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 # needed for sanitizing filenames in restricted mode
1719 ACCENT_CHARS
= dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
1720 itertools
.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1721 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1744 '%Y/%m/%d %H:%M:%S',
1746 '%Y-%m-%d %H:%M:%S',
1747 '%Y-%m-%d %H:%M:%S.%f',
1750 '%Y-%m-%dT%H:%M:%SZ',
1751 '%Y-%m-%dT%H:%M:%S.%fZ',
1752 '%Y-%m-%dT%H:%M:%S.%f0Z',
1753 '%Y-%m-%dT%H:%M:%S',
1754 '%Y-%m-%dT%H:%M:%S.%f',
1756 '%b %d %Y at %H:%M',
1757 '%b %d %Y at %H:%M:%S',
1758 '%B %d %Y at %H:%M',
1759 '%B %d %Y at %H:%M:%S',
1762 DATE_FORMATS_DAY_FIRST
= list(DATE_FORMATS
)
1763 DATE_FORMATS_DAY_FIRST
.extend([
1769 '%d/%m/%Y %H:%M:%S',
1772 DATE_FORMATS_MONTH_FIRST
= list(DATE_FORMATS
)
1773 DATE_FORMATS_MONTH_FIRST
.extend([
1778 '%m/%d/%Y %H:%M:%S',
1781 PACKED_CODES_RE
= r
"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1782 JSON_LD_RE
= r
'(?is)<script[^>]+type=(["\']?
)application
/ld\
+json\
1[^
>]*>(?P
<json_ld
>.+?
)</script
>'
1785 def preferredencoding():
1786 """Get preferred encoding.
1788 Returns the best encoding scheme for the system, based on
1789 locale.getpreferredencoding() and some further tweaks.
1792 pref = locale.getpreferredencoding()
1800 def write_json_file(obj, fn):
1801 """ Encode obj as JSON and write it to fn, atomically if possible """
1803 fn = encodeFilename(fn)
1804 if sys.version_info < (3, 0) and sys.platform != 'win32
':
1805 encoding = get_filesystem_encoding()
1806 # os.path.basename returns a bytes object, but NamedTemporaryFile
1807 # will fail if the filename contains non ascii characters unless we
1808 # use a unicode object
1809 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1810 # the same for os.path.dirname
1811 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1813 path_basename = os.path.basename
1814 path_dirname = os.path.dirname
1818 'prefix
': path_basename(fn) + '.',
1819 'dir': path_dirname(fn),
1823 # In Python 2.x, json.dump expects a bytestream.
1824 # In Python 3.x, it writes to a character stream
1825 if sys.version_info < (3, 0):
1830 'encoding
': 'utf
-8',
1833 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1838 if sys.platform == 'win32
':
1839 # Need to remove existing file on Windows, else os.rename raises
1840 # WindowsError or FileExistsError.
1848 os.chmod(tf.name, 0o666 & ~mask)
1851 os.rename(tf.name, fn)
1860 if sys.version_info >= (2, 7):
1861 def find_xpath_attr(node, xpath, key, val=None):
1862 """ Find the xpath xpath[@key=val] """
1863 assert re.match(r'^
[a
-zA
-Z_
-]+$
', key)
1864 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1865 return node.find(expr)
1867 def find_xpath_attr(node, xpath, key, val=None):
1868 for f in node.findall(compat_xpath(xpath)):
1869 if key not in f.attrib:
1871 if val is None or f.attrib.get(key) == val:
1875 # On python2.6 the xml.etree.ElementTree.Element methods don't support
1876 # the namespace parameter
1879 def xpath_with_ns(path
, ns_map
):
1880 components
= [c
.split(':') for c
in path
.split('/')]
1882 for c
in components
:
1884 replaced
.append(c
[0])
1887 replaced
.append('{%s}%s' % (ns_map
[ns
], tag
))
1888 return '/'.join(replaced
)
1891 def xpath_element(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1892 def _find_xpath(xpath
):
1893 return node
.find(compat_xpath(xpath
))
1895 if isinstance(xpath
, (str, compat_str
)):
1896 n
= _find_xpath(xpath
)
1904 if default
is not NO_DEFAULT
:
1907 name
= xpath
if name
is None else name
1908 raise ExtractorError('Could not find XML element %s' % name
)
1914 def xpath_text(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1915 n
= xpath_element(node
, xpath
, name
, fatal
=fatal
, default
=default
)
1916 if n
is None or n
== default
:
1919 if default
is not NO_DEFAULT
:
1922 name
= xpath
if name
is None else name
1923 raise ExtractorError('Could not find XML element\'s text %s' % name
)
1929 def xpath_attr(node
, xpath
, key
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1930 n
= find_xpath_attr(node
, xpath
, key
)
1932 if default
is not NO_DEFAULT
:
1935 name
= '%s[@%s]' % (xpath
, key
) if name
is None else name
1936 raise ExtractorError('Could not find XML attribute %s' % name
)
1939 return n
.attrib
[key
]
1942 def get_element_by_id(id, html
):
1943 """Return the content of the tag with the specified ID in the passed HTML document"""
1944 return get_element_by_attribute('id', id, html
)
1947 def get_element_by_class(class_name
, html
):
1948 """Return the content of the first tag with the specified class in the passed HTML document"""
1949 retval
= get_elements_by_class(class_name
, html
)
1950 return retval
[0] if retval
else None
1953 def get_element_by_attribute(attribute
, value
, html
, escape_value
=True):
1954 retval
= get_elements_by_attribute(attribute
, value
, html
, escape_value
)
1955 return retval
[0] if retval
else None
1958 def get_elements_by_class(class_name
, html
):
1959 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1960 return get_elements_by_attribute(
1961 'class', r
'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1962 html, escape_value=False)
1965 def get_elements_by_attribute(attribute, value, html, escape_value=True):
1966 """Return the content of the tag with the specified attribute in the passed HTML document"""
1968 value = re.escape(value) if escape_value else value
1971 for m in re.finditer(r'''(?xs)
1973 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^
']*'|
))*?
1975 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^
"]*"|
='[^']*'|))*?
1979 ''' % (re.escape(attribute), value), html):
1980 res = m.group('content
')
1982 if res.startswith('"') or res.startswith("'"):
1985 retlist.append(unescapeHTML(res))
1990 class HTMLAttributeParser(compat_HTMLParser):
1991 """Trivial HTML parser to gather the attributes for a single element"""
1995 compat_HTMLParser.__init__(self)
1997 def handle_starttag(self, tag, attrs):
1998 self.attrs = dict(attrs)
2001 def extract_attributes(html_element):
2002 """Given a string for an HTML element such as
2004 a="foo" B="bar" c="&98;az" d=boz
2005 empty= noval entity="&"
2008 Decode and return a dictionary of attributes.
2010 'a
': 'foo
', 'b
': 'bar
', c: 'baz
', d: 'boz
',
2011 'empty
': '', 'noval
': None, 'entity
': '&',
2012 'sq
': '"', 'dq': '\''
2014 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2015 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2017 parser = HTMLAttributeParser()
2019 parser.feed(html_element)
2021 # Older Python may throw HTMLParseError in case of malformed HTML
2022 except compat_HTMLParseError:
2027 def clean_html(html):
2028 """Clean an HTML snippet into a readable string"""
2030 if html is None: # Convenience for sanitizing descriptions etc.
2034 html = html.replace('\n', ' ')
2035 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2036 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2038 html = re.sub('<.*?>', '', html)
2039 # Replace html entities
2040 html = unescapeHTML(html)
2044 def sanitize_open(filename, open_mode):
2045 """Try to open the given filename, and slightly tweak it if this fails.
2047 Attempts to open the given filename. If this fails, it tries to change
2048 the filename slightly, step by step, until it's either able to open it
2049 or it fails and raises a final exception, like the standard open()
2052 It returns the tuple (stream, definitive_file_name).
2056 if sys.platform == 'win32':
2058 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2059 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2060 stream = open(encodeFilename(filename), open_mode)
2061 return (stream, filename)
2062 except (IOError, OSError) as err:
2063 if err.errno in (errno.EACCES,):
2066 # In case of error, try to remove win32 forbidden chars
2067 alt_filename = sanitize_path(filename)
2068 if alt_filename == filename:
2071 # An exception here should be caught in the caller
2072 stream = open(encodeFilename(alt_filename), open_mode)
2073 return (stream, alt_filename)
2076 def timeconvert(timestr):
2077 """Convert RFC 2822 defined time string into system timestamp"""
2079 timetuple = email.utils.parsedate_tz(timestr)
2080 if timetuple is not None:
2081 timestamp = email.utils.mktime_tz(timetuple)
2085 def sanitize_filename(s, restricted=False, is_id=False):
2086 """Sanitizes a string so it could be used as part of a filename.
