':
2707 socks_type
= ProxyType
.SOCKS5
2708 elif url_components
.scheme
.lower() in ('socks', 'socks4'):
2709 socks_type
= ProxyType
.SOCKS4
2710 elif url_components
.scheme
.lower() == 'socks4a':
2711 socks_type
= ProxyType
.SOCKS4A
2713 def unquote_if_non_empty(s
):
2716 return compat_urllib_parse_unquote_plus(s
)
2720 url_components
.hostname
, url_components
.port
or 1080,
2722 unquote_if_non_empty(url_components
.username
),
2723 unquote_if_non_empty(url_components
.password
),
2726 class SocksConnection(base_class
):
2728 self
.sock
= sockssocket()
2729 self
.sock
.setproxy(*proxy_args
)
2730 if type(self
.timeout
) in (int, float):
2731 self
.sock
.settimeout(self
.timeout
)
2732 self
.sock
.connect((self
.host
, self
.port
))
2734 if isinstance(self
, compat_http_client
.HTTPSConnection
):
2735 if hasattr(self
, '_context'): # Python > 2.6
2736 self
.sock
= self
._context
.wrap_socket(
2737 self
.sock
, server_hostname
=self
.host
)
2739 self
.sock
= ssl
.wrap_socket(self
.sock
)
2741 return SocksConnection
2744 class YoutubeDLHTTPSHandler(compat_urllib_request
.HTTPSHandler
):
2745 def __init__(self
, params
, https_conn_class
=None, *args
, **kwargs
):
2746 compat_urllib_request
.HTTPSHandler
.__init
__(self
, *args
, **kwargs
)
2747 self
._https
_conn
_class
= https_conn_class
or compat_http_client
.HTTPSConnection
2748 self
._params
= params
2750 def https_open(self
, req
):
2752 conn_class
= self
._https
_conn
_class
2754 if hasattr(self
, '_context'): # python > 2.6
2755 kwargs
['context'] = self
._context
2756 if hasattr(self
, '_check_hostname'): # python 3.x
2757 kwargs
['check_hostname'] = self
._check
_hostname
2759 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2761 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2762 del req
.headers
['Ytdl-socks-proxy']
2764 return self
.do_open(functools
.partial(
2765 _create_http_connection
, self
, conn_class
, True),
2769 class YoutubeDLCookieJar(compat_cookiejar
.MozillaCookieJar
):
2771 See [1] for cookie file format.
2773 1. https://curl.haxx.se/docs/http-cookies.html
2775 _HTTPONLY_PREFIX
= '#HttpOnly_'
2777 _HEADER
= '''# Netscape HTTP Cookie File
2778 # This file is generated by youtube-dlc. Do not edit.
2781 _CookieFileEntry
= collections
.namedtuple(
2783 ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
2785 def save(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2787 Save cookies to a file.
2789 Most of the code is taken from CPython 3.8 and slightly adapted
2790 to support cookie files with UTF-8 in both python 2 and 3.
2792 if filename
is None:
2793 if self
.filename
is not None:
2794 filename
= self
.filename
2796 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2798 # Store session cookies with `expires` set to 0 instead of an empty
2801 if cookie
.expires
is None:
2804 with io
.open(filename
, 'w', encoding
='utf-8') as f
:
2805 f
.write(self
._HEADER
)
2808 if not ignore_discard
and cookie
.discard
:
2810 if not ignore_expires
and cookie
.is_expired(now
):
2816 if cookie
.domain
.startswith('.'):
2817 initial_dot
= 'TRUE'
2819 initial_dot
= 'FALSE'
2820 if cookie
.expires
is not None:
2821 expires
= compat_str(cookie
.expires
)
2824 if cookie
.value
is None:
2825 # cookies.txt regards 'Set-Cookie: foo' as a cookie
2826 # with no name, whereas http.cookiejar regards it as a
2827 # cookie with no value.
2832 value
= cookie
.value
2834 '\t'.join([cookie
.domain
, initial_dot
, cookie
.path
,
2835 secure
, expires
, name
, value
]) + '\n')
2837 def load(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2838 """Load cookies from a file."""
2839 if filename
is None:
2840 if self
.filename
is not None:
2841 filename
= self
.filename
2843 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2845 def prepare_line(line
):
2846 if line
.startswith(self
._HTTPONLY
_PREFIX
):
2847 line
= line
[len(self
._HTTPONLY
_PREFIX
):]
2848 # comments and empty lines are fine
2849 if line
.startswith('#') or not line
.strip():
2851 cookie_list
= line
.split('\t')
2852 if len(cookie_list
) != self
._ENTRY
_LEN
:
2853 raise compat_cookiejar
.LoadError('invalid length %d' % len(cookie_list
))
2854 cookie
= self
._CookieFileEntry
(*cookie_list
)
2855 if cookie
.expires_at
and not cookie
.expires_at
.isdigit():
2856 raise compat_cookiejar
.LoadError('invalid expires at %s' % cookie
.expires_at
)
2860 with io
.open(filename
, encoding
='utf-8') as f
:
2863 cf
.write(prepare_line(line
))
2864 except compat_cookiejar
.LoadError
as e
:
2866 'WARNING: skipping cookie file entry due to %s: %r\n'
2867 % (e
, line
), sys
.stderr
)
2870 self
._really
_load
(cf
, filename
, ignore_discard
, ignore_expires
)
2871 # Session cookies are denoted by either `expires` field set to
2872 # an empty string or 0. MozillaCookieJar only recognizes the former
2873 # (see [1]). So we need force the latter to be recognized as session
2874 # cookies on our own.
2875 # Session cookies may be important for cookies-based authentication,
2876 # e.g. usually, when user does not check 'Remember me' check box while
2877 # logging in on a site, some important cookies are stored as session
2878 # cookies so that not recognizing them will result in failed login.
2879 # 1. https://bugs.python.org/issue17164
2881 # Treat `expires=0` cookies as session cookies
2882 if cookie
.expires
== 0:
2883 cookie
.expires
= None
2884 cookie
.discard
= True
2887 class YoutubeDLCookieProcessor(compat_urllib_request
.HTTPCookieProcessor
):
2888 def __init__(self
, cookiejar
=None):
2889 compat_urllib_request
.HTTPCookieProcessor
.__init
__(self
, cookiejar
)
2891 def http_response(self
, request
, response
):
2892 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2893 # characters in Set-Cookie HTTP header of last response (see
2894 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2895 # In order to at least prevent crashing we will percent encode Set-Cookie
2896 # header before HTTPCookieProcessor starts processing it.
2897 # if sys.version_info < (3, 0) and response.headers:
2898 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2899 # set_cookie = response.headers.get(set_cookie_header)
2901 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2902 # if set_cookie != set_cookie_escaped:
2903 # del response.headers[set_cookie_header]
2904 # response.headers[set_cookie_header] = set_cookie_escaped
2905 return compat_urllib_request
.HTTPCookieProcessor
.http_response(self
, request
, response
)
2907 https_request
= compat_urllib_request
.HTTPCookieProcessor
.http_request
2908 https_response
= http_response
2911 class YoutubeDLRedirectHandler(compat_urllib_request
.HTTPRedirectHandler
):
2912 if sys
.version_info
[0] < 3:
2913 def redirect_request(self
, req
, fp
, code
, msg
, headers
, newurl
):
2914 # On python 2 urlh.geturl() may sometimes return redirect URL
2915 # as byte string instead of unicode. This workaround allows
2916 # to force it always return unicode.
2917 return compat_urllib_request
.HTTPRedirectHandler
.redirect_request(self
, req
, fp
, code
, msg
, headers
, compat_str(newurl
))
2920 def extract_timezone(date_str
):
2922 r
'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
2925 timezone
= datetime
.timedelta()
2927 date_str
= date_str
[:-len(m
.group('tz'))]
2928 if not m
.group('sign'):
2929 timezone
= datetime
.timedelta()
2931 sign
= 1 if m
.group('sign') == '+' else -1
2932 timezone
= datetime
.timedelta(
2933 hours
=sign
* int(m
.group('hours')),
2934 minutes
=sign
* int(m
.group('minutes')))
2935 return timezone
, date_str
2938 def parse_iso8601(date_str
, delimiter
='T', timezone
=None):
2939 """ Return a UNIX timestamp from the given date """
2941 if date_str
is None:
2944 date_str
= re
.sub(r
'\.[0-9]+', '', date_str
)
2946 if timezone
is None:
2947 timezone
, date_str
= extract_timezone(date_str
)
2950 date_format
= '%Y-%m-%d{0}%H:%M:%S'.format(delimiter
)
2951 dt
= datetime
.datetime
.strptime(date_str
, date_format
) - timezone
2952 return calendar
.timegm(dt
.timetuple())
2957 def date_formats(day_first
=True):
2958 return DATE_FORMATS_DAY_FIRST
if day_first
else DATE_FORMATS_MONTH_FIRST
2961 def unified_strdate(date_str
, day_first
=True):
2962 """Return a string with the date in the format YYYYMMDD"""
2964 if date_str
is None:
2968 date_str
= date_str
.replace(',', ' ')
2969 # Remove AM/PM + timezone
2970 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
2971 _
, date_str
= extract_timezone(date_str
)
2973 for expression
in date_formats(day_first
):
2975 upload_date
= datetime
.datetime
.strptime(date_str
, expression
).strftime('%Y%m%d')
2978 if upload_date
is None:
2979 timetuple
= email
.utils
.parsedate_tz(date_str
)
2982 upload_date
= datetime
.datetime(*timetuple
[:6]).strftime('%Y%m%d')
2985 if upload_date
is not None:
2986 return compat_str(upload_date
)
2989 def unified_timestamp(date_str
, day_first
=True):
2990 if date_str
is None:
2993 date_str
= re
.sub(r
'[,|]', '', date_str
)
2995 pm_delta
= 12 if re
.search(r
'(?i)PM', date_str
) else 0
2996 timezone
, date_str
= extract_timezone(date_str
)
2998 # Remove AM/PM + timezone
2999 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
3001 # Remove unrecognized timezones from ISO 8601 alike timestamps
3002 m
= re
.search(r
'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str
)
3004 date_str
= date_str
[:-len(m
.group('tz'))]
3006 # Python only supports microseconds, so remove nanoseconds
3007 m
= re
.search(r
'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str
)
3009 date_str
= m
.group(1)
3011 for expression
in date_formats(day_first
):
3013 dt
= datetime
.datetime
.strptime(date_str
, expression
) - timezone
+ datetime
.timedelta(hours
=pm_delta
)
3014 return calendar
.timegm(dt
.timetuple())
3017 timetuple
= email
.utils
.parsedate_tz(date_str
)
3019 return calendar
.timegm(timetuple
) + pm_delta
* 3600
3022 def determine_ext(url
, default_ext
='unknown_video'):
3023 if url
is None or '.' not in url
:
3025 guess
= url
.partition('?')[0].rpartition('.')[2]
3026 if re
.match(r
'^[A-Za-z0-9]+$', guess
):
3028 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
3029 elif guess
.rstrip('/') in KNOWN_EXTENSIONS
:
3030 return guess
.rstrip('/')
3035 def subtitles_filename(filename
, sub_lang
, sub_format
, expected_real_ext
=None):
3036 return replace_extension(filename
, sub_lang
+ '.' + sub_format
, expected_real_ext
)
3039 def date_from_str(date_str
):
3041 Return a datetime object from a string in the format YYYYMMDD or
3042 (now|today)[+-][0-9](day|week|month|year)(s)?"""