2087 If restricted is set, use a stricter subset of allowed characters.
2088 Set is_id if this is not an arbitrary string, but an ID that should be kept
2091 def replace_insane(char):
2092 if restricted and char in ACCENT_CHARS:
2093 return ACCENT_CHARS[char]
2094 if char == '?' or ord(char) < 32 or ord(char) == 127:
2097 return '' if restricted else '\''
2099 return '_
-' if restricted else ' -'
2100 elif char in '\\/|
*<>':
2102 if restricted and (char in '!&\'()[]{}$
;`^
,#' or char.isspace()):
2104 if restricted
and ord(char
) > 127:
2109 s
= re
.sub(r
'[0-9]+(?::[0-9]+)+', lambda m
: m
.group(0).replace(':', '_'), s
)
2110 result
= ''.join(map(replace_insane
, s
))
2112 while '__' in result
:
2113 result
= result
.replace('__', '_')
2114 result
= result
.strip('_')
2115 # Common case of "Foreign band name - English song title"
2116 if restricted
and result
.startswith('-_'):
2118 if result
.startswith('-'):
2119 result
= '_' + result
[len('-'):]
2120 result
= result
.lstrip('.')
2126 def sanitize_path(s
):
2127 """Sanitizes and normalizes path on Windows"""
2128 if sys
.platform
!= 'win32':
2130 drive_or_unc
, _
= os
.path
.splitdrive(s
)
2131 if sys
.version_info
< (2, 7) and not drive_or_unc
:
2132 drive_or_unc
, _
= os
.path
.splitunc(s
)
2133 norm_path
= os
.path
.normpath(remove_start(s
, drive_or_unc
)).split(os
.path
.sep
)
2137 path_part
if path_part
in ['.', '..'] else re
.sub(r
'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part
)
2138 for path_part
in norm_path
]
2140 sanitized_path
.insert(0, drive_or_unc
+ os
.path
.sep
)
2141 return os
.path
.join(*sanitized_path
)
2144 def sanitize_url(url
):
2145 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2146 # the number of unwanted failures due to missing protocol
2147 if url
.startswith('//'):
2148 return 'http:%s' % url
2149 # Fix some common typos seen so far
2151 # https://github.com/ytdl-org/youtube-dl/issues/15649
2152 (r
'^httpss://', r
'https://'),
2153 # https://bx1.be/lives/direct-tv/
2154 (r
'^rmtp([es]?)://', r
'rtmp\1://'),
2156 for mistake
, fixup
in COMMON_TYPOS
:
2157 if re
.match(mistake
, url
):
2158 return re
.sub(mistake
, fixup
, url
)
2162 def sanitized_Request(url
, *args
, **kwargs
):
2163 return compat_urllib_request
.Request(sanitize_url(url
), *args
, **kwargs
)
2167 """Expand shell variables and ~"""
2168 return os
.path
.expandvars(compat_expanduser(s
))
2171 def orderedSet(iterable
):
2172 """ Remove all duplicates from the input iterable """
2180 def _htmlentity_transform(entity_with_semicolon
):
2181 """Transforms an HTML entity to a character."""
2182 entity
= entity_with_semicolon
[:-1]
2184 # Known non-numeric HTML entity
2185 if entity
in compat_html_entities
.name2codepoint
:
2186 return compat_chr(compat_html_entities
.name2codepoint
[entity
])
2188 # TODO: HTML5 allows entities without a semicolon. For example,
2189 # 'Éric' should be decoded as 'Éric'.
2190 if entity_with_semicolon
in compat_html_entities_html5
:
2191 return compat_html_entities_html5
[entity_with_semicolon
]
2193 mobj
= re
.match(r
'#(x[0-9a-fA-F]+|[0-9]+)', entity
)
2194 if mobj
is not None:
2195 numstr
= mobj
.group(1)
2196 if numstr
.startswith('x'):
2198 numstr
= '0%s' % numstr
2201 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2203 return compat_chr(int(numstr
, base
))
2207 # Unknown entity in name, return its literal representation
2208 return '&%s;' % entity
2211 def unescapeHTML(s
):
2214 assert type(s
) == compat_str
2217 r
'&([^&;]+;)', lambda m
: _htmlentity_transform(m
.group(1)), s
)
2220 def process_communicate_or_kill(p
, *args
, **kwargs
):
2222 return p
.communicate(*args
, **kwargs
)
2223 except BaseException
: # Including KeyboardInterrupt
2229 def get_subprocess_encoding():
2230 if sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2231 # For subprocess calls, encode with locale encoding
2232 # Refer to http://stackoverflow.com/a/9951851/35070
2233 encoding
= preferredencoding()
2235 encoding
= sys
.getfilesystemencoding()
2236 if encoding
is None:
2241 def encodeFilename(s
, for_subprocess
=False):
2243 @param s The name of the file
2246 assert type(s
) == compat_str
2248 # Python 3 has a Unicode API
2249 if sys
.version_info
>= (3, 0):
2252 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2253 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2254 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2255 if not for_subprocess
and sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2258 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2259 if sys
.platform
.startswith('java'):
2262 return s
.encode(get_subprocess_encoding(), 'ignore')
2265 def decodeFilename(b
, for_subprocess
=False):
2267 if sys
.version_info
>= (3, 0):
2270 if not isinstance(b
, bytes):
2273 return b
.decode(get_subprocess_encoding(), 'ignore')
2276 def encodeArgument(s
):
2277 if not isinstance(s
, compat_str
):
2278 # Legacy code that uses byte strings
2279 # Uncomment the following line after fixing all post processors
2280 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2281 s
= s
.decode('ascii')
2282 return encodeFilename(s
, True)
2285 def decodeArgument(b
):
2286 return decodeFilename(b
, True)
2289 def decodeOption(optval
):
2292 if isinstance(optval
, bytes):
2293 optval
= optval
.decode(preferredencoding())
2295 assert isinstance(optval
, compat_str
)
2299 def formatSeconds(secs
, delim
=':'):
2301 return '%d%s%02d%s%02d' % (secs
// 3600, delim
, (secs
% 3600) // 60, delim
, secs
% 60)
2303 return '%d%s%02d' % (secs
// 60, delim
, secs
% 60)
2308 def make_HTTPS_handler(params
, **kwargs
):
2309 opts_no_check_certificate
= params
.get('nocheckcertificate', False)
2310 if hasattr(ssl
, 'create_default_context'): # Python >= 3.4 or 2.7.9
2311 context
= ssl
.create_default_context(ssl
.Purpose
.SERVER_AUTH
)
2312 if opts_no_check_certificate
:
2313 context
.check_hostname
= False
2314 context
.verify_mode
= ssl
.CERT_NONE
2316 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2319 # (create_default_context present but HTTPSHandler has no context=)
2322 if sys
.version_info
< (3, 2):
2323 return YoutubeDLHTTPSHandler(params
, **kwargs
)
2324 else: # Python < 3.4
2325 context
= ssl
.SSLContext(ssl
.PROTOCOL_TLSv1
)
2326 context
.verify_mode
= (ssl
.CERT_NONE
2327 if opts_no_check_certificate
2328 else ssl
.CERT_REQUIRED
)
2329 context
.set_default_verify_paths()
2330 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2333 def bug_reports_message():
2334 if ytdl_is_updateable():
2335 update_cmd
= 'type youtube-dlc -U to update'
2337 update_cmd
= 'see https://github.com/pukkandan/yt-dlp on how to update'
2338 msg
= '; please report this issue on https://github.com/pukkandan/yt-dlp .'
2339 msg
+= ' Make sure you are using the latest version; %s.' % update_cmd
2340 msg
+= ' Be sure to call youtube-dlc with the --verbose flag and include its complete output.'
2344 class YoutubeDLError(Exception):
2345 """Base exception for YoutubeDL errors."""
2349 class ExtractorError(YoutubeDLError
):
2350 """Error during info extraction."""
2352 def __init__(self
, msg
, tb
=None, expected
=False, cause
=None, video_id
=None):
2353 """ tb, if given, is the original traceback (so that it can be printed out).
2354 If expected is set, this is a normal error message and most likely not a bug in youtube-dlc.
2357 if sys
.exc_info()[0] in (compat_urllib_error
.URLError
, socket
.timeout
, UnavailableVideoError
):
2359 if video_id
is not None:
2360 msg
= video_id
+ ': ' + msg
2362 msg
+= ' (caused by %r)' % cause
2364 msg
+= bug_reports_message()
2365 super(ExtractorError
, self
).__init
__(msg
)
2368 self
.exc_info
= sys
.exc_info() # preserve original exception
2370 self
.video_id
= video_id
2372 def format_traceback(self
):
2373 if self
.traceback
is None:
2375 return ''.join(traceback
.format_tb(self
.traceback
))
2378 class UnsupportedError(ExtractorError
):
2379 def __init__(self
, url
):
2380 super(UnsupportedError
, self
).__init
__(
2381 'Unsupported URL: %s' % url
, expected
=True)
2385 class RegexNotFoundError(ExtractorError
):
2386 """Error when a regex didn't match"""
2390 class GeoRestrictedError(ExtractorError
):
2391 """Geographic restriction Error exception.
2393 This exception may be thrown when a video is not available from your
2394 geographic location due to geographic restrictions imposed by a website.
2397 def __init__(self
, msg
, countries
=None):
2398 super(GeoRestrictedError
, self
).__init
__(msg
, expected
=True)
2400 self
.countries
= countries
2403 class DownloadError(YoutubeDLError
):
2404 """Download Error exception.
2406 This exception may be thrown by FileDownloader objects if they are not
2407 configured to continue on errors. They will contain the appropriate
2411 def __init__(self
, msg
, exc_info
=None):
2412 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2413 super(DownloadError
, self
).__init
__(msg
)
2414 self
.exc_info
= exc_info
2417 class SameFileError(YoutubeDLError
):
2418 """Same File exception.
2420 This exception will be thrown by FileDownloader objects if they detect
2421 multiple files would have to be downloaded to the same file on disk.
2426 class PostProcessingError(YoutubeDLError
):
2427 """Post Processing exception.
2429 This exception may be raised by PostProcessor's .run() method to
2430 indicate an error in the postprocessing task.
2433 def __init__(self
, msg
):
2434 super(PostProcessingError
, self
).__init
__(msg
)
2438 class ExistingVideoReached(YoutubeDLError
):
2439 """ --max-downloads limit has been reached. """
2443 class RejectedVideoReached(YoutubeDLError
):
2444 """ --max-downloads limit has been reached. """
2448 class MaxDownloadsReached(YoutubeDLError
):
2449 """ --max-downloads limit has been reached. """
2453 class UnavailableVideoError(YoutubeDLError
):
2454 """Unavailable Format exception.
2456 This exception will be thrown when a video is requested
2457 in a format that is not available for that video.
2462 class ContentTooShortError(YoutubeDLError
):
2463 """Content Too Short exception.
2465 This exception may be raised by FileDownloader objects when a file they
2466 download is too small for what the server announced first, indicating
2467 the connection was probably interrupted.
2470 def __init__(self
, downloaded
, expected
):
2471 super(ContentTooShortError
, self
).__init
__(
2472 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded
, expected
)
2475 self
.downloaded
= downloaded
2476 self
.expected
= expected
2479 class XAttrMetadataError(YoutubeDLError
):
2480 def __init__(self
, code
=None, msg
='Unknown error'):
2481 super(XAttrMetadataError
, self
).__init
__(msg
)
2485 # Parsing code and msg
2486 if (self
.code
in (errno
.ENOSPC
, errno
.EDQUOT
)
2487 or 'No space left' in self
.msg
or 'Disk quota exceeded' in self
.msg
):
2488 self
.reason
= 'NO_SPACE'
2489 elif self
.code
== errno
.E2BIG
or 'Argument list too long' in self
.msg
:
2490 self
.reason
= 'VALUE_TOO_LONG'
2492 self
.reason
= 'NOT_SUPPORTED'
2495 class XAttrUnavailableError(YoutubeDLError
):
2499 def _create_http_connection(ydl_handler
, http_class
, is_https
, *args
, **kwargs
):
2500 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2501 # expected HTTP responses to meet HTTP/1.0 or later (see also
2502 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2503 if sys
.version_info
< (3, 0):
2504 kwargs
['strict'] = True
2505 hc
= http_class(*args
, **compat_kwargs(kwargs
))
2506 source_address
= ydl_handler
._params
.get('source_address')
2508 if source_address
is not None:
2509 # This is to workaround _create_connection() from socket where it will try all
2510 # address data from getaddrinfo() including IPv6. This filters the result from
2511 # getaddrinfo() based on the source_address value.
2512 # This is based on the cpython socket.create_connection() function.
2513 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2514 def _create_connection(address
, timeout
=socket
._GLOBAL
_DEFAULT
_TIMEOUT
, source_address
=None):
2515 host
, port
= address
2517 addrs
= socket
.getaddrinfo(host
, port
, 0, socket
.SOCK_STREAM
)
2518 af
= socket
.AF_INET
if '.' in source_address
[0] else socket
.AF_INET6
2519 ip_addrs
= [addr
for addr
in addrs
if addr
[0] == af
]
2520 if addrs
and not ip_addrs
:
2521 ip_version
= 'v4' if af
== socket
.AF_INET
else 'v6'
2523 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2524 % (ip_version
, source_address
[0]))
2525 for res
in ip_addrs
:
2526 af
, socktype
, proto
, canonname
, sa
= res
2529 sock
= socket
.socket(af
, socktype
, proto
)
2530 if timeout
is not socket
._GLOBAL
_DEFAULT
_TIMEOUT
:
2531 sock
.settimeout(timeout
)
2532 sock
.bind(source_address
)
2534 err
= None # Explicitly break reference cycle
2536 except socket
.error
as _
:
2538 if sock
is not None:
2543 raise socket
.error('getaddrinfo returns an empty list')
2544 if hasattr(hc
, '_create_connection'):
2545 hc
._create
_connection
= _create_connection
2546 sa
= (source_address
, 0)
2547 if hasattr(hc
, 'source_address'): # Python 2.7+
2548 hc
.source_address
= sa
2550 def _hc_connect(self
, *args
, **kwargs
):
2551 sock
= _create_connection(
2552 (self
.host
, self
.port
), self
.timeout
, sa
)
2554 self
.sock
= ssl
.wrap_socket(
2555 sock
, self
.key_file
, self
.cert_file
,
2556 ssl_version
=ssl
.PROTOCOL_TLSv1
)
2559 hc
.connect
= functools
.partial(_hc_connect
, hc
)
2564 def handle_youtubedl_headers(headers
):
2565 filtered_headers
= headers
2567 if 'Youtubedl-no-compression' in filtered_headers
:
2568 filtered_headers
= dict((k
, v
) for k
, v
in filtered_headers
.items() if k
.lower() != 'accept-encoding')
2569 del filtered_headers
['Youtubedl-no-compression']
2571 return filtered_headers
2574 class YoutubeDLHandler(compat_urllib_request
.HTTPHandler
):
2575 """Handler for HTTP requests and responses.
2577 This class, when installed with an OpenerDirector, automatically adds
2578 the standard headers to every HTTP request and handles gzipped and
2579 deflated responses from web servers. If compression is to be avoided in
2580 a particular request, the original request in the program code only has
2581 to include the HTTP header "Youtubedl-no-compression", which will be
2582 removed before making the real request.
2584 Part of this code was copied from:
2586 http://techknack.net/python-urllib2-handlers/
2588 Andrew Rowls, the author of that code, agreed to release it to the
2592 def __init__(self
, params
, *args
, **kwargs
):
2593 compat_urllib_request
.HTTPHandler
.__init
__(self
, *args
, **kwargs
)
2594 self
._params
= params
2596 def http_open(self
, req
):
2597 conn_class
= compat_http_client
.HTTPConnection
2599 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2601 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2602 del req
.headers
['Ytdl-socks-proxy']
2604 return self
.do_open(functools
.partial(
2605 _create_http_connection
, self
, conn_class
, False),
2611 return zlib
.decompress(data
, -zlib
.MAX_WBITS
)
2613 return zlib
.decompress(data
)
2615 def http_request(self
, req
):
2616 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2617 # always respected by websites, some tend to give out URLs with non percent-encoded
2618 # non-ASCII characters (see telemb.py, ard.py [#3412])
2619 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2620 # To work around aforementioned issue we will replace request's original URL with
2621 # percent-encoded one
2622 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2623 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2624 url
= req
.get_full_url()
2625 url_escaped
= escape_url(url
)
2627 # Substitute URL if any change after escaping
2628 if url
!= url_escaped
:
2629 req
= update_Request(req
, url
=url_escaped
)
2631 for h
, v
in std_headers
.items():
2632 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2633 # The dict keys are capitalized because of this bug by urllib
2634 if h
.capitalize() not in req
.headers
:
2635 req
.add_header(h
, v
)
2637 req
.headers
= handle_youtubedl_headers(req
.headers
)
2639 if sys
.version_info
< (2, 7) and '#' in req
.get_full_url():
2640 # Python 2.6 is brain-dead when it comes to fragments
2641 req
._Request
__original
= req
._Request
__original
.partition('#')[0]
2642 req
._Request
__r
_type
= req
._Request
__r
_type
.partition('#')[0]
2646 def http_response(self
, req
, resp
):
2649 if resp
.headers
.get('Content-encoding', '') == 'gzip':
2650 content
= resp
.read()
2651 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
), mode
='rb')
2653 uncompressed
= io
.BytesIO(gz
.read())
2654 except IOError as original_ioerror
:
2655 # There may be junk add the end of the file
2656 # See http://stackoverflow.com/q/4928560/35070 for details
2657 for i
in range(1, 1024):
2659 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
[:-i
]), mode
='rb')
2660 uncompressed
= io
.BytesIO(gz
.read())
2665 raise original_ioerror
2666 resp
= compat_urllib_request
.addinfourl(uncompressed
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2667 resp
.msg
= old_resp
.msg
2668 del resp
.headers
['Content-encoding']
2670 if resp
.headers
.get('Content-encoding', '') == 'deflate':
2671 gz
= io
.BytesIO(self
.deflate(resp
.read()))
2672 resp
= compat_urllib_request
.addinfourl(gz
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2673 resp
.msg
= old_resp
.msg
2674 del resp
.headers
['Content-encoding']
2675 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2676 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2677 if 300 <= resp
.code
< 400:
2678 location
= resp
.headers
.get('Location')
2680 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2681 if sys
.version_info
>= (3, 0):
2682 location
= location
.encode('iso-8859-1').decode('utf-8')
2684 location
= location
.decode('utf-8')
2685 location_escaped
= escape_url(location
)
2686 if location
!= location_escaped
:
2687 del resp
.headers
['Location']
2688 if sys
.version_info
< (3, 0):
2689 location_escaped
= location_escaped
.encode('utf-8')
2690 resp
.headers
['Location'] = location_escaped
2693 https_request
= http_request
2694 https_response
= http_response
2697 def make_socks_conn_class(base_class
, socks_proxy
):
2698 assert issubclass(base_class
, (
2699 compat_http_client
.HTTPConnection
, compat_http_client
.HTTPSConnection
))
2701 url_components
= compat_urlparse
.urlparse(socks_proxy
)
2702 if url_components
.scheme
.lower() == 'socks5':
2703 socks_type
= ProxyType
.SOCKS5
2704 elif url_components
.scheme
.lower() in ('socks', 'socks4'):
2705 socks_type
= ProxyType
.SOCKS4
2706 elif url_components
.scheme
.lower() == 'socks4a':
2707 socks_type
= ProxyType
.SOCKS4A
2709 def unquote_if_non_empty(s
):
2712 return compat_urllib_parse_unquote_plus(s
)
2716 url_components
.hostname
, url_components
.port
or 1080,
2718 unquote_if_non_empty(url_components
.username
),
2719 unquote_if_non_empty(url_components
.password
),
2722 class SocksConnection(base_class
):
2724 self
.sock
= sockssocket()
2725 self
.sock
.setproxy(*proxy_args
)
2726 if type(self
.timeout
) in (int, float):
2727 self
.sock
.settimeout(self
.timeout
)
2728 self
.sock
.connect((self
.host
, self
.port
))
2730 if isinstance(self
, compat_http_client
.HTTPSConnection
):
2731 if hasattr(self
, '_context'): # Python > 2.6
2732 self
.sock
= self
._context
.wrap_socket(
2733 self
.sock
, server_hostname
=self
.host
)
2735 self
.sock
= ssl
.wrap_socket(self
.sock
)
2737 return SocksConnection
2740 class YoutubeDLHTTPSHandler(compat_urllib_request
.HTTPSHandler
):
2741 def __init__(self
, params
, https_conn_class
=None, *args
, **kwargs
):
2742 compat_urllib_request
.HTTPSHandler
.__init
__(self
, *args
, **kwargs
)
2743 self
._https
_conn
_class
= https_conn_class
or compat_http_client
.HTTPSConnection
2744 self
._params
= params
2746 def https_open(self
, req
):
2748 conn_class
= self
._https
_conn
_class
2750 if hasattr(self
, '_context'): # python > 2.6
2751 kwargs
['context'] = self
._context
2752 if hasattr(self
, '_check_hostname'): # python 3.x
2753 kwargs
['check_hostname'] = self
._check
_hostname
2755 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2757 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2758 del req
.headers
['Ytdl-socks-proxy']
2760 return self
.do_open(functools
.partial(
2761 _create_http_connection
, self
, conn_class
, True),
2765 class YoutubeDLCookieJar(compat_cookiejar
.MozillaCookieJar
):