3043 today
= datetime
.date
.today()
3044 if date_str
in ('now', 'today'):
3046 if date_str
== 'yesterday':
3047 return today
- datetime
.timedelta(days
=1)
3048 match
= re
.match(r
'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str
)
3049 if match
is not None:
3050 sign
= match
.group('sign')
3051 time
= int(match
.group('time'))
3054 unit
= match
.group('unit')
3055 # A bad approximation?
3059 elif unit
== 'year':
3063 delta
= datetime
.timedelta(**{unit: time}
)
3064 return today
+ delta
3065 return datetime
.datetime
.strptime(date_str
, '%Y%m%d').date()
3068 def hyphenate_date(date_str
):
3070 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
3071 match
= re
.match(r
'^(\d\d\d\d)(\d\d)(\d\d)$', date_str
)
3072 if match
is not None:
3073 return '-'.join(match
.groups())
3078 class DateRange(object):
3079 """Represents a time interval between two dates"""
3081 def __init__(self
, start
=None, end
=None):
3082 """start and end must be strings in the format accepted by date"""
3083 if start
is not None:
3084 self
.start
= date_from_str(start
)
3086 self
.start
= datetime
.datetime
.min.date()
3088 self
.end
= date_from_str(end
)
3090 self
.end
= datetime
.datetime
.max.date()
3091 if self
.start
> self
.end
:
3092 raise ValueError('Date range: "%s" , the start date must be before the end date' % self
)
3096 """Returns a range that only contains the given day"""
3097 return cls(day
, day
)
3099 def __contains__(self
, date
):
3100 """Check if the date is in the range"""
3101 if not isinstance(date
, datetime
.date
):
3102 date
= date_from_str(date
)
3103 return self
.start
<= date
<= self
.end
3106 return '%s - %s' % (self
.start
.isoformat(), self
.end
.isoformat())
3109 def platform_name():
3110 """ Returns the platform name as a compat_str """
3111 res
= platform
.platform()
3112 if isinstance(res
, bytes):
3113 res
= res
.decode(preferredencoding())
3115 assert isinstance(res
, compat_str
)
3119 def _windows_write_string(s
, out
):
3120 """ Returns True if the string was written using special methods,
3121 False if it has yet to be written out."""
3122 # Adapted from http://stackoverflow.com/a/3259271/35070
3125 import ctypes
.wintypes
3133 fileno
= out
.fileno()
3134 except AttributeError:
3135 # If the output stream doesn't have a fileno, it's virtual
3137 except io
.UnsupportedOperation
:
3138 # Some strange Windows pseudo files?
3140 if fileno
not in WIN_OUTPUT_IDS
:
3143 GetStdHandle
= compat_ctypes_WINFUNCTYPE(
3144 ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.DWORD
)(
3145 ('GetStdHandle', ctypes
.windll
.kernel32
))
3146 h
= GetStdHandle(WIN_OUTPUT_IDS
[fileno
])
3148 WriteConsoleW
= compat_ctypes_WINFUNCTYPE(
3149 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.LPWSTR
,
3150 ctypes
.wintypes
.DWORD
, ctypes
.POINTER(ctypes
.wintypes
.DWORD
),
3151 ctypes
.wintypes
.LPVOID
)(('WriteConsoleW', ctypes
.windll
.kernel32
))
3152 written
= ctypes
.wintypes
.DWORD(0)
3154 GetFileType
= compat_ctypes_WINFUNCTYPE(ctypes
.wintypes
.DWORD
, ctypes
.wintypes
.DWORD
)(('GetFileType', ctypes
.windll
.kernel32
))
3155 FILE_TYPE_CHAR
= 0x0002
3156 FILE_TYPE_REMOTE
= 0x8000
3157 GetConsoleMode
= compat_ctypes_WINFUNCTYPE(
3158 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
,
3159 ctypes
.POINTER(ctypes
.wintypes
.DWORD
))(
3160 ('GetConsoleMode', ctypes
.windll
.kernel32
))
3161 INVALID_HANDLE_VALUE
= ctypes
.wintypes
.DWORD(-1).value
3163 def not_a_console(handle
):
3164 if handle
== INVALID_HANDLE_VALUE
or handle
is None:
3166 return ((GetFileType(handle
) & ~FILE_TYPE_REMOTE
) != FILE_TYPE_CHAR
3167 or GetConsoleMode(handle
, ctypes
.byref(ctypes
.wintypes
.DWORD())) == 0)
3169 if not_a_console(h
):
3172 def next_nonbmp_pos(s
):
3174 return next(i
for i
, c
in enumerate(s
) if ord(c
) > 0xffff)
3175 except StopIteration:
3179 count
= min(next_nonbmp_pos(s
), 1024)
3181 ret
= WriteConsoleW(
3182 h
, s
, count
if count
else 2, ctypes
.byref(written
), None)
3184 raise OSError('Failed to write string')
3185 if not count
: # We just wrote a non-BMP character
3186 assert written
.value
== 2
3189 assert written
.value
> 0
3190 s
= s
[written
.value
:]
3194 def write_string(s
, out
=None, encoding
=None):
3197 assert type(s
) == compat_str
3199 if sys
.platform
== 'win32' and encoding
is None and hasattr(out
, 'fileno'):
3200 if _windows_write_string(s
, out
):
3203 if ('b' in getattr(out
, 'mode', '')
3204 or sys
.version_info
[0] < 3): # Python 2 lies about mode of sys.stderr
3205 byt
= s
.encode(encoding
or preferredencoding(), 'ignore')
3207 elif hasattr(out
, 'buffer'):
3208 enc
= encoding
or getattr(out
, 'encoding', None) or preferredencoding()
3209 byt
= s
.encode(enc
, 'ignore')
3210 out
.buffer.write(byt
)
3216 def bytes_to_intlist(bs
):
3219 if isinstance(bs
[0], int): # Python 3
3222 return [ord(c
) for c
in bs
]
3225 def intlist_to_bytes(xs
):
3228 return compat_struct_pack('%dB' % len(xs
), *xs
)
3231 # Cross-platform file locking
3232 if sys
.platform
== 'win32':
3233 import ctypes
.wintypes
3236 class OVERLAPPED(ctypes
.Structure
):
3238 ('Internal', ctypes
.wintypes
.LPVOID
),
3239 ('InternalHigh', ctypes
.wintypes
.LPVOID
),
3240 ('Offset', ctypes
.wintypes
.DWORD
),
3241 ('OffsetHigh', ctypes
.wintypes
.DWORD
),
3242 ('hEvent', ctypes
.wintypes
.HANDLE
),
3245 kernel32
= ctypes
.windll
.kernel32
3246 LockFileEx
= kernel32
.LockFileEx
3247 LockFileEx
.argtypes
= [
3248 ctypes
.wintypes
.HANDLE
, # hFile
3249 ctypes
.wintypes
.DWORD
, # dwFlags
3250 ctypes
.wintypes
.DWORD
, # dwReserved
3251 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3252 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3253 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3255 LockFileEx
.restype
= ctypes
.wintypes
.BOOL
3256 UnlockFileEx
= kernel32
.UnlockFileEx
3257 UnlockFileEx
.argtypes
= [
3258 ctypes
.wintypes
.HANDLE
, # hFile
3259 ctypes
.wintypes
.DWORD
, # dwReserved
3260 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3261 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3262 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3264 UnlockFileEx
.restype
= ctypes
.wintypes
.BOOL
3265 whole_low
= 0xffffffff
3266 whole_high
= 0x7fffffff
3268 def _lock_file(f
, exclusive
):
3269 overlapped
= OVERLAPPED()
3270 overlapped
.Offset
= 0
3271 overlapped
.OffsetHigh
= 0
3272 overlapped
.hEvent
= 0
3273 f
._lock
_file
_overlapped
_p
= ctypes
.pointer(overlapped
)
3274 handle
= msvcrt
.get_osfhandle(f
.fileno())
3275 if not LockFileEx(handle
, 0x2 if exclusive
else 0x0, 0,
3276 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3277 raise OSError('Locking file failed: %r' % ctypes
.FormatError())
3279 def _unlock_file(f
):
3280 assert f
._lock
_file
_overlapped
_p
3281 handle
= msvcrt
.get_osfhandle(f
.fileno())
3282 if not UnlockFileEx(handle
, 0,
3283 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3284 raise OSError('Unlocking file failed: %r' % ctypes
.FormatError())
3287 # Some platforms, such as Jython, is missing fcntl
3291 def _lock_file(f
, exclusive
):
3292 fcntl
.flock(f
, fcntl
.LOCK_EX
if exclusive
else fcntl
.LOCK_SH
)
3294 def _unlock_file(f
):
3295 fcntl
.flock(f
, fcntl
.LOCK_UN
)
3297 UNSUPPORTED_MSG
= 'file locking is not supported on this platform'
3299 def _lock_file(f
, exclusive
):
3300 raise IOError(UNSUPPORTED_MSG
)
3302 def _unlock_file(f
):
3303 raise IOError(UNSUPPORTED_MSG
)
3306 class locked_file(object):
3307 def __init__(self
, filename
, mode
, encoding
=None):
3308 assert mode
in ['r', 'a', 'w']
3309 self
.f
= io
.open(filename
, mode
, encoding
=encoding
)
3312 def __enter__(self
):
3313 exclusive
= self
.mode
!= 'r'
3315 _lock_file(self
.f
, exclusive
)
3321 def __exit__(self
, etype
, value
, traceback
):
3323 _unlock_file(self
.f
)
3330 def write(self
, *args
):
3331 return self
.f
.write(*args
)
3333 def read(self
, *args
):
3334 return self
.f
.read(*args
)
3337 def get_filesystem_encoding():
3338 encoding
= sys
.getfilesystemencoding()
3339 return encoding
if encoding
is not None else 'utf-8'
3342 def shell_quote(args
):
3344 encoding
= get_filesystem_encoding()
3346 if isinstance(a
, bytes):
3347 # We may get a filename encoded with 'encodeFilename'
3348 a
= a
.decode(encoding
)
3349 quoted_args
.append(compat_shlex_quote(a
))
3350 return ' '.join(quoted_args
)
3353 def smuggle_url(url
, data
):
3354 """ Pass additional data in a URL for internal use. """
3356 url
, idata
= unsmuggle_url(url
, {})
3358 sdata
= compat_urllib_parse_urlencode(
3359 {'__youtubedl_smuggle': json.dumps(data)}
)
3360 return url
+ '#' + sdata
3363 def unsmuggle_url(smug_url
, default
=None):
3364 if '#__youtubedl_smuggle' not in smug_url
:
3365 return smug_url
, default
3366 url
, _
, sdata
= smug_url
.rpartition('#')
3367 jsond
= compat_parse_qs(sdata
)['__youtubedl_smuggle'][0]
3368 data
= json
.loads(jsond
)
3372 def format_bytes(bytes):
3375 if type(bytes) is str:
3376 bytes = float(bytes)
3380 exponent
= int(math
.log(bytes, 1024.0))
3381 suffix
= ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent
]
3382 converted
= float(bytes) / float(1024 ** exponent
)
3383 return '%.2f%s' % (converted
, suffix
)
3386 def lookup_unit_table(unit_table
, s
):
3387 units_re
= '|'.join(re
.escape(u
) for u
in unit_table
)
3389 r
'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re
, s
)
3392 num_str
= m
.group('num').replace(',', '.')