2767 See [1] for cookie file format.
2769 1. https://curl.haxx.se/docs/http-cookies.html
2771 _HTTPONLY_PREFIX
= '#HttpOnly_'
2773 _HEADER
= '''# Netscape HTTP Cookie File
2774 # This file is generated by youtube-dlc. Do not edit.
2777 _CookieFileEntry
= collections
.namedtuple(
2779 ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
2781 def save(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2783 Save cookies to a file.
2785 Most of the code is taken from CPython 3.8 and slightly adapted
2786 to support cookie files with UTF-8 in both python 2 and 3.
2788 if filename
is None:
2789 if self
.filename
is not None:
2790 filename
= self
.filename
2792 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2794 # Store session cookies with `expires` set to 0 instead of an empty
2797 if cookie
.expires
is None:
2800 with io
.open(filename
, 'w', encoding
='utf-8') as f
:
2801 f
.write(self
._HEADER
)
2804 if not ignore_discard
and cookie
.discard
:
2806 if not ignore_expires
and cookie
.is_expired(now
):
2812 if cookie
.domain
.startswith('.'):
2813 initial_dot
= 'TRUE'
2815 initial_dot
= 'FALSE'
2816 if cookie
.expires
is not None:
2817 expires
= compat_str(cookie
.expires
)
2820 if cookie
.value
is None:
2821 # cookies.txt regards 'Set-Cookie: foo' as a cookie
2822 # with no name, whereas http.cookiejar regards it as a
2823 # cookie with no value.
2828 value
= cookie
.value
2830 '\t'.join([cookie
.domain
, initial_dot
, cookie
.path
,
2831 secure
, expires
, name
, value
]) + '\n')
2833 def load(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2834 """Load cookies from a file."""
2835 if filename
is None:
2836 if self
.filename
is not None:
2837 filename
= self
.filename
2839 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2841 def prepare_line(line
):
2842 if line
.startswith(self
._HTTPONLY
_PREFIX
):
2843 line
= line
[len(self
._HTTPONLY
_PREFIX
):]
2844 # comments and empty lines are fine
2845 if line
.startswith('#') or not line
.strip():
2847 cookie_list
= line
.split('\t')
2848 if len(cookie_list
) != self
._ENTRY
_LEN
:
2849 raise compat_cookiejar
.LoadError('invalid length %d' % len(cookie_list
))
2850 cookie
= self
._CookieFileEntry
(*cookie_list
)
2851 if cookie
.expires_at
and not cookie
.expires_at
.isdigit():
2852 raise compat_cookiejar
.LoadError('invalid expires at %s' % cookie
.expires_at
)
2856 with io
.open(filename
, encoding
='utf-8') as f
:
2859 cf
.write(prepare_line(line
))
2860 except compat_cookiejar
.LoadError
as e
:
2862 'WARNING: skipping cookie file entry due to %s: %r\n'
2863 % (e
, line
), sys
.stderr
)
2866 self
._really
_load
(cf
, filename
, ignore_discard
, ignore_expires
)
2867 # Session cookies are denoted by either `expires` field set to
2868 # an empty string or 0. MozillaCookieJar only recognizes the former
2869 # (see [1]). So we need force the latter to be recognized as session
2870 # cookies on our own.
2871 # Session cookies may be important for cookies-based authentication,
2872 # e.g. usually, when user does not check 'Remember me' check box while
2873 # logging in on a site, some important cookies are stored as session
2874 # cookies so that not recognizing them will result in failed login.
2875 # 1. https://bugs.python.org/issue17164
2877 # Treat `expires=0` cookies as session cookies
2878 if cookie
.expires
== 0:
2879 cookie
.expires
= None
2880 cookie
.discard
= True
2883 class YoutubeDLCookieProcessor(compat_urllib_request
.HTTPCookieProcessor
):
2884 def __init__(self
, cookiejar
=None):
2885 compat_urllib_request
.HTTPCookieProcessor
.__init
__(self
, cookiejar
)
2887 def http_response(self
, request
, response
):
2888 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2889 # characters in Set-Cookie HTTP header of last response (see
2890 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2891 # In order to at least prevent crashing we will percent encode Set-Cookie
2892 # header before HTTPCookieProcessor starts processing it.
2893 # if sys.version_info < (3, 0) and response.headers:
2894 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2895 # set_cookie = response.headers.get(set_cookie_header)
2897 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2898 # if set_cookie != set_cookie_escaped:
2899 # del response.headers[set_cookie_header]
2900 # response.headers[set_cookie_header] = set_cookie_escaped
2901 return compat_urllib_request
.HTTPCookieProcessor
.http_response(self
, request
, response
)
2903 https_request
= compat_urllib_request
.HTTPCookieProcessor
.http_request
2904 https_response
= http_response
2907 class YoutubeDLRedirectHandler(compat_urllib_request
.HTTPRedirectHandler
):
2908 if sys
.version_info
[0] < 3:
2909 def redirect_request(self
, req
, fp
, code
, msg
, headers
, newurl
):
2910 # On python 2 urlh.geturl() may sometimes return redirect URL
2911 # as byte string instead of unicode. This workaround allows
2912 # to force it always return unicode.