3393 mult
= unit_table
[m
.group('unit')]
3394 return int(float(num_str
) * mult
)
3397 def parse_filesize(s
):
3401 # The lower-case forms are of course incorrect and unofficial,
3402 # but we support those too
3419 'megabytes': 1000 ** 2,
3420 'mebibytes': 1024 ** 2,
3426 'gigabytes': 1000 ** 3,
3427 'gibibytes': 1024 ** 3,
3433 'terabytes': 1000 ** 4,
3434 'tebibytes': 1024 ** 4,
3440 'petabytes': 1000 ** 5,
3441 'pebibytes': 1024 ** 5,
3447 'exabytes': 1000 ** 6,
3448 'exbibytes': 1024 ** 6,
3454 'zettabytes': 1000 ** 7,
3455 'zebibytes': 1024 ** 7,
3461 'yottabytes': 1000 ** 8,
3462 'yobibytes': 1024 ** 8,
3465 return lookup_unit_table(_UNIT_TABLE
, s
)
3474 if re
.match(r
'^[\d,.]+$', s
):
3475 return str_to_int(s
)
3486 return lookup_unit_table(_UNIT_TABLE
, s
)
3489 def parse_resolution(s
):
3493 mobj
= re
.search(r
'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s
)
3496 'width': int(mobj
.group('w')),
3497 'height': int(mobj
.group('h')),
3500 mobj
= re
.search(r
'\b(\d+)[pPiI]\b', s
)
3502 return {'height': int(mobj.group(1))}
3504 mobj
= re
.search(r
'\b([48])[kK]\b', s
)
3506 return {'height': int(mobj.group(1)) * 540}
3511 def parse_bitrate(s
):
3512 if not isinstance(s
, compat_str
):
3514 mobj
= re
.search(r
'\b(\d+)\s*kbps', s
)
3516 return int(mobj
.group(1))
3519 def month_by_name(name
, lang
='en'):
3520 """ Return the number of a month by (locale-independently) English name """
3522 month_names
= MONTH_NAMES
.get(lang
, MONTH_NAMES
['en'])
3525 return month_names
.index(name
) + 1
3530 def month_by_abbreviation(abbrev
):
3531 """ Return the number of a month by (locale-independently) English
3535 return [s
[:3] for s
in ENGLISH_MONTH_NAMES
].index(abbrev
) + 1
3540 def fix_xml_ampersands(xml_str
):
3541 """Replace all the '&' by '&' in XML"""
3543 r
'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3548 def setproctitle(title
):
3549 assert isinstance(title
, compat_str
)
3551 # ctypes in Jython is not complete
3552 # http://bugs.jython.org/issue2148
3553 if sys
.platform
.startswith('java'):
3557 libc
= ctypes
.cdll
.LoadLibrary('libc.so.6')
3561 # LoadLibrary in Windows Python 2.7.13 only expects
3562 # a bytestring, but since unicode_literals turns
3563 # every string into a unicode string, it fails.
3565 title_bytes
= title
.encode('utf-8')
3566 buf
= ctypes
.create_string_buffer(len(title_bytes
))
3567 buf
.value
= title_bytes
3569 libc
.prctl(15, buf
, 0, 0, 0)
3570 except AttributeError:
3571 return # Strange libc, just skip this
3574 def remove_start(s
, start
):
3575 return s
[len(start
):] if s
is not None and s
.startswith(start
) else s
3578 def remove_end(s
, end
):
3579 return s
[:-len(end
)] if s
is not None and s
.endswith(end
) else s
3582 def remove_quotes(s
):
3583 if s
is None or len(s
) < 2:
3585 for quote
in ('"', "'", ):
3586 if s
[0] == quote
and s
[-1] == quote
:
3591 def get_domain(url
):
3592 domain
= re
.match(r
'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url
)
3593 return domain
.group('domain') if domain
else None
3596 def url_basename(url
):
3597 path
= compat_urlparse
.urlparse(url
).path
3598 return path
.strip('/').split('/')[-1]
3602 return re
.match(r
'https?://[^?#&]+/', url
).group()
3605 def urljoin(base
, path
):
3606 if isinstance(path
, bytes):
3607 path
= path
.decode('utf-8')
3608 if not isinstance(path
, compat_str
) or not path
:
3610 if re
.match(r
'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path
):
3612 if isinstance(base
, bytes):
3613 base
= base
.decode('utf-8')
3614 if not isinstance(base
, compat_str
) or not re
.match(
3615 r
'^(?:https?:)?//', base
):
3617 return compat_urlparse
.urljoin(base
, path
)
3620 class HEADRequest(compat_urllib_request
.Request
):
3621 def get_method(self
):
3625 class PUTRequest(compat_urllib_request
.Request
):
3626 def get_method(self
):
3630 def int_or_none(v
, scale
=1, default
=None, get_attr
=None, invscale
=1):
3633 v
= getattr(v
, get_attr
, None)
3639 return int(v
) * invscale
// scale
3640 except (ValueError, TypeError):
3644 def str_or_none(v
, default
=None):
3645 return default
if v
is None else compat_str(v
)
3648 def str_to_int(int_str
):
3649 """ A more relaxed version of int_or_none """
3650 if isinstance(int_str
, compat_integer_types
):
3652 elif isinstance(int_str
, compat_str
):
3653 int_str
= re
.sub(r
'[,\.\+]', '', int_str
)
3654 return int_or_none(int_str
)
3657 def float_or_none(v
, scale
=1, invscale
=1, default
=None):
3661 return float(v
) * invscale
/ scale
3662 except (ValueError, TypeError):
3666 def bool_or_none(v
, default
=None):
3667 return v
if isinstance(v
, bool) else default
3670 def strip_or_none(v
, default
=None):
3671 return v
.strip() if isinstance(v
, compat_str
) else default
3674 def url_or_none(url
):
3675 if not url
or not isinstance(url
, compat_str
):
3678 return url
if re
.match(r
'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url
) else None
3681 def strftime_or_none(timestamp
, date_format
, default
=None):
3682 datetime_object
= None
3684 if isinstance(timestamp
, compat_numeric_types
): # unix timestamp
3685 datetime_object
= datetime
.datetime
.utcfromtimestamp(timestamp
)
3686 elif isinstance(timestamp
, compat_str
): # assume YYYYMMDD
3687 datetime_object
= datetime
.datetime
.strptime(timestamp
, '%Y%m%d')
3688 return datetime_object
.strftime(date_format
)
3689 except (ValueError, TypeError, AttributeError):
3693 def parse_duration(s
):
3694 if not isinstance(s
, compat_basestring
):
3699 days
, hours
, mins
, secs
, ms
= [None] * 5
3700 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
)
3702 days
, hours
, mins
, secs
, ms
= m
.groups()
3707 [0-9]+\s*y(?:ears?)?\s*
3710 [0-9]+\s*m(?:onths?)?\s*
3713 [0-9]+\s*w(?:eeks?)?\s*
3716 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3720 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3723 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3726 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3729 days
, hours
, mins
, secs
, ms
= m
.groups()
3731 m
= re
.match(r
'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s
)
3733 hours
, mins
= m
.groups()
3739 duration
+= float(secs
)
3741 duration
+= float(mins
) * 60
3743 duration
+= float(hours
) * 60 * 60
3745 duration
+= float(days
) * 24 * 60 * 60
3747 duration
+= float(ms
)
3751 def prepend_extension(filename
, ext
, expected_real_ext
=None):
3752 name
, real_ext
= os
.path
.splitext(filename
)
3754 '{0}.{1}{2}'.format(name
, ext
, real_ext
)
3755 if not expected_real_ext
or real_ext
[1:] == expected_real_ext
3756 else '{0}.{1}'.format(filename
, ext
))
3759 def replace_extension(filename
, ext
, expected_real_ext
=None):
3760 name
, real_ext
= os
.path
.splitext(filename
)
3761 return '{0}.{1}'.format(
3762 name
if not expected_real_ext
or real_ext
[1:] == expected_real_ext
else filename
,
3766 def check_executable(exe
, args
=[]):
3767 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3768 args can be a list of arguments for a short output (like -version) """
3770 process_communicate_or_kill(subprocess
.Popen(
3771 [exe
] + args
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
))
3777 def get_exe_version(exe
, args
=['--version'],
3778 version_re
=None, unrecognized
='present'):
3779 """ Returns the version of the specified executable,
3780 or False if the executable is not present """
3782 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3783 # SIGTTOU if youtube-dlc is run in the background.
3784 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3785 out
, _
= process_communicate_or_kill(subprocess
.Popen(
3786 [encodeArgument(exe
)] + args
,
3787 stdin
=subprocess
.PIPE
,
3788 stdout
=subprocess
.PIPE
, stderr
=subprocess
.STDOUT
))
3791 if isinstance(out
, bytes): # Python 2.x
3792 out
= out
.decode('ascii', 'ignore')
3793 return detect_exe_version(out
, version_re
, unrecognized
)
3796 def detect_exe_version(output
, version_re
=None, unrecognized
='present'):
3797 assert isinstance(output
, compat_str
)
3798 if version_re
is None:
3799 version_re
= r
'version\s+([-0-9._a-zA-Z]+)'
3800 m
= re
.search(version_re
, output
)
3807 class PagedList(object):
3809 # This is only useful for tests
3810 return len(self
.getslice())
3813 class OnDemandPagedList(PagedList
):
3814 def __init__(self
, pagefunc
, pagesize
, use_cache
=True):
3815 self
._pagefunc
= pagefunc
3816 self
._pagesize
= pagesize
3817 self
._use
_cache
= use_cache
3821 def getslice(self
, start
=0, end
=None):
3823 for pagenum
in itertools
.count(start
// self
._pagesize
):
3824 firstid
= pagenum
* self
._pagesize
3825 nextfirstid
= pagenum
* self
._pagesize
+ self
._pagesize
3826 if start
>= nextfirstid
:
3831 page_results
= self
._cache
.get(pagenum
)
3832 if page_results
is None:
3833 page_results
= list(self
._pagefunc
(pagenum
))
3835 self
._cache
[pagenum
] = page_results
3838 start
% self
._pagesize
3839 if firstid
<= start
< nextfirstid
3843 ((end
- 1) % self
._pagesize
) + 1
3844 if (end
is not None and firstid
<= end
<= nextfirstid
)
3847 if startv
!= 0 or endv
is not None:
3848 page_results
= page_results
[startv
:endv
]
3849 res
.extend(page_results
)
3851 # A little optimization - if current page is not "full", ie. does
3852 # not contain page_size videos then we can assume that this page
3853 # is the last one - there are no more ids on further pages -
3854 # i.e. no need to query again.