2913 return compat_urllib_request
.HTTPRedirectHandler
.redirect_request(self
, req
, fp
, code
, msg
, headers
, compat_str(newurl
))
2916 def extract_timezone(date_str
):
2918 r
'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
2921 timezone
= datetime
.timedelta()
2923 date_str
= date_str
[:-len(m
.group('tz'))]
2924 if not m
.group('sign'):
2925 timezone
= datetime
.timedelta()
2927 sign
= 1 if m
.group('sign') == '+' else -1
2928 timezone
= datetime
.timedelta(
2929 hours
=sign
* int(m
.group('hours')),
2930 minutes
=sign
* int(m
.group('minutes')))
2931 return timezone
, date_str
2934 def parse_iso8601(date_str
, delimiter
='T', timezone
=None):
2935 """ Return a UNIX timestamp from the given date """
2937 if date_str
is None:
2940 date_str
= re
.sub(r
'\.[0-9]+', '', date_str
)
2942 if timezone
is None:
2943 timezone
, date_str
= extract_timezone(date_str
)
2946 date_format
= '%Y-%m-%d{0}%H:%M:%S'.format(delimiter
)
2947 dt
= datetime
.datetime
.strptime(date_str
, date_format
) - timezone
2948 return calendar
.timegm(dt
.timetuple())
2953 def date_formats(day_first
=True):
2954 return DATE_FORMATS_DAY_FIRST
if day_first
else DATE_FORMATS_MONTH_FIRST
2957 def unified_strdate(date_str
, day_first
=True):
2958 """Return a string with the date in the format YYYYMMDD"""
2960 if date_str
is None:
2964 date_str
= date_str
.replace(',', ' ')
2965 # Remove AM/PM + timezone
2966 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
2967 _
, date_str
= extract_timezone(date_str
)
2969 for expression
in date_formats(day_first
):
2971 upload_date
= datetime
.datetime
.strptime(date_str
, expression
).strftime('%Y%m%d')
2974 if upload_date
is None:
2975 timetuple
= email
.utils
.parsedate_tz(date_str
)
2978 upload_date
= datetime
.datetime(*timetuple
[:6]).strftime('%Y%m%d')
2981 if upload_date
is not None:
2982 return compat_str(upload_date
)
2985 def unified_timestamp(date_str
, day_first
=True):
2986 if date_str
is None:
2989 date_str
= re
.sub(r
'[,|]', '', date_str
)
2991 pm_delta
= 12 if re
.search(r
'(?i)PM', date_str
) else 0
2992 timezone
, date_str
= extract_timezone(date_str
)
2994 # Remove AM/PM + timezone
2995 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
2997 # Remove unrecognized timezones from ISO 8601 alike timestamps
2998 m
= re
.search(r
'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str
)
3000 date_str
= date_str
[:-len(m
.group('tz'))]
3002 # Python only supports microseconds, so remove nanoseconds
3003 m
= re
.search(r
'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str
)
3005 date_str
= m
.group(1)
3007 for expression
in date_formats(day_first
):
3009 dt
= datetime
.datetime
.strptime(date_str
, expression
) - timezone
+ datetime
.timedelta(hours
=pm_delta
)
3010 return calendar
.timegm(dt
.timetuple())
3013 timetuple
= email
.utils
.parsedate_tz(date_str
)
3015 return calendar
.timegm(timetuple
) + pm_delta
* 3600
3018 def determine_ext(url
, default_ext
='unknown_video'):
3019 if url
is None or '.' not in url
:
3021 guess
= url
.partition('?')[0].rpartition('.')[2]
3022 if re
.match(r
'^[A-Za-z0-9]+$', guess
):
3024 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
3025 elif guess
.rstrip('/') in KNOWN_EXTENSIONS
:
3026 return guess
.rstrip('/')
3031 def subtitles_filename(filename
, sub_lang
, sub_format
, expected_real_ext
=None):
3032 return replace_extension(filename
, sub_lang
+ '.' + sub_format
, expected_real_ext
)
3035 def date_from_str(date_str
):
3037 Return a datetime object from a string in the format YYYYMMDD or
3038 (now|today)[+-][0-9](day|week|month|year)(s)?"""
3039 today
= datetime
.date
.today()
3040 if date_str
in ('now', 'today'):
3042 if date_str
== 'yesterday':
3043 return today
- datetime
.timedelta(days
=1)
3044 match
= re
.match(r
'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str
)
3045 if match
is not None:
3046 sign
= match
.group('sign')
3047 time
= int(match
.group('time'))
3050 unit
= match
.group('unit')
3051 # A bad approximation?
3055 elif unit
== 'year':
3059 delta
= datetime
.timedelta(**{unit: time}
)
3060 return today
+ delta
3061 return datetime
.datetime
.strptime(date_str
, '%Y%m%d').date()
3064 def hyphenate_date(date_str
):
3066 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
3067 match
= re
.match(r
'^(\d\d\d\d)(\d\d)(\d\d)$', date_str
)
3068 if match
is not None:
3069 return '-'.join(match
.groups())
3074 class DateRange(object):
3075 """Represents a time interval between two dates"""
3077 def __init__(self
, start
=None, end
=None):
3078 """start and end must be strings in the format accepted by date"""
3079 if start
is not None:
3080 self
.start
= date_from_str(start
)
3082 self
.start
= datetime
.datetime
.min.date()
3084 self
.end
= date_from_str(end
)
3086 self
.end
= datetime
.datetime
.max.date()
3087 if self
.start
> self
.end
:
3088 raise ValueError('Date range: "%s" , the start date must be before the end date' % self
)
3092 """Returns a range that only contains the given day"""
3093 return cls(day
, day
)
3095 def __contains__(self
, date
):
3096 """Check if the date is in the range"""
3097 if not isinstance(date
, datetime
.date
):
3098 date
= date_from_str(date
)
3099 return self
.start
<= date
<= self
.end
3102 return '%s - %s' % (self
.start
.isoformat(), self
.end
.isoformat())
3105 def platform_name():
3106 """ Returns the platform name as a compat_str """
3107 res
= platform
.platform()
3108 if isinstance(res
, bytes):
3109 res
= res
.decode(preferredencoding())
3111 assert isinstance(res
, compat_str
)
3115 def _windows_write_string(s
, out
):
3116 """ Returns True if the string was written using special methods,
3117 False if it has yet to be written out."""
3118 # Adapted from http://stackoverflow.com/a/3259271/35070
3121 import ctypes
.wintypes
3129 fileno
= out
.fileno()
3130 except AttributeError:
3131 # If the output stream doesn't have a fileno, it's virtual
3133 except io
.UnsupportedOperation
:
3134 # Some strange Windows pseudo files?
3136 if fileno
not in WIN_OUTPUT_IDS
:
3139 GetStdHandle
= compat_ctypes_WINFUNCTYPE(
3140 ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.DWORD
)(
3141 ('GetStdHandle', ctypes
.windll
.kernel32
))
3142 h
= GetStdHandle(WIN_OUTPUT_IDS
[fileno
])
3144 WriteConsoleW
= compat_ctypes_WINFUNCTYPE(
3145 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.LPWSTR
,
3146 ctypes
.wintypes
.DWORD
, ctypes
.POINTER(ctypes
.wintypes
.DWORD
),
3147 ctypes
.wintypes
.LPVOID
)(('WriteConsoleW', ctypes
.windll
.kernel32
))
3148 written
= ctypes
.wintypes
.DWORD(0)
3150 GetFileType
= compat_ctypes_WINFUNCTYPE(ctypes
.wintypes
.DWORD
, ctypes
.wintypes
.DWORD
)(('GetFileType', ctypes
.windll
.kernel32
))
3151 FILE_TYPE_CHAR
= 0x0002
3152 FILE_TYPE_REMOTE
= 0x8000
3153 GetConsoleMode
= compat_ctypes_WINFUNCTYPE(
3154 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
,
3155 ctypes
.POINTER(ctypes
.wintypes
.DWORD
))(
3156 ('GetConsoleMode', ctypes
.windll
.kernel32
))
3157 INVALID_HANDLE_VALUE
= ctypes
.wintypes
.DWORD(-1).value
3159 def not_a_console(handle
):
3160 if handle
== INVALID_HANDLE_VALUE
or handle
is None:
3162 return ((GetFileType(handle
) & ~FILE_TYPE_REMOTE
) != FILE_TYPE_CHAR
3163 or GetConsoleMode(handle
, ctypes
.byref(ctypes
.wintypes
.DWORD())) == 0)
3165 if not_a_console(h
):
3168 def next_nonbmp_pos(s
):
3170 return next(i
for i
, c
in enumerate(s
) if ord(c
) > 0xffff)
3171 except StopIteration:
3175 count
= min(next_nonbmp_pos(s
), 1024)
3177 ret
= WriteConsoleW(
3178 h
, s
, count
if count
else 2, ctypes
.byref(written
), None)
3180 raise OSError('Failed to write string')
3181 if not count
: # We just wrote a non-BMP character
3182 assert written
.value
== 2
3185 assert written
.value
> 0
3186 s
= s
[written
.value
:]
3190 def write_string(s
, out
=None, encoding
=None):
3193 assert type(s
) == compat_str
3195 if sys
.platform
== 'win32' and encoding
is None and hasattr(out
, 'fileno'):
3196 if _windows_write_string(s
, out
):
3199 if ('b' in getattr(out
, 'mode', '')
3200 or sys
.version_info
[0] < 3): # Python 2 lies about mode of sys.stderr
3201 byt
= s
.encode(encoding
or preferredencoding(), 'ignore')
3203 elif hasattr(out
, 'buffer'):
3204 enc
= encoding
or getattr(out
, 'encoding', None) or preferredencoding()
3205 byt
= s
.encode(enc
, 'ignore')
3206 out
.buffer.write(byt
)
3212 def bytes_to_intlist(bs
):
3215 if isinstance(bs
[0], int): # Python 3
3218 return [ord(c
) for c
in bs
]
3221 def intlist_to_bytes(xs
):
3224 return compat_struct_pack('%dB' % len(xs
), *xs
)
3227 # Cross-platform file locking
3228 if sys
.platform
== 'win32':
3229 import ctypes
.wintypes
3232 class OVERLAPPED(ctypes
.Structure
):
3234 ('Internal', ctypes
.wintypes
.LPVOID
),
3235 ('InternalHigh', ctypes
.wintypes
.LPVOID
),
3236 ('Offset', ctypes
.wintypes
.DWORD
),
3237 ('OffsetHigh', ctypes
.wintypes
.DWORD
),
3238 ('hEvent', ctypes
.wintypes
.HANDLE
),
3241 kernel32
= ctypes
.windll
.kernel32
3242 LockFileEx
= kernel32
.LockFileEx
3243 LockFileEx
.argtypes
= [
3244 ctypes
.wintypes
.HANDLE
, # hFile
3245 ctypes
.wintypes
.DWORD
, # dwFlags
3246 ctypes
.wintypes
.DWORD
, # dwReserved
3247 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3248 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3249 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3251 LockFileEx
.restype
= ctypes
.wintypes
.BOOL
3252 UnlockFileEx
= kernel32
.UnlockFileEx
3253 UnlockFileEx
.argtypes
= [
3254 ctypes
.wintypes
.HANDLE
, # hFile
3255 ctypes
.wintypes
.DWORD
, # dwReserved
3256 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3257 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3258 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3260 UnlockFileEx
.restype
= ctypes
.wintypes
.BOOL
3261 whole_low
= 0xffffffff
3262 whole_high
= 0x7fffffff
3264 def _lock_file(f
, exclusive
):
3265 overlapped
= OVERLAPPED()
3266 overlapped
.Offset
= 0
3267 overlapped
.OffsetHigh
= 0
3268 overlapped
.hEvent
= 0
3269 f
._lock
_file
_overlapped
_p
= ctypes
.pointer(overlapped
)
3270 handle
= msvcrt
.get_osfhandle(f
.fileno())
3271 if not LockFileEx(handle
, 0x2 if exclusive
else 0x0, 0,
3272 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3273 raise OSError('Locking file failed: %r' % ctypes
.FormatError())
3275 def _unlock_file(f
):
3276 assert f
._lock
_file
_overlapped
_p
3277 handle
= msvcrt
.get_osfhandle(f
.fileno())
3278 if not UnlockFileEx(handle
, 0,
3279 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3280 raise OSError('Unlocking file failed: %r' % ctypes
.FormatError())
3283 # Some platforms, such as Jython, is missing fcntl
3287 def _lock_file(f
, exclusive
):
3288 fcntl
.flock(f
, fcntl
.LOCK_EX
if exclusive
else fcntl
.LOCK_SH
)
3290 def _unlock_file(f
):
3291 fcntl
.flock(f
, fcntl
.LOCK_UN
)
3293 UNSUPPORTED_MSG
= 'file locking is not supported on this platform'
3295 def _lock_file(f
, exclusive
):
3296 raise IOError(UNSUPPORTED_MSG
)
3298 def _unlock_file(f
):
3299 raise IOError(UNSUPPORTED_MSG
)
3302 class locked_file(object):
3303 def __init__(self
, filename
, mode
, encoding
=None):
3304 assert mode
in ['r', 'a', 'w']
3305 self
.f
= io
.open(filename
, mode
, encoding
=encoding
)
3308 def __enter__(self
):
3309 exclusive
= self
.mode
!= 'r'
3311 _lock_file(self
.f
, exclusive
)
3317 def __exit__(self
, etype
, value
, traceback
):
3319 _unlock_file(self
.f
)
3326 def write(self
, *args
):
3327 return self
.f
.write(*args
)
3329 def read(self
, *args
):
3330 return self
.f
.read(*args
)
3333 def get_filesystem_encoding():
3334 encoding
= sys
.getfilesystemencoding()
3335 return encoding
if encoding
is not None else 'utf-8'
3338 def shell_quote(args
):
3340 encoding
= get_filesystem_encoding()
3342 if isinstance(a
, bytes):
3343 # We may get a filename encoded with 'encodeFilename'
3344 a
= a
.decode(encoding
)
3345 quoted_args
.append(compat_shlex_quote(a
))
3346 return ' '.join(quoted_args
)
3349 def smuggle_url(url
, data
):
3350 """ Pass additional data in a URL for internal use. """
3352 url
, idata
= unsmuggle_url(url
, {})
3354 sdata
= compat_urllib_parse_urlencode(
3355 {'__youtubedl_smuggle': json.dumps(data)}
)
3356 return url
+ '#' + sdata
3359 def unsmuggle_url(smug_url
, default
=None):
3360 if '#__youtubedl_smuggle' not in smug_url
:
3361 return smug_url
, default
3362 url
, _
, sdata
= smug_url
.rpartition('#')
3363 jsond
= compat_parse_qs(sdata
)['__youtubedl_smuggle'][0]
3364 data
= json
.loads(jsond
)
3368 def format_bytes(bytes):
3371 if type(bytes) is str:
3372 bytes = float(bytes)
3376 exponent
= int(math
.log(bytes, 1024.0))
3377 suffix
= ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent
]
3378 converted
= float(bytes) / float(1024 ** exponent
)
3379 return '%.2f%s' % (converted
, suffix
)
3382 def lookup_unit_table(unit_table
, s
):
3383 units_re
= '|'.join(re
.escape(u
) for u
in unit_table
)
3385 r
'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re
, s
)
3388 num_str
= m
.group('num').replace(',', '.')
3389 mult
= unit_table
[m
.group('unit')]
3390 return int(float(num_str
) * mult
)
3393 def parse_filesize(s
):
3397 # The lower-case forms are of course incorrect and unofficial,
3398 # but we support those too
3415 'megabytes': 1000 ** 2,
3416 'mebibytes': 1024 ** 2,
3422 'gigabytes': 1000 ** 3,
3423 'gibibytes': 1024 ** 3,
3429 'terabytes': 1000 ** 4,
3430 'tebibytes': 1024 ** 4,
3436 'petabytes': 1000 ** 5,
3437 'pebibytes': 1024 ** 5,
3443 'exabytes': 1000 ** 6,
3444 'exbibytes': 1024 ** 6,
3450 'zettabytes': 1000 ** 7,
3451 'zebibytes': 1024 ** 7,
3457 'yottabytes': 1000 ** 8,
3458 'yobibytes': 1024 ** 8,
3461 return lookup_unit_table(_UNIT_TABLE
, s
)
3470 if re
.match(r
'^[\d,.]+$', s
):
3471 return str_to_int(s
)
3482 return lookup_unit_table(_UNIT_TABLE
, s
)
3485 def parse_resolution(s
):
3489 mobj
= re
.search(r
'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s
)
3492 'width': int(mobj
.group('w')),
3493 'height': int(mobj
.group('h')),
3496 mobj
= re
.search(r
'\b(\d+)[pPiI]\b', s
)
3498 return {'height': int(mobj.group(1))}
3500 mobj
= re
.search(r
'\b([48])[kK]\b', s
)
3502 return {'height': int(mobj.group(1)) * 540}
3507 def parse_bitrate(s
):
3508 if not isinstance(s
, compat_str
):
3510 mobj
= re
.search(r
'\b(\d+)\s*kbps', s
)
3512 return int(mobj
.group(1))
3515 def month_by_name(name
, lang
='en'):
3516 """ Return the number of a month by (locale-independently) English name """
3518 month_names
= MONTH_NAMES
.get(lang
, MONTH_NAMES
['en'])
3521 return month_names
.index(name
) + 1
3526 def month_by_abbreviation(abbrev
):
3527 """ Return the number of a month by (locale-independently) English
3531 return [s
[:3] for s
in ENGLISH_MONTH_NAMES
].index(abbrev
) + 1
3536 def fix_xml_ampersands(xml_str
):
3537 """Replace all the '&' by '&' in XML"""
3539 r
'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3544 def setproctitle(title
):
3545 assert isinstance(title
, compat_str
)
3547 # ctypes in Jython is not complete
3548 # http://bugs.jython.org/issue2148
3549 if sys
.platform
.startswith('java'):
3553 libc
= ctypes
.cdll
.LoadLibrary('libc.so.6')
3557 # LoadLibrary in Windows Python 2.7.13 only expects
3558 # a bytestring, but since unicode_literals turns
3559 # every string into a unicode string, it fails.