3855 if len(page_results
) + startv
< self
._pagesize
:
3858 # If we got the whole page, but the next page is not interesting,
3859 # break out early as well
3860 if end
== nextfirstid
:
3865 class InAdvancePagedList(PagedList
):
3866 def __init__(self
, pagefunc
, pagecount
, pagesize
):
3867 self
._pagefunc
= pagefunc
3868 self
._pagecount
= pagecount
3869 self
._pagesize
= pagesize
3871 def getslice(self
, start
=0, end
=None):
3873 start_page
= start
// self
._pagesize
3875 self
._pagecount
if end
is None else (end
// self
._pagesize
+ 1))
3876 skip_elems
= start
- start_page
* self
._pagesize
3877 only_more
= None if end
is None else end
- start
3878 for pagenum
in range(start_page
, end_page
):
3879 page
= list(self
._pagefunc
(pagenum
))
3881 page
= page
[skip_elems
:]
3883 if only_more
is not None:
3884 if len(page
) < only_more
:
3885 only_more
-= len(page
)
3887 page
= page
[:only_more
]
3894 def uppercase_escape(s
):
3895 unicode_escape
= codecs
.getdecoder('unicode_escape')
3897 r
'\\U[0-9a-fA-F]{8}',
3898 lambda m
: unicode_escape(m
.group(0))[0],
3902 def lowercase_escape(s
):
3903 unicode_escape
= codecs
.getdecoder('unicode_escape')
3905 r
'\\u[0-9a-fA-F]{4}',
3906 lambda m
: unicode_escape(m
.group(0))[0],
3910 def escape_rfc3986(s
):
3911 """Escape non-ASCII characters as suggested by RFC 3986"""
3912 if sys
.version_info
< (3, 0) and isinstance(s
, compat_str
):
3913 s
= s
.encode('utf-8')
3914 return compat_urllib_parse
.quote(s
, b
"%/;:@&=+$,!~*'()?#[]")
3917 def escape_url(url
):
3918 """Escape URL as suggested by RFC 3986"""
3919 url_parsed
= compat_urllib_parse_urlparse(url
)
3920 return url_parsed
._replace
(
3921 netloc
=url_parsed
.netloc
.encode('idna').decode('ascii'),
3922 path
=escape_rfc3986(url_parsed
.path
),
3923 params
=escape_rfc3986(url_parsed
.params
),
3924 query
=escape_rfc3986(url_parsed
.query
),
3925 fragment
=escape_rfc3986(url_parsed
.fragment
)
3929 def read_batch_urls(batch_fd
):
3931 if not isinstance(url
, compat_str
):
3932 url
= url
.decode('utf-8', 'replace')
3933 BOM_UTF8
= ('\xef\xbb\xbf', '\ufeff')
3934 for bom
in BOM_UTF8
:
3935 if url
.startswith(bom
):
3936 url
= url
[len(bom
):]
3938 if not url
or url
.startswith(('#', ';', ']')):
3940 # "#" cannot be stripped out since it is part of the URI
3941 # However, it can be safely stipped out if follwing a whitespace
3942 return re
.split(r
'\s#', url
, 1)[0].rstrip()
3944 with contextlib
.closing(batch_fd
) as fd
:
3945 return [url
for url
in map(fixup
, fd
) if url
]
3948 def urlencode_postdata(*args
, **kargs
):
3949 return compat_urllib_parse_urlencode(*args
, **kargs
).encode('ascii')
3952 def update_url_query(url
, query
):
3955 parsed_url
= compat_urlparse
.urlparse(url
)
3956 qs
= compat_parse_qs(parsed_url
.query
)
3958 return compat_urlparse
.urlunparse(parsed_url
._replace
(
3959 query
=compat_urllib_parse_urlencode(qs
, True)))
3962 def update_Request(req
, url
=None, data
=None, headers
={}, query={}
):
3963 req_headers
= req
.headers
.copy()
3964 req_headers
.update(headers
)
3965 req_data
= data
or req
.data
3966 req_url
= update_url_query(url
or req
.get_full_url(), query
)
3967 req_get_method
= req
.get_method()
3968 if req_get_method
== 'HEAD':
3969 req_type
= HEADRequest
3970 elif req_get_method
== 'PUT':
3971 req_type
= PUTRequest
3973 req_type
= compat_urllib_request
.Request
3975 req_url
, data
=req_data
, headers
=req_headers
,
3976 origin_req_host
=req
.origin_req_host
, unverifiable
=req
.unverifiable
)
3977 if hasattr(req
, 'timeout'):
3978 new_req
.timeout
= req
.timeout
3982 def _multipart_encode_impl(data
, boundary
):
3983 content_type
= 'multipart/form-data; boundary=%s' % boundary
3986 for k
, v
in data
.items():
3987 out
+= b
'--' + boundary
.encode('ascii') + b
'\r\n'
3988 if isinstance(k
, compat_str
):
3989 k
= k
.encode('utf-8')
3990 if isinstance(v
, compat_str
):
3991 v
= v
.encode('utf-8')
3992 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
3993 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
3994 content
= b
'Content-Disposition: form-data; name="' + k
+ b
'"\r\n\r\n' + v
+ b
'\r\n'
3995 if boundary
.encode('ascii') in content
:
3996 raise ValueError('Boundary overlaps with data')
3999 out
+= b
'--' + boundary
.encode('ascii') + b
'--\r\n'
4001 return out
, content_type
4004 def multipart_encode(data
, boundary
=None):
4006 Encode a dict to RFC 7578-compliant form-data
4009 A dict where keys and values can be either Unicode or bytes-like
4012 If specified a Unicode object, it's used as the boundary. Otherwise
4013 a random boundary is generated.
4015 Reference: https://tools.ietf.org/html/rfc7578
4017 has_specified_boundary
= boundary
is not None
4020 if boundary
is None:
4021 boundary
= '---------------' + str(random
.randrange(0x0fffffff, 0xffffffff))
4024 out
, content_type
= _multipart_encode_impl(data
, boundary
)
4027 if has_specified_boundary
:
4031 return out
, content_type
4034 def dict_get(d
, key_or_keys
, default
=None, skip_false_values
=True):
4035 if isinstance(key_or_keys
, (list, tuple)):
4036 for key
in key_or_keys
:
4037 if key
not in d
or d
[key
] is None or skip_false_values
and not d
[key
]:
4041 return d
.get(key_or_keys
, default
)
4044 def try_get(src
, getter
, expected_type
=None):
4045 if not isinstance(getter
, (list, tuple)):
4050 except (AttributeError, KeyError, TypeError, IndexError):
4053 if expected_type
is None or isinstance(v
, expected_type
):
4057 def merge_dicts(*dicts
):
4059 for a_dict
in dicts
:
4060 for k
, v
in a_dict
.items():
4064 or (isinstance(v
, compat_str
) and v
4065 and isinstance(merged
[k
], compat_str
)
4066 and not merged
[k
])):
4071 def encode_compat_str(string
, encoding
=preferredencoding(), errors
='strict'):
4072 return string
if isinstance(string
, compat_str
) else compat_str(string
, encoding
, errors
)
4084 TV_PARENTAL_GUIDELINES
= {
4094 def parse_age_limit(s
):
4096 return s
if 0 <= s
<= 21 else None
4097 if not isinstance(s
, compat_basestring
):
4099 m
= re
.match(r
'^(?P<age>\d{1,2})\+?$', s
)
4101 return int(m
.group('age'))
4103 return US_RATINGS
[s
]
4104 m
= re
.match(r
'^TV[_-]?(%s)$' % '|'.join(k
[3:] for k
in TV_PARENTAL_GUIDELINES
), s
)
4106 return TV_PARENTAL_GUIDELINES
['TV-' + m
.group(1)]
4110 def strip_jsonp(code
):
4113 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
4114 (?:\s*&&\s*(?P=func_name))?
4115 \s*\(\s*(?P<callback_data>.*)\);?