3561 title_bytes
= title
.encode('utf-8')
3562 buf
= ctypes
.create_string_buffer(len(title_bytes
))
3563 buf
.value
= title_bytes
3565 libc
.prctl(15, buf
, 0, 0, 0)
3566 except AttributeError:
3567 return # Strange libc, just skip this
3570 def remove_start(s
, start
):
3571 return s
[len(start
):] if s
is not None and s
.startswith(start
) else s
3574 def remove_end(s
, end
):
3575 return s
[:-len(end
)] if s
is not None and s
.endswith(end
) else s
3578 def remove_quotes(s
):
3579 if s
is None or len(s
) < 2:
3581 for quote
in ('"', "'", ):
3582 if s
[0] == quote
and s
[-1] == quote
:
3587 def get_domain(url
):
3588 domain
= re
.match(r
'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url
)
3589 return domain
.group('domain') if domain
else None
3592 def url_basename(url
):
3593 path
= compat_urlparse
.urlparse(url
).path
3594 return path
.strip('/').split('/')[-1]
3598 return re
.match(r
'https?://[^?#&]+/', url
).group()
3601 def urljoin(base
, path
):
3602 if isinstance(path
, bytes):
3603 path
= path
.decode('utf-8')
3604 if not isinstance(path
, compat_str
) or not path
:
3606 if re
.match(r
'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path
):
3608 if isinstance(base
, bytes):
3609 base
= base
.decode('utf-8')
3610 if not isinstance(base
, compat_str
) or not re
.match(
3611 r
'^(?:https?:)?//', base
):
3613 return compat_urlparse
.urljoin(base
, path
)
3616 class HEADRequest(compat_urllib_request
.Request
):
3617 def get_method(self
):
3621 class PUTRequest(compat_urllib_request
.Request
):
3622 def get_method(self
):
3626 def int_or_none(v
, scale
=1, default
=None, get_attr
=None, invscale
=1):
3629 v
= getattr(v
, get_attr
, None)
3635 return int(v
) * invscale
// scale
3636 except (ValueError, TypeError):
3640 def str_or_none(v
, default
=None):
3641 return default
if v
is None else compat_str(v
)
3644 def str_to_int(int_str
):
3645 """ A more relaxed version of int_or_none """
3646 if isinstance(int_str
, compat_integer_types
):
3648 elif isinstance(int_str
, compat_str
):
3649 int_str
= re
.sub(r
'[,\.\+]', '', int_str
)
3650 return int_or_none(int_str
)
3653 def float_or_none(v
, scale
=1, invscale
=1, default
=None):
3657 return float(v
) * invscale
/ scale
3658 except (ValueError, TypeError):
3662 def bool_or_none(v
, default
=None):
3663 return v
if isinstance(v
, bool) else default
3666 def strip_or_none(v
, default
=None):
3667 return v
.strip() if isinstance(v
, compat_str
) else default
3670 def url_or_none(url
):
3671 if not url
or not isinstance(url
, compat_str
):
3674 return url
if re
.match(r
'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url
) else None
3677 def strftime_or_none(timestamp
, date_format
, default
=None):
3678 datetime_object
= None
3680 if isinstance(timestamp
, compat_numeric_types
): # unix timestamp
3681 datetime_object
= datetime
.datetime
.utcfromtimestamp(timestamp
)
3682 elif isinstance(timestamp
, compat_str
): # assume YYYYMMDD
3683 datetime_object
= datetime
.datetime
.strptime(timestamp
, '%Y%m%d')
3684 return datetime_object
.strftime(date_format
)
3685 except (ValueError, TypeError, AttributeError):
3689 def parse_duration(s
):
3690 if not isinstance(s
, compat_basestring
):
3695 days
, hours
, mins
, secs
, ms
= [None] * 5
3696 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
)
3698 days
, hours
, mins
, secs
, ms
= m
.groups()
3703 [0-9]+\s*y(?:ears?)?\s*
3706 [0-9]+\s*m(?:onths?)?\s*
3709 [0-9]+\s*w(?:eeks?)?\s*
3712 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3716 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3719 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3722 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3725 days
, hours
, mins
, secs
, ms
= m
.groups()
3727 m
= re
.match(r
'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s
)
3729 hours
, mins
= m
.groups()
3735 duration
+= float(secs
)
3737 duration
+= float(mins
) * 60
3739 duration
+= float(hours
) * 60 * 60
3741 duration
+= float(days
) * 24 * 60 * 60
3743 duration
+= float(ms
)
3747 def prepend_extension(filename
, ext
, expected_real_ext
=None):
3748 name
, real_ext
= os
.path
.splitext(filename
)
3750 '{0}.{1}{2}'.format(name
, ext
, real_ext
)
3751 if not expected_real_ext
or real_ext
[1:] == expected_real_ext
3752 else '{0}.{1}'.format(filename
, ext
))
3755 def replace_extension(filename
, ext
, expected_real_ext
=None):
3756 name
, real_ext
= os
.path
.splitext(filename
)
3757 return '{0}.{1}'.format(
3758 name
if not expected_real_ext
or real_ext
[1:] == expected_real_ext
else filename
,
3762 def check_executable(exe
, args
=[]):
3763 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3764 args can be a list of arguments for a short output (like -version) """
3766 process_communicate_or_kill(subprocess
.Popen(
3767 [exe
] + args
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
))
3773 def get_exe_version(exe
, args
=['--version'],
3774 version_re
=None, unrecognized
='present'):
3775 """ Returns the version of the specified executable,
3776 or False if the executable is not present """
3778 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3779 # SIGTTOU if youtube-dlc is run in the background.
3780 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3781 out
, _
= process_communicate_or_kill(subprocess
.Popen(
3782 [encodeArgument(exe
)] + args
,
3783 stdin
=subprocess
.PIPE
,
3784 stdout
=subprocess
.PIPE
, stderr
=subprocess
.STDOUT
))
3787 if isinstance(out
, bytes): # Python 2.x
3788 out
= out
.decode('ascii', 'ignore')
3789 return detect_exe_version(out
, version_re
, unrecognized
)
3792 def detect_exe_version(output
, version_re
=None, unrecognized
='present'):
3793 assert isinstance(output
, compat_str
)
3794 if version_re
is None:
3795 version_re
= r
'version\s+([-0-9._a-zA-Z]+)'
3796 m
= re
.search(version_re
, output
)
3803 class PagedList(object):
3805 # This is only useful for tests
3806 return len(self
.getslice())
3809 class OnDemandPagedList(PagedList
):
3810 def __init__(self
, pagefunc
, pagesize
, use_cache
=True):
3811 self
._pagefunc
= pagefunc
3812 self
._pagesize
= pagesize
3813 self
._use
_cache
= use_cache
3817 def getslice(self
, start
=0, end
=None):
3819 for pagenum
in itertools
.count(start
// self
._pagesize
):
3820 firstid
= pagenum
* self
._pagesize
3821 nextfirstid
= pagenum
* self
._pagesize
+ self
._pagesize
3822 if start
>= nextfirstid
:
3827 page_results
= self
._cache
.get(pagenum
)
3828 if page_results
is None:
3829 page_results
= list(self
._pagefunc
(pagenum
))
3831 self
._cache
[pagenum
] = page_results
3834 start
% self
._pagesize
3835 if firstid
<= start
< nextfirstid
3839 ((end
- 1) % self
._pagesize
) + 1
3840 if (end
is not None and firstid
<= end
<= nextfirstid
)
3843 if startv
!= 0 or endv
is not None:
3844 page_results
= page_results
[startv
:endv
]
3845 res
.extend(page_results
)
3847 # A little optimization - if current page is not "full", ie. does
3848 # not contain page_size videos then we can assume that this page
3849 # is the last one - there are no more ids on further pages -
3850 # i.e. no need to query again.
3851 if len(page_results
) + startv
< self
._pagesize
:
3854 # If we got the whole page, but the next page is not interesting,
3855 # break out early as well
3856 if end
== nextfirstid
:
3861 class InAdvancePagedList(PagedList
):
3862 def __init__(self
, pagefunc
, pagecount
, pagesize
):
3863 self
._pagefunc
= pagefunc
3864 self
._pagecount
= pagecount
3865 self
._pagesize
= pagesize
3867 def getslice(self
, start
=0, end
=None):
3869 start_page
= start
// self
._pagesize
3871 self
._pagecount
if end
is None else (end
// self
._pagesize
+ 1))
3872 skip_elems
= start
- start_page
* self
._pagesize
3873 only_more
= None if end
is None else end
- start
3874 for pagenum
in range(start_page
, end_page
):
3875 page
= list(self
._pagefunc
(pagenum
))
3877 page
= page
[skip_elems
:]
3879 if only_more
is not None:
3880 if len(page
) < only_more
:
3881 only_more
-= len(page
)
3883 page
= page
[:only_more
]
3890 def uppercase_escape(s
):
3891 unicode_escape
= codecs
.getdecoder('unicode_escape')
3893 r
'\\U[0-9a-fA-F]{8}',
3894 lambda m
: unicode_escape(m
.group(0))[0],
3898 def lowercase_escape(s
):
3899 unicode_escape
= codecs
.getdecoder('unicode_escape')
3901 r
'\\u[0-9a-fA-F]{4}',
3902 lambda m
: unicode_escape(m
.group(0))[0],
3906 def escape_rfc3986(s
):
3907 """Escape non-ASCII characters as suggested by RFC 3986"""
3908 if sys
.version_info
< (3, 0) and isinstance(s
, compat_str
):
3909 s
= s
.encode('utf-8')
3910 return compat_urllib_parse
.quote(s
, b
"%/;:@&=+$,!~*'()?#[]")
3913 def escape_url(url
):
3914 """Escape URL as suggested by RFC 3986"""
3915 url_parsed
= compat_urllib_parse_urlparse(url
)
3916 return url_parsed
._replace
(
3917 netloc
=url_parsed
.netloc
.encode('idna').decode('ascii'),
3918 path
=escape_rfc3986(url_parsed
.path
),
3919 params
=escape_rfc3986(url_parsed
.params
),
3920 query
=escape_rfc3986(url_parsed
.query
),
3921 fragment
=escape_rfc3986(url_parsed
.fragment
)
3925 def read_batch_urls(batch_fd
):
3927 if not isinstance(url
, compat_str
):
3928 url
= url
.decode('utf-8', 'replace')
3929 BOM_UTF8
= ('\xef\xbb\xbf', '\ufeff')
3930 for bom
in BOM_UTF8
:
3931 if url
.startswith(bom
):
3932 url
= url
[len(bom
):]
3934 if not url
or url
.startswith(('#', ';', ']')):
3936 # "#" cannot be stripped out since it is part of the URI
3937 # However, it can be safely stipped out if follwing a whitespace
3938 return re
.split(r
'\s#', url
, 1)[0].rstrip()
3940 with contextlib
.closing(batch_fd
) as fd
:
3941 return [url
for url
in map(fixup
, fd
) if url
]
3944 def urlencode_postdata(*args
, **kargs
):
3945 return compat_urllib_parse_urlencode(*args
, **kargs
).encode('ascii')
3948 def update_url_query(url
, query
):
3951 parsed_url
= compat_urlparse
.urlparse(url
)
3952 qs
= compat_parse_qs(parsed_url
.query
)
3954 return compat_urlparse
.urlunparse(parsed_url
._replace
(
3955 query
=compat_urllib_parse_urlencode(qs
, True)))
3958 def update_Request(req
, url
=None, data
=None, headers
={}, query={}
):
3959 req_headers
= req
.headers
.copy()
3960 req_headers
.update(headers
)
3961 req_data
= data
or req
.data
3962 req_url
= update_url_query(url
or req
.get_full_url(), query
)
3963 req_get_method
= req
.get_method()
3964 if req_get_method
== 'HEAD':
3965 req_type
= HEADRequest
3966 elif req_get_method
== 'PUT':
3967 req_type
= PUTRequest
3969 req_type
= compat_urllib_request
.Request
3971 req_url
, data
=req_data
, headers
=req_headers
,
3972 origin_req_host
=req
.origin_req_host
, unverifiable
=req
.unverifiable
)
3973 if hasattr(req
, 'timeout'):
3974 new_req
.timeout
= req
.timeout
3978 def _multipart_encode_impl(data
, boundary
):
3979 content_type
= 'multipart/form-data; boundary=%s' % boundary
3982 for k
, v
in data
.items():
3983 out
+= b
'--' + boundary
.encode('ascii') + b
'\r\n'
3984 if isinstance(k
, compat_str
):
3985 k
= k
.encode('utf-8')
3986 if isinstance(v
, compat_str
):
3987 v
= v
.encode('utf-8')
3988 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
3989 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
3990 content
= b
'Content-Disposition: form-data; name="' + k
+ b
'"\r\n\r\n' + v
+ b
'\r\n'
3991 if boundary
.encode('ascii') in content
:
3992 raise ValueError('Boundary overlaps with data')
3995 out
+= b
'--' + boundary
.encode('ascii') + b
'--\r\n'
3997 return out
, content_type
4000 def multipart_encode(data
, boundary
=None):
4002 Encode a dict to RFC 7578-compliant form-data
4005 A dict where keys and values can be either Unicode or bytes-like
4008 If specified a Unicode object, it's used as the boundary. Otherwise
4009 a random boundary is generated.