4116 \s*?(?://[^\n]*)*$''',
4117 r
'\g<callback_data>', code
)
4120 def js_to_json(code
, vars={}):
4121 # vars is a dict of var, val pairs to substitute
4122 COMMENT_RE
= r
'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
4123 SKIP_RE
= r
'\s*(?:{comment})?\s*'.format(comment
=COMMENT_RE
)
4125 (r
'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip
=SKIP_RE
), 16),
4126 (r
'(?s)^(0+[0-7]+){skip}:?$'.format(skip
=SKIP_RE
), 8),
4131 if v
in ('true', 'false', 'null'):
4133 elif v
.startswith('/*') or v
.startswith('//') or v
.startswith('!') or v
== ',':
4136 if v
[0] in ("'", '"'):
4137 v
= re
.sub(r
'(?s)\\.|"', lambda m
: {
4142 }.get(m
.group(0), m
.group(0)), v
[1:-1])
4144 for regex
, base
in INTEGER_TABLE
:
4145 im
= re
.match(regex
, v
)
4147 i
= int(im
.group(1), base
)
4148 return '"%d":' % i
if v
.endswith(':') else '%d' % i
4155 return re
.sub(r
'''(?sx)
4156 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4157 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4158 {comment}|,(?={skip}[\]}}])|
4159 (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
4160 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4163 '''.format(comment
=COMMENT_RE
, skip
=SKIP_RE
), fix_kv
, code
)
4166 def qualities(quality_ids
):
4167 """ Get a numeric quality value out of a list of possible values """
4170 return quality_ids
.index(qid
)
4177 'default': '%(title)s [%(id)s].%(ext)s',
4182 'description': 'description',
4183 'annotation': 'annotations.xml',
4184 'infojson': 'info.json',
4185 'pl_description': 'description',
4186 'pl_infojson': 'info.json',
4190 def limit_length(s
, length
):
4191 """ Add ellipses to overly long strings """
4196 return s
[:length
- len(ELLIPSES
)] + ELLIPSES
4200 def version_tuple(v
):
4201 return tuple(int(e
) for e
in re
.split(r
'[-.]', v
))
4204 def is_outdated_version(version
, limit
, assume_new
=True):
4206 return not assume_new
4208 return version_tuple(version
) < version_tuple(limit
)
4210 return not assume_new
4213 def ytdl_is_updateable():
4214 """ Returns if youtube-dlc can be updated with -U """
4217 from zipimport
import zipimporter
4219 return isinstance(globals().get('__loader__'), zipimporter
) or hasattr(sys
, 'frozen')
4222 def args_to_str(args
):
4223 # Get a short string representation for a subprocess command
4224 return ' '.join(compat_shlex_quote(a
) for a
in args
)
4227 def error_to_compat_str(err
):
4229 # On python 2 error byte string must be decoded with proper
4230 # encoding rather than ascii
4231 if sys
.version_info
[0] < 3:
4232 err_str
= err_str
.decode(preferredencoding())
4236 def mimetype2ext(mt
):
4242 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4243 # it's the most popular one
4244 'audio/mpeg': 'mp3',
4245 'audio/x-wav': 'wav',
4250 _
, _
, res
= mt
.rpartition('/')
4251 res
= res
.split(';')[0].strip().lower()
4255 'smptett+xml': 'tt',
4259 'x-mp4-fragmented': 'mp4',
4260 'x-ms-sami': 'sami',
4263 'x-mpegurl': 'm3u8',
4264 'vnd.apple.mpegurl': 'm3u8',
4268 'vnd.ms-sstr+xml': 'ism',
4275 def parse_codecs(codecs_str
):
4276 # http://tools.ietf.org/html/rfc6381
4279 split_codecs
= list(filter(None, map(
4280 lambda str: str.strip(), codecs_str
.strip().strip(',').split(','))))
4281 vcodec
, acodec
= None, None
4282 for full_codec
in split_codecs
:
4283 codec
= full_codec
.split('.')[0]
4284 if codec
in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4287 elif codec
in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4291 write_string('WARNING: Unknown codec %s\n' % full_codec
, sys
.stderr
)
4292 if not vcodec
and not acodec
:
4293 if len(split_codecs
) == 2:
4295 'vcodec': split_codecs
[0],
4296 'acodec': split_codecs
[1],
4300 'vcodec': vcodec
or 'none',
4301 'acodec': acodec
or 'none',
4306 def urlhandle_detect_ext(url_handle
):
4307 getheader
= url_handle
.headers
.get
4309 cd
= getheader('Content-Disposition')
4311 m
= re
.match(r
'attachment;\s*filename="(?P<filename>[^"]+)"', cd
)
4313 e
= determine_ext(m
.group('filename'), default_ext
=None)
4317 return mimetype2ext(getheader('Content-Type'))
4320 def encode_data_uri(data
, mime_type
):
4321 return 'data:%s;base64,%s' % (mime_type
, base64
.b64encode(data
).decode('ascii'))
4324 def age_restricted(content_limit
, age_limit
):
4325 """ Returns True iff the content should be blocked """
4327 if age_limit
is None: # No limit set
4329 if content_limit
is None:
4330 return False # Content available for everyone
4331 return age_limit
< content_limit
4334 def is_html(first_bytes
):
4335 """ Detect whether a file contains HTML by examining its first bytes. """
4338 (b
'\xef\xbb\xbf', 'utf-8'),
4339 (b
'\x00\x00\xfe\xff', 'utf-32-be'),
4340 (b
'\xff\xfe\x00\x00', 'utf-32-le'),
4341 (b
'\xff\xfe', 'utf-16-le'),
4342 (b
'\xfe\xff', 'utf-16-be'),
4344 for bom
, enc
in BOMS
:
4345 if first_bytes
.startswith(bom
):
4346 s
= first_bytes
[len(bom
):].decode(enc
, 'replace')
4349 s
= first_bytes
.decode('utf-8', 'replace')
4351 return re
.match(r
'^\s*<', s
)
4354 def determine_protocol(info_dict
):
4355 protocol
= info_dict
.get('protocol')
4356 if protocol
is not None:
4359 url
= info_dict
['url']
4360 if url
.startswith('rtmp'):
4362 elif url
.startswith('mms'):
4364 elif url
.startswith('rtsp'):
4367 ext
= determine_ext(url
)
4373 return compat_urllib_parse_urlparse(url
).scheme
4376 def render_table(header_row
, data
, delim
=False, extraGap
=0, hideEmpty
=False):
4377 """ Render a list of rows, each as a list of values """
4379 def get_max_lens(table
):
4380 return [max(len(compat_str(v
)) for v
in col
) for col
in zip(*table
)]
4382 def filter_using_list(row
, filterArray
):
4383 return [col
for (take
, col
) in zip(filterArray
, row
) if take
]
4386 max_lens
= get_max_lens(data
)
4387 header_row
= filter_using_list(header_row
, max_lens
)
4388 data
= [filter_using_list(row
, max_lens
) for row
in data
]
4390 table
= [header_row
] + data
4391 max_lens
= get_max_lens(table
)
4393 table
= [header_row
] + [['-' * ml
for ml
in max_lens
]] + data
4394 format_str
= ' '.join('%-' + compat_str(ml
+ extraGap
) + 's' for ml
in max_lens
[:-1]) + ' %s'
4395 return '\n'.join(format_str
% tuple(row
) for row
in table
)
4398 def _match_one(filter_part
, dct
):
4399 COMPARISON_OPERATORS
= {
4407 operator_rex
= re
.compile(r
'''(?x)\s*
4409 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4411 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4412 (?P<quote>["\'])(?P
<quotedstrval
>(?
:\\.|
(?
!(?P
=quote
)|
\\).)+?
)(?P
=quote
)|
4413 (?P
<strval
>(?
![0-9.])[a
-z0
-9A
-Z
]*)
4416 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4417 m = operator_rex.search(filter_part)
4419 op = COMPARISON_OPERATORS[m.group('op')]
4420 actual_value = dct.get(m.group('key'))
4421 if (m.group('quotedstrval') is not None
4422 or m.group('strval') is not None
4423 # If the original field is a string and matching comparisonvalue is
4424 # a number we should respect the origin of the original field
4425 # and process comparison value as a string (see
4426 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4427 or actual_value is not None and m.group('intval') is not None
4428 and isinstance(actual_value, compat_str)):
4429 if m.group('op') not in ('=', '!='):
4431 'Operator %s does not support string values!' % m.group('op'))
4432 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4433 quote = m.group('quote')
4434 if quote is not None:
4435 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4438 comparison_value = int(m.group('intval'))
4440 comparison_value = parse_filesize(m.group('intval'))
4441 if comparison_value is None:
4442 comparison_value = parse_filesize(m.group('intval') + 'B')
4443 if comparison_value is None:
4445 'Invalid integer value %r in filter part %r' % (
4446 m.group('intval'), filter_part))
4447 if actual_value is None:
4448 return m.group('none_inclusive')
4449 return op(actual_value, comparison_value)
4452 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4453 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4455 operator_rex = re.compile(r'''(?x
)\s
*
4456 (?P
<op
>%s)\s
*(?P
<key
>[a
-z_
]+)
4458 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4459 m = operator_rex.search(filter_part)
4461 op = UNARY_OPERATORS[m.group('op')]
4462 actual_value = dct.get(m.group('key'))
4463 return op(actual_value)
4465 raise ValueError('Invalid filter part %r' % filter_part)
4468 def match_str(filter_str, dct):
4469 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
4472 _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
4475 def match_filter_func(filter_str):
4476 def _match_func(info_dict):
4477 if match_str(filter_str, info_dict):
4480 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4481 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4485 def parse_dfxp_time_expr(time_expr):
4489 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4491 return float(mobj.group('time_offset'))
4493 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4495 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4498 def srt_subtitles_timecode(seconds):
4499 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4502 def dfxp2srt(dfxp_data):
4504 @param dfxp_data A
bytes-like
object containing DFXP data
4505 @returns A
unicode object containing converted SRT data
4507 LEGACY_NAMESPACES = (
4508 (b'http://www.w3.org/ns/ttml', [
4509 b'http://www.w3.org/2004/11/ttaf1',
4510 b'http://www.w3.org/2006/04/ttaf1',
4511 b'http://www.w3.org/2006/10/ttaf1',
4513 (b'http://www.w3.org/ns/ttml#styling', [
4514 b'http://www.w3.org/ns/ttml#style',
4518 SUPPORTED_STYLING = [
4527 _x = functools.partial(xpath_with_ns, ns_map={
4528 'xml': 'http://www.w3.org/XML/1998/namespace',
4529 'ttml': 'http://www.w3.org/ns/ttml',
4530 'tts': 'http://www.w3.org/ns/ttml#styling',
4536 class TTMLPElementParser(object):
4538 _unclosed_elements = []
4539 _applied_styles = []
4541 def start(self, tag, attrib):
4542 if tag in (_x('ttml:br'), 'br'):
4545 unclosed_elements = []
4547 element_style_id = attrib.get('style')
4549 style.update(default_style)
4550 if element_style_id:
4551 style.update(styles.get(element_style_id, {}))
4552 for prop in SUPPORTED_STYLING:
4553 prop_val = attrib.get(_x('tts:' + prop))
4555 style[prop] = prop_val
4558 for k, v in sorted(style.items()):
4559 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4562 font += ' color="%s"' % v
4563 elif k == 'fontSize':
4564 font += ' size="%s"' % v
4565 elif k == 'fontFamily':
4566 font += ' face="%s"' % v
4567 elif k == 'fontWeight' and v == 'bold':
4569 unclosed_elements.append('b')
4570 elif k == 'fontStyle' and v == 'italic':
4572 unclosed_elements.append('i')
4573 elif k == 'textDecoration' and v == 'underline':
4575 unclosed_elements.append('u')
4577 self._out += '<font' + font + '>'
4578 unclosed_elements.append('font')
4580 if self._applied_styles:
4581 applied_style.update(self._applied_styles[-1])
4582 applied_style.update(style)
4583 self._applied_styles.append(applied_style)
4584 self._unclosed_elements.append(unclosed_elements)
4587 if tag not in (_x('ttml:br'), 'br'):
4588 unclosed_elements = self._unclosed_elements.pop()
4589 for element in reversed(unclosed_elements):
4590 self._out += '</%s>' % element
4591 if unclosed_elements and self._applied_styles:
4592 self._applied_styles.pop()
4594 def data(self, data):
4598 return self._out.strip()
4600 def parse_node(node):
4601 target = TTMLPElementParser()
4602 parser = xml.etree.ElementTree.XMLParser(target=target)
4603 parser.feed(xml.etree.ElementTree.tostring(node))
4604 return parser.close()
4606 for k, v in LEGACY_NAMESPACES:
4608 dfxp_data = dfxp_data.replace(ns, k)
4610 dfxp = compat_etree_fromstring(dfxp_data)
4612 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4615 raise ValueError('Invalid dfxp/TTML subtitle')
4619 for style in dfxp.findall(_x('.//ttml:style')):
4620 style_id = style.get('id') or style.get(_x('xml:id'))
4623 parent_style_id = style.get('style')
4625 if parent_style_id not in styles:
4628 styles[style_id] = styles[parent_style_id].copy()
4629 for prop in SUPPORTED_STYLING:
4630 prop_val = style.get(_x('tts:' + prop))
4632 styles.setdefault(style_id, {})[prop] = prop_val
4638 for p in ('body', 'div'):
4639 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4642 style = styles.get(ele.get('style'))
4645 default_style.update(style)
4647 for para, index in zip(paras, itertools.count(1)):
4648 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
4649 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
4650 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
4651 if begin_time is None:
4656 end_time = begin_time + dur
4657 out.append('%d\n%s --> %s\n%s\n\n' % (
4659 srt_subtitles_timecode(begin_time),
4660 srt_subtitles_timecode(end_time),
4666 def cli_option(params, command_option, param):
4667 param = params.get(param)
4669 param = compat_str(param)
4670 return [command_option, param] if param is not None else []
4673 def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
4674 param = params.get(param)
4677 assert isinstance(param, bool)
4679 return [command_option + separator + (true_value if param else false_value)]
4680 return [command_option, true_value if param else false_value]
4683 def cli_valueless_option(params, command_option, param, expected_value=True):
4684 param = params.get(param)
4685 return [command_option] if param == expected_value else []
4688 def cli_configuration_args(params, arg_name, key, default=[], exe=None): # returns arg, for_compat
4689 argdict = params.get(arg_name, {})
4690 if isinstance(argdict, (list, tuple)): # for backward compatibility
4691 return argdict, True
4694 return default, False
4695 assert isinstance(argdict, dict)
4698 args = exe_args = None
4700 assert isinstance(exe, compat_str)
4702 args = argdict.get('%s+%s' % (key, exe))
4704 exe_args = argdict.get(exe)
4707 args = argdict.get(key) if key != exe else None
4708 if args is None and exe_args is None:
4709 args = argdict.get('default', default)
4711 args, exe_args = args or [], exe_args or []
4712 assert isinstance(args, (list, tuple))
4713 assert isinstance(exe_args, (list, tuple))
4714 return args + exe_args, False
4717 class ISO639Utils(object):
4718 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
4777 'iw': 'heb', # Replaced by he in 1989 revision
4787 'in': 'ind', # Replaced by id in 1989 revision
4902 'ji': 'yid', # Replaced by yi in 1989 revision
4910 def short2long(cls, code):
4911 """Convert language code from ISO 639-1 to ISO 639-2/T"""
4912 return cls._lang_map.get(code[:2])
4915 def long2short(cls, code):
4916 """Convert language code from ISO 639-2/T to ISO 639-1"""
4917 for short_name, long_name in cls._lang_map.items():
4918 if long_name == code:
4922 class ISO3166Utils(object):
4923 # From http://data.okfn.org/data/core/country-list
4925 'AF': 'Afghanistan',
4926 'AX': 'Åland Islands',
4929 'AS': 'American Samoa',
4934 'AG': 'Antigua and Barbuda',
4951 'BO': 'Bolivia, Plurinational State of',
4952 'BQ': 'Bonaire, Sint Eustatius and Saba',
4953 'BA': 'Bosnia and Herzegovina',
4955 'BV': 'Bouvet Island',
4957 'IO': 'British Indian Ocean Territory',
4958 'BN': 'Brunei Darussalam',
4960 'BF': 'Burkina Faso',
4966 'KY': 'Cayman Islands',
4967 'CF': 'Central African Republic',
4971 'CX': 'Christmas Island',
4972 'CC': 'Cocos (Keeling) Islands',
4976 'CD': 'Congo, the Democratic Republic of the',
4977 'CK': 'Cook Islands',
4979 'CI': 'Côte d\'Ivoire',
4984 'CZ': 'Czech Republic',
4988 'DO': 'Dominican Republic',
4991 'SV': 'El Salvador',
4992 'GQ': 'Equatorial Guinea',
4996 'FK': 'Falkland Islands (Malvinas)',
4997 'FO': 'Faroe Islands',
5001 'GF': 'French Guiana',
5002 'PF': 'French Polynesia',
5003 'TF': 'French Southern Territories',
5018 'GW': 'Guinea-Bissau',
5021 'HM': 'Heard Island and McDonald Islands',
5022 'VA': 'Holy See (Vatican City State)',
5029 'IR': 'Iran, Islamic Republic of',
5032 'IM': 'Isle of Man',
5042 'KP': 'Korea, Democratic People\'s Republic of',
5043 'KR': 'Korea, Republic of',
5046 'LA': 'Lao People\'s Democratic Republic',
5052 'LI': 'Liechtenstein',
5056 'MK': 'Macedonia, the Former Yugoslav Republic of',
5063 'MH': 'Marshall Islands',
5069 'FM': 'Micronesia, Federated States of',
5070 'MD': 'Moldova, Republic of',
5081 'NL': 'Netherlands',
5082 'NC': 'New Caledonia',
5083 'NZ': 'New Zealand',
5088 'NF': 'Norfolk Island',
5089 'MP': 'Northern Mariana Islands',
5094 'PS': 'Palestine, State of',
5096 'PG': 'Papua New Guinea',
5099 'PH': 'Philippines',
5103 'PR': 'Puerto Rico',
5107 'RU': 'Russian Federation',
5109 'BL': 'Saint Barthélemy',
5110 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
5111 'KN': 'Saint Kitts and Nevis',
5112 'LC': 'Saint Lucia',
5113 'MF': 'Saint Martin (French part)',
5114 'PM': 'Saint Pierre and Miquelon',
5115 'VC': 'Saint Vincent and the Grenadines',
5118 'ST': 'Sao Tome and Principe',
5119 'SA': 'Saudi Arabia',
5123 'SL': 'Sierra Leone',
5125 'SX': 'Sint Maarten (Dutch part)',
5128 'SB': 'Solomon Islands',
5130 'ZA': 'South Africa',
5131 'GS': 'South Georgia and the South Sandwich Islands',
5132 'SS': 'South Sudan',
5137 'SJ': 'Svalbard and Jan Mayen',
5140 'CH': 'Switzerland',
5141 'SY': 'Syrian Arab Republic',
5142 'TW': 'Taiwan, Province of China',
5144 'TZ': 'Tanzania, United Republic of',
5146 'TL': 'Timor-Leste',
5150 'TT': 'Trinidad and Tobago',
5153 'TM': 'Turkmenistan',
5154 'TC': 'Turks and Caicos Islands',
5158 'AE': 'United Arab Emirates',
5159 'GB': 'United Kingdom',
5160 'US': 'United States',
5161 'UM': 'United States Minor Outlying Islands',
5165 'VE': 'Venezuela, Bolivarian Republic of',
5167 'VG': 'Virgin Islands, British',
5168 'VI': 'Virgin Islands, U.S.',
5169 'WF': 'Wallis and Futuna',
5170 'EH': 'Western Sahara',
5177 def short2full(cls, code):
5178 """Convert an ISO 3166-2 country code to the corresponding full name"""
5179 return cls._country_map.get(code.upper())
5182 class GeoUtils(object):
5183 # Major IPv4 address blocks per country
5185 'AD': '46.172.224.0/19',
5186 'AE': '94.200.0.0/13',
5187 'AF': '149.54.0.0/17',
5188 'AG': '209.59.64.0/18',
5189 'AI': '204.14.248.0/21',
5190 'AL': '46.99.0.0/16',
5191 'AM': '46.70.0.0/15',
5192 'AO': '105.168.0.0/13',
5193 'AP': '182.50.184.0/21',
5194 'AQ': '23.154.160.0/24',
5195 'AR': '181.0.0.0/12',
5196 'AS': '202.70.112.0/20',
5197 'AT': '77.116.0.0/14',
5198 'AU': '1.128.0.0/11',
5199 'AW': '181.41.0.0/18',
5200 'AX': '185.217.4.0/22',
5201 'AZ': '5.197.0.0/16',
5202 'BA': '31.176.128.0/17',
5203 'BB': '65.48.128.0/17',
5204 'BD': '114.130.0.0/16',
5206 'BF': '102.178.0.0/15',
5207 'BG': '95.42.0.0/15',
5208 'BH': '37.131.0.0/17',
5209 'BI': '154.117.192.0/18',
5210 'BJ': '137.255.0.0/16',
5211 'BL': '185.212.72.0/23',
5212 'BM': '196.12.64.0/18',
5213 'BN': '156.31.0.0/16',
5214 'BO': '161.56.0.0/16',
5215 'BQ': '161.0.80.0/20',
5216 'BR': '191.128.0.0/12',
5217 'BS': '24.51.64.0/18',
5218 'BT': '119.2.96.0/19',
5219 'BW': '168.167.0.0/16',
5220 'BY': '178.120.0.0/13',
5221 'BZ': '179.42.192.0/18',
5222 'CA': '99.224.0.0/11',
5223 'CD': '41.243.0.0/16',
5224 'CF': '197.242.176.0/21',
5225 'CG': '160.113.0.0/16',
5226 'CH': '85.0.0.0/13',
5227 'CI': '102.136.0.0/14',
5228 'CK': '202.65.32.0/19',
5229 'CL': '152.172.0.0/14',
5230 'CM': '102.244.0.0/14',
5231 'CN': '36.128.0.0/10',
5232 'CO': '181.240.0.0/12',
5233 'CR': '201.192.0.0/12',
5234 'CU': '152.206.0.0/15',
5235 'CV': '165.90.96.0/19',
5236 'CW': '190.88.128.0/17',
5237 'CY': '31.153.0.0/16',
5238 'CZ': '88.100.0.0/14',
5240 'DJ': '197.241.0.0/17',
5241 'DK': '87.48.0.0/12',
5242 'DM': '192.243.48.0/20',
5243 'DO': '152.166.0.0/15',
5244 'DZ': '41.96.0.0/12',
5245 'EC': '186.68.0.0/15',
5246 'EE': '90.190.0.0/15',
5247 'EG': '156.160.0.0/11',
5248 'ER': '196.200.96.0/20',
5249 'ES': '88.0.0.0/11',
5250 'ET': '196.188.0.0/14',
5251 'EU': '2.16.0.0/13',
5252 'FI': '91.152.0.0/13',
5253 'FJ': '144.120.0.0/16',
5254 'FK': '80.73.208.0/21',
5255 'FM': '119.252.112.0/20',
5256 'FO': '88.85.32.0/19',
5258 'GA': '41.158.0.0/15',
5260 'GD': '74.122.88.0/21',
5261 'GE': '31.146.0.0/16',
5262 'GF': '161.22.64.0/18',
5263 'GG': '62.68.160.0/19',
5264 'GH': '154.160.0.0/12',
5265 'GI': '95.164.0.0/16',
5266 'GL': '88.83.0.0/19',
5267 'GM': '160.182.0.0/15',
5268 'GN': '197.149.192.0/18',
5269 'GP': '104.250.0.0/19',
5270 'GQ': '105.235.224.0/20',
5271 'GR': '94.64.0.0/13',
5272 'GT': '168.234.0.0/16',
5273 'GU': '168.123.0.0/16',
5274 'GW': '197.214.80.0/20',
5275 'GY': '181.41.64.0/18',
5276 'HK': '113.252.0.0/14',
5277 'HN': '181.210.0.0/16',
5278 'HR': '93.136.0.0/13',
5279 'HT': '148.102.128.0/17',
5280 'HU': '84.0.0.0/14',
5281 'ID': '39.192.0.0/10',
5282 'IE': '87.32.0.0/12',
5283 'IL': '79.176.0.0/13',
5284 'IM': '5.62.80.0/20',
5285 'IN': '117.192.0.0/10',
5286 'IO': '203.83.48.0/21',
5287 'IQ': '37.236.0.0/14',
5288 'IR': '2.176.0.0/12',
5289 'IS': '82.221.0.0/16',
5290 'IT': '79.0.0.0/10',
5291 'JE': '87.244.64.0/18',
5292 'JM': '72.27.0.0/17',
5293 'JO': '176.29.0.0/16',
5294 'JP': '133.0.0.0/8',
5295 'KE': '105.48.0.0/12',
5296 'KG': '158.181.128.0/17',
5297 'KH': '36.37.128.0/17',
5298 'KI': '103.25.140.0/22',
5299 'KM': '197.255.224.0/20',
5300 'KN': '198.167.192.0/19',
5301 'KP': '175.45.176.0/22',
5302 'KR': '175.192.0.0/10',
5303 'KW': '37.36.0.0/14',
5304 'KY': '64.96.0.0/15',
5305 'KZ': '2.72.0.0/13',
5306 'LA': '115.84.64.0/18',
5307 'LB': '178.135.0.0/16',
5308 'LC': '24.92.144.0/20',
5309 'LI': '82.117.0.0/19',
5310 'LK': '112.134.0.0/15',
5311 'LR': '102.183.0.0/16',
5312 'LS': '129.232.0.0/17',
5313 'LT': '78.56.0.0/13',
5314 'LU': '188.42.0.0/16',
5315 'LV': '46.109.0.0/16',
5316 'LY': '41.252.0.0/14',
5317 'MA': '105.128.0.0/11',
5318 'MC': '88.209.64.0/18',
5319 'MD': '37.246.0.0/16',
5320 'ME': '178.175.0.0/17',
5321 'MF': '74.112.232.0/21',
5322 'MG': '154.126.0.0/17',
5323 'MH': '117.103.88.0/21',
5324 'MK': '77.28.0.0/15',
5325 'ML': '154.118.128.0/18',
5326 'MM': '37.111.0.0/17',
5327 'MN': '49.0.128.0/17',
5328 'MO': '60.246.0.0/16',
5329 'MP': '202.88.64.0/20',
5330 'MQ': '109.203.224.0/19',
5331 'MR': '41.188.64.0/18',
5332 'MS': '208.90.112.0/22',
5333 'MT': '46.11.0.0/16',
5334 'MU': '105.16.0.0/12',
5335 'MV': '27.114.128.0/18',
5336 'MW': '102.70.0.0/15',
5337 'MX': '187.192.0.0/11',
5338 'MY': '175.136.0.0/13',
5339 'MZ': '197.218.0.0/15',
5340 'NA': '41.182.0.0/16',
5341 'NC': '101.101.0.0/18',
5342 'NE': '197.214.0.0/18',
5343 'NF': '203.17.240.0/22',
5344 'NG': '105.112.0.0/12',
5345 'NI': '186.76.0.0/15',
5346 'NL': '145.96.0.0/11',
5347 'NO': '84.208.0.0/13',
5348 'NP': '36.252.0.0/15',
5349 'NR': '203.98.224.0/19',
5350 'NU': '49.156.48.0/22',
5351 'NZ': '49.224.0.0/14',
5352 'OM': '5.36.0.0/15',
5353 'PA': '186.72.0.0/15',
5354 'PE': '186.160.0.0/14',
5355 'PF': '123.50.64.0/18',
5356 'PG': '124.240.192.0/19',
5357 'PH': '49.144.0.0/13',
5358 'PK': '39.32.0.0/11',
5359 'PL': '83.0.0.0/11',
5360 'PM': '70.36.0.0/20',
5361 'PR': '66.50.0.0/16',
5362 'PS': '188.161.0.0/16',
5363 'PT': '85.240.0.0/13',
5364 'PW': '202.124.224.0/20',
5365 'PY': '181.120.0.0/14',
5366 'QA': '37.210.0.0/15',
5367 'RE': '102.35.0.0/16',
5368 'RO': '79.112.0.0/13',
5369 'RS': '93.86.0.0/15',
5370 'RU': '5.136.0.0/13',
5371 'RW': '41.186.0.0/16',
5372 'SA': '188.48.0.0/13',
5373 'SB': '202.1.160.0/19',
5374 'SC': '154.192.0.0/11',
5375 'SD': '102.120.0.0/13',
5376 'SE': '78.64.0.0/12',
5377 'SG': '8.128.0.0/10',
5378 'SI': '188.196.0.0/14',
5379 'SK': '78.98.0.0/15',
5380 'SL': '102.143.0.0/17',
5381 'SM': '89.186.32.0/19',
5382 'SN': '41.82.0.0/15',
5383 'SO': '154.115.192.0/18',
5384 'SR': '186.179.128.0/17',
5385 'SS': '105.235.208.0/21',
5386 'ST': '197.159.160.0/19',
5387 'SV': '168.243.0.0/16',
5388 'SX': '190.102.0.0/20',
5390 'SZ': '41.84.224.0/19',
5391 'TC': '65.255.48.0/20',
5392 'TD': '154.68.128.0/19',
5393 'TG': '196.168.0.0/14',
5394 'TH': '171.96.0.0/13',
5395 'TJ': '85.9.128.0/18',
5396 'TK': '27.96.24.0/21',
5397 'TL': '180.189.160.0/20',
5398 'TM': '95.85.96.0/19',
5399 'TN': '197.0.0.0/11',
5400 'TO': '175.176.144.0/21',
5401 'TR': '78.160.0.0/11',
5402 'TT': '186.44.0.0/15',
5403 'TV': '202.2.96.0/19',
5404 'TW': '120.96.0.0/11',
5405 'TZ': '156.156.0.0/14',
5406 'UA': '37.52.0.0/14',
5407 'UG': '102.80.0.0/13',
5409 'UY': '167.56.0.0/13',
5410 'UZ': '84.54.64.0/18',
5411 'VA': '212.77.0.0/19',
5412 'VC': '207.191.240.0/21',
5413 'VE': '186.88.0.0/13',
5414 'VG': '66.81.192.0/20',
5415 'VI': '146.226.0.0/16',
5416 'VN': '14.160.0.0/11',
5417 'VU': '202.80.32.0/20',
5418 'WF': '117.20.32.0/21',
5419 'WS': '202.4.32.0/19',
5420 'YE': '134.35.0.0/16',
5421 'YT': '41.242.116.0/22',
5422 'ZA': '41.0.0.0/11',
5423 'ZM': '102.144.0.0/13',
5424 'ZW': '102.177.192.0/18',
5428 def random_ipv4(cls, code_or_block):
5429 if len(code_or_block) == 2:
5430 block = cls._country_ip_map.get(code_or_block.upper())
5434 block = code_or_block
5435 addr, preflen = block.split('/')
5436 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5437 addr_max = addr_min | (0xffffffff >> int(preflen))
5438 return compat_str(socket.inet_ntoa(
5439 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5442 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5443 def __init__(self, proxies=None):
5444 # Set default handlers
5445 for type in ('http', 'https'):
5446 setattr(self, '%s_open' % type,
5447 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5448 meth(r, proxy, type))
5449 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5451 def proxy_open(self, req, proxy, type):
5452 req_proxy = req.headers.get('Ytdl-request-proxy')
5453 if req_proxy is not None:
5455 del req.headers['Ytdl-request-proxy']
5457 if proxy == '__noproxy__':
5458 return None # No Proxy
5459 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5460 req.add_header('Ytdl-socks-proxy', proxy)
5461 # youtube-dlc's http/https handlers do wrapping the socket with socks
5463 return compat_urllib_request.ProxyHandler.proxy_open(
5464 self, req, proxy, type)
5467 # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5468 # released into Public Domain
5469 # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5471 def long_to_bytes(n, blocksize=0):
5472 """long_to_bytes(n:long, blocksize:int) : string
5473 Convert a long integer to a byte string.
5475 If optional blocksize is given and greater than zero, pad the front of the
5476 byte string with binary zeros so that the length is a multiple of
5479 # after much testing, this algorithm was deemed to be the fastest
5483 s = compat_struct_pack('>I', n & 0xffffffff) + s
5485 # strip off leading zeros
5486 for i in range(len(s)):
5487 if s[i] != b'\000'[0]:
5490 # only happens when n == 0
5494 # add back some pad bytes. this could be done more efficiently w.r.t. the
5495 # de-padding being done above, but sigh...
5496 if blocksize > 0 and len(s) % blocksize:
5497 s = (blocksize - len(s) % blocksize) * b'\000' + s
5501 def bytes_to_long(s):
5502 """bytes_to_long(string) : long
5503 Convert a byte string to a long integer.
5505 This is (essentially) the inverse of long_to_bytes().
5510 extra = (4 - length % 4)
5511 s = b'\000' * extra + s
5512 length = length + extra
5513 for i in range(0, length, 4):
5514 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5518 def ohdave_rsa_encrypt(data, exponent, modulus):
5520 Implement OHDave
's RSA algorithm. See http://www.ohdave.com/rsa/
5523 data: data to encrypt, bytes-like object
5524 exponent, modulus: parameter e and N of RSA algorithm, both integer
5525 Output: hex string of encrypted data
5527 Limitation: supports one block encryption only
5530 payload = int(binascii.hexlify(data[::-1]), 16)
5531 encrypted = pow(payload, exponent, modulus)
5532 return '%x' % encrypted
5535 def pkcs1pad(data, length):
5537 Padding input data with PKCS#1 scheme
5539 @param {int[]} data input data
5540 @param {int} length target length
5541 @returns {int[]} padded data
5543 if len(data) > length - 11:
5544 raise ValueError('Input data too
long for PKCS
#1 padding')
5546 pseudo_random
= [random
.randint(0, 254) for _
in range(length
- len(data
) - 3)]
5547 return [0, 2] + pseudo_random
+ [0] + data
5550 def encode_base_n(num
, n
, table
=None):
5551 FULL_TABLE
= '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5553 table
= FULL_TABLE
[:n
]
5556 raise ValueError('base %d exceeds table length %d' % (n
, len(table
)))
5563 ret
= table
[num
% n
] + ret
5568 def decode_packed_codes(code
):
5569 mobj
= re
.search(PACKED_CODES_RE
, code
)
5570 obfuscated_code
, base
, count
, symbols
= mobj
.groups()
5573 symbols
= symbols
.split('|')
5578 base_n_count
= encode_base_n(count
, base
)
5579 symbol_table
[base_n_count
] = symbols
[count
] or base_n_count
5582 r
'\b(\w+)\b', lambda mobj
: symbol_table
[mobj
.group(0)],
5586 def caesar(s
, alphabet
, shift
):
5591 alphabet
[(alphabet
.index(c
) + shift
) % l
] if c
in alphabet
else c
5596 return caesar(s
, r
'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5599 def parse_m3u8_attributes(attrib
):
5601 for (key
, val
) in re
.findall(r
'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib
):
5602 if val
.startswith('"'):
5608 def urshift(val
, n
):
5609 return val
>> n
if val
>= 0 else (val
+ 0x100000000) >> n
5612 # Based on png2str() written by @gdkchan and improved by @yokrysty
5613 # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5614 def decode_png(png_data
):
5615 # Reference: https://www.w3.org/TR/PNG/
5616 header
= png_data
[8:]
5618 if png_data
[:8] != b
'\x89PNG\x0d\x0a\x1a\x0a' or header
[4:8] != b
'IHDR':
5619 raise IOError('Not a valid PNG file.')
5621 int_map
= {1: '>B', 2: '>H', 4: '>I'}
5622 unpack_integer
= lambda x
: compat_struct_unpack(int_map
[len(x
)], x
)[0]
5627 length
= unpack_integer(header
[:4])
5630 chunk_type
= header
[:4]
5633 chunk_data
= header
[:length
]
5634 header
= header
[length
:]
5636 header
= header
[4:] # Skip CRC
5644 ihdr
= chunks
[0]['data']
5646 width
= unpack_integer(ihdr
[:4])
5647 height
= unpack_integer(ihdr
[4:8])
5651 for chunk
in chunks
:
5652 if chunk
['type'] == b
'IDAT':
5653 idat
+= chunk
['data']
5656 raise IOError('Unable to read PNG data.')
5658 decompressed_data
= bytearray(zlib
.decompress(idat
))
5663 def _get_pixel(idx
):
5668 for y
in range(height
):
5669 basePos
= y
* (1 + stride
)
5670 filter_type
= decompressed_data
[basePos
]
5674 pixels
.append(current_row
)
5676 for x
in range(stride
):
5677 color
= decompressed_data
[1 + basePos
+ x
]
5678 basex
= y
* stride
+ x
5683 left
= _get_pixel(basex
- 3)
5685 up
= _get_pixel(basex
- stride
)
5687 if filter_type
== 1: # Sub
5688 color
= (color
+ left
) & 0xff
5689 elif filter_type
== 2: # Up
5690 color
= (color
+ up
) & 0xff
5691 elif filter_type
== 3: # Average
5692 color
= (color
+ ((left
+ up
) >> 1)) & 0xff
5693 elif filter_type
== 4: # Paeth
5699 c
= _get_pixel(basex
- stride
- 3)
5707 if pa
<= pb
and pa
<= pc
:
5708 color
= (color
+ a
) & 0xff
5710 color
= (color
+ b
) & 0xff
5712 color
= (color
+ c
) & 0xff
5714 current_row
.append(color
)
5716 return width
, height
, pixels
5719 def write_xattr(path
, key
, value
):
5720 # This mess below finds the best xattr tool for the job
5722 # try the pyxattr module...
5725 if hasattr(xattr
, 'set'): # pyxattr
5726 # Unicode arguments are not supported in python-pyxattr until
5728 # See https://github.com/ytdl-org/youtube-dl/issues/5498
5729 pyxattr_required_version
= '0.5.0'
5730 if version_tuple(xattr
.__version
__) < version_tuple(pyxattr_required_version
):
5731 # TODO: fallback to CLI tools
5732 raise XAttrUnavailableError(
5733 'python-pyxattr is detected but is too old. '
5734 'youtube-dlc requires %s or above while your version is %s. '
5735 'Falling back to other xattr implementations' % (
5736 pyxattr_required_version
, xattr
.__version
__))
5738 setxattr
= xattr
.set
5740 setxattr
= xattr
.setxattr
5743 setxattr(path
, key
, value
)
5744 except EnvironmentError as e
:
5745 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5748 if compat_os_name
== 'nt':
5749 # Write xattrs to NTFS Alternate Data Streams:
5750 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
5751 assert ':' not in key
5752 assert os
.path
.exists(path
)
5754 ads_fn
= path
+ ':' + key
5756 with open(ads_fn
, 'wb') as f
:
5758 except EnvironmentError as e
:
5759 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5761 user_has_setfattr
= check_executable('setfattr', ['--version'])
5762 user_has_xattr
= check_executable('xattr', ['-h'])
5764 if user_has_setfattr
or user_has_xattr
:
5766 value
= value
.decode('utf-8')
5767 if user_has_setfattr
:
5768 executable
= 'setfattr'
5769 opts
= ['-n', key
, '-v', value
]
5770 elif user_has_xattr
:
5771 executable
= 'xattr'
5772 opts
= ['-w', key
, value
]
5774 cmd
= ([encodeFilename(executable
, True)]
5775 + [encodeArgument(o
) for o
in opts
]
5776 + [encodeFilename(path
, True)])
5779 p
= subprocess
.Popen(
5780 cmd
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
, stdin
=subprocess
.PIPE
)
5781 except EnvironmentError as e
:
5782 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5783 stdout
, stderr
= process_communicate_or_kill(p
)
5784 stderr
= stderr
.decode('utf-8', 'replace')
5785 if p
.returncode
!= 0:
5786 raise XAttrMetadataError(p
.returncode
, stderr
)
5789 # On Unix, and can't find pyxattr, setfattr, or xattr.
5790 if sys
.platform
.startswith('linux'):
5791 raise XAttrUnavailableError(
5792 "Couldn't find a tool to set the xattrs. "
5793 "Install either the python 'pyxattr' or 'xattr' "
5794 "modules, or the GNU 'attr' package "
5795 "(which contains the 'setfattr' tool).")
5797 raise XAttrUnavailableError(
5798 "Couldn't find a tool to set the xattrs. "
5799 "Install either the python 'xattr' module, "
5800 "or the 'xattr' binary.")
5803 def random_birthday(year_field
, month_field
, day_field
):
5804 start_date
= datetime
.date(1950, 1, 1)
5805 end_date
= datetime
.date(1995, 12, 31)
5806 offset
= random
.randint(0, (end_date
- start_date
).days
)
5807 random_date
= start_date
+ datetime
.timedelta(offset
)
5809 year_field
: str(random_date
.year
),
5810 month_field
: str(random_date
.month
),
5811 day_field
: str(random_date
.day
),
5815 # Templates for internet shortcut files, which are plain text files.
5816 DOT_URL_LINK_TEMPLATE
= '''
5821 DOT_WEBLOC_LINK_TEMPLATE
= '''
5822 <?xml version="1.0" encoding="UTF-8"?>
5823 <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
5824 <plist version="1.0">
5827 \t<string>%(url)s</string>
5832 DOT_DESKTOP_LINK_TEMPLATE
= '''
5842 def iri_to_uri(iri
):
5844 Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
5846 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.
5849 iri_parts
= compat_urllib_parse_urlparse(iri
)
5851 if '[' in iri_parts
.netloc
:
5852 raise ValueError('IPv6 URIs are not, yet, supported.')
5853 # Querying `.netloc`, when there's only one bracket, also raises a ValueError.
5855 # 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.
5858 if iri_parts
.username
:
5859 net_location
+= compat_urllib_parse_quote(iri_parts
.username
, safe
=r
"!$%&'()*+,~")
5860 if iri_parts
.password
is not None:
5861 net_location
+= ':' + compat_urllib_parse_quote(iri_parts
.password
, safe
=r
"!$%&'()*+,~")
5864 net_location
+= iri_parts
.hostname
.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
5865 # The 'idna' encoding produces ASCII text.
5866 if iri_parts
.port
is not None and iri_parts
.port
!= 80:
5867 net_location
+= ':' + str(iri_parts
.port
)
5869 return compat_urllib_parse_urlunparse(
5873 compat_urllib_parse_quote_plus(iri_parts
.path
, safe
=r
"!$%&'()*+,/:;=@|~"),
5875 # Unsure about the `safe` argument, since this is a legacy way of handling parameters.
5876 compat_urllib_parse_quote_plus(iri_parts
.params
, safe
=r
"!$%&'()*+,/:;=@|~"),
5878 # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
5879 compat_urllib_parse_quote_plus(iri_parts
.query
, safe
=r
"!$%&'()*+,/:;=?@{|}~"),
5881 compat_urllib_parse_quote_plus(iri_parts
.fragment
, safe
=r
"!#$%&'()*+,/:;=?@{|}~")))
5883 # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
5886 def to_high_limit_path(path
):
5887 if sys
.platform
in ['win32', 'cygwin']:
5888 # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
5889 return r
'\\?\ '.rstrip() + os
.path
.abspath(path
)
5894 def format_field(obj
, field
, template
='%s', ignore
=(None, ''), default
='', func
=None):
5895 val
= obj
.get(field
, default
)
5896 if func
and val
not in ignore
:
5898 return template
% val
if val
not in ignore
else default
5901 def clean_podcast_url(url
):
5902 return re
.sub(r
'''(?x)
5906 media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
5909 (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
5912 cn\.co| # https://podcorn.com/analytics-prefix/
5913 st\.fm # https://podsights.com/docs/
5918 _HEX_TABLE
= '0123456789abcdef'
5921 def random_uuidv4():
5922 return re
.sub(r
'[xy]', lambda x
: _HEX_TABLE
[random
.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
5925 def make_dir(path
, to_screen
=None):
5927 dn
= os
.path
.dirname(path
)
5928 if dn
and not os
.path
.exists(dn
):
5931 except (OSError, IOError) as err
:
5932 if callable(to_screen
) is not None:
5933 to_screen('unable to create directory ' + error_to_compat_str(err
))
5937 def get_executable_path():
5938 path
= os
.path
.dirname(sys
.argv
[0])
5939 if os
.path
.abspath(sys
.argv
[0]) != os
.path
.abspath(sys
.executable
): # Not packaged
5940 path
= os
.path
.join(path
, '..')
5941 return os
.path
.abspath(path
)
5944 def load_plugins(name
, type, namespace
):
5945 plugin_info
= [None]
5948 plugin_info
= imp
.find_module(
5949 name
, [os
.path
.join(get_executable_path(), 'ytdlp_plugins')])
5950 plugins
= imp
.load_module(name
, *plugin_info
)
5951 for name
in dir(plugins
):
5952 if not name
.endswith(type):
5954 klass
= getattr(plugins
, name
)
5955 classes
.append(klass
)
5956 namespace
[name
] = klass
5960 if plugin_info
[0] is not None:
5961 plugin_info
[0].close()
5965 def traverse_dict(dictn
, keys
, casesense
=True):
5966 if not isinstance(dictn
, dict):
5970 dictn
= {key.lower(): val for key, val in dictn.items()}
5971 first_key
= first_key
.lower()
5972 value
= dictn
.get(first_key
, None)
5973 return value
if len(keys
) < 2 else traverse_dict(value
, keys
[1:], casesense
)
projects / yt-dlp.git / blob