4011 Reference: https://tools.ietf.org/html/rfc7578
4013 has_specified_boundary
= boundary
is not None
4016 if boundary
is None:
4017 boundary
= '---------------' + str(random
.randrange(0x0fffffff, 0xffffffff))
4020 out
, content_type
= _multipart_encode_impl(data
, boundary
)
4023 if has_specified_boundary
:
4027 return out
, content_type
4030 def dict_get(d
, key_or_keys
, default
=None, skip_false_values
=True):
4031 if isinstance(key_or_keys
, (list, tuple)):
4032 for key
in key_or_keys
:
4033 if key
not in d
or d
[key
] is None or skip_false_values
and not d
[key
]:
4037 return d
.get(key_or_keys
, default
)
4040 def try_get(src
, getter
, expected_type
=None):
4041 if not isinstance(getter
, (list, tuple)):
4046 except (AttributeError, KeyError, TypeError, IndexError):
4049 if expected_type
is None or isinstance(v
, expected_type
):
4053 def merge_dicts(*dicts
):
4055 for a_dict
in dicts
:
4056 for k
, v
in a_dict
.items():
4060 or (isinstance(v
, compat_str
) and v
4061 and isinstance(merged
[k
], compat_str
)
4062 and not merged
[k
])):
4067 def encode_compat_str(string
, encoding
=preferredencoding(), errors
='strict'):
4068 return string
if isinstance(string
, compat_str
) else compat_str(string
, encoding
, errors
)
4080 TV_PARENTAL_GUIDELINES
= {
4090 def parse_age_limit(s
):
4092 return s
if 0 <= s
<= 21 else None
4093 if not isinstance(s
, compat_basestring
):
4095 m
= re
.match(r
'^(?P<age>\d{1,2})\+?$', s
)
4097 return int(m
.group('age'))
4099 return US_RATINGS
[s
]
4100 m
= re
.match(r
'^TV[_-]?(%s)$' % '|'.join(k
[3:] for k
in TV_PARENTAL_GUIDELINES
), s
)
4102 return TV_PARENTAL_GUIDELINES
['TV-' + m
.group(1)]
4106 def strip_jsonp(code
):
4109 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
4110 (?:\s*&&\s*(?P=func_name))?
4111 \s*\(\s*(?P<callback_data>.*)\);?
4112 \s*?(?://[^\n]*)*$''',
4113 r
'\g<callback_data>', code
)
4116 def js_to_json(code
, vars={}):
4117 # vars is a dict of var, val pairs to substitute
4118 COMMENT_RE
= r
'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
4119 SKIP_RE
= r
'\s*(?:{comment})?\s*'.format(comment
=COMMENT_RE
)
4121 (r
'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip
=SKIP_RE
), 16),
4122 (r
'(?s)^(0+[0-7]+){skip}:?$'.format(skip
=SKIP_RE
), 8),
4127 if v
in ('true', 'false', 'null'):
4129 elif v
.startswith('/*') or v
.startswith('//') or v
.startswith('!') or v
== ',':
4132 if v
[0] in ("'", '"'):
4133 v
= re
.sub(r
'(?s)\\.|"', lambda m
: {
4138 }.get(m
.group(0), m
.group(0)), v
[1:-1])
4140 for regex
, base
in INTEGER_TABLE
:
4141 im
= re
.match(regex
, v
)
4143 i
= int(im
.group(1), base
)
4144 return '"%d":' % i
if v
.endswith(':') else '%d' % i
4151 return re
.sub(r
'''(?sx)
4152 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4153 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4154 {comment}|,(?={skip}[\]}}])|
4155 (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
4156 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4159 '''.format(comment
=COMMENT_RE
, skip
=SKIP_RE
), fix_kv
, code
)
4162 def qualities(quality_ids
):
4163 """ Get a numeric quality value out of a list of possible values """
4166 return quality_ids
.index(qid
)
4172 DEFAULT_OUTTMPL
= '%(title)s [%(id)s].%(ext)s'
4175 def limit_length(s
, length
):
4176 """ Add ellipses to overly long strings """
4181 return s
[:length
- len(ELLIPSES
)] + ELLIPSES
4185 def version_tuple(v
):
4186 return tuple(int(e
) for e
in re
.split(r
'[-.]', v
))
4189 def is_outdated_version(version
, limit
, assume_new
=True):
4191 return not assume_new
4193 return version_tuple(version
) < version_tuple(limit
)
4195 return not assume_new
4198 def ytdl_is_updateable():
4199 """ Returns if youtube-dlc can be updated with -U """
4202 from zipimport
import zipimporter
4204 return isinstance(globals().get('__loader__'), zipimporter
) or hasattr(sys
, 'frozen')
4207 def args_to_str(args
):
4208 # Get a short string representation for a subprocess command
4209 return ' '.join(compat_shlex_quote(a
) for a
in args
)
4212 def error_to_compat_str(err
):
4214 # On python 2 error byte string must be decoded with proper
4215 # encoding rather than ascii
4216 if sys
.version_info
[0] < 3:
4217 err_str
= err_str
.decode(preferredencoding())
4221 def mimetype2ext(mt
):
4227 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4228 # it's the most popular one
4229 'audio/mpeg': 'mp3',
4230 'audio/x-wav': 'wav',
4235 _
, _
, res
= mt
.rpartition('/')
4236 res
= res
.split(';')[0].strip().lower()
4240 'smptett+xml': 'tt',
4244 'x-mp4-fragmented': 'mp4',
4245 'x-ms-sami': 'sami',
4248 'x-mpegurl': 'm3u8',
4249 'vnd.apple.mpegurl': 'm3u8',
4253 'vnd.ms-sstr+xml': 'ism',
4260 def parse_codecs(codecs_str
):
4261 # http://tools.ietf.org/html/rfc6381
4264 split_codecs
= list(filter(None, map(
4265 lambda str: str.strip(), codecs_str
.strip().strip(',').split(','))))
4266 vcodec
, acodec
= None, None
4267 for full_codec
in split_codecs
:
4268 codec
= full_codec
.split('.')[0]
4269 if codec
in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4272 elif codec
in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4276 write_string('WARNING: Unknown codec %s\n' % full_codec
, sys
.stderr
)
4277 if not vcodec
and not acodec
:
4278 if len(split_codecs
) == 2:
4280 'vcodec': split_codecs
[0],
4281 'acodec': split_codecs
[1],
4285 'vcodec': vcodec
or 'none',
4286 'acodec': acodec
or 'none',
4291 def urlhandle_detect_ext(url_handle
):
4292 getheader
= url_handle
.headers
.get
4294 cd
= getheader('Content-Disposition')
4296 m
= re
.match(r
'attachment;\s*filename="(?P<filename>[^"]+)"', cd
)
4298 e
= determine_ext(m
.group('filename'), default_ext
=None)
4302 return mimetype2ext(getheader('Content-Type'))
4305 def encode_data_uri(data
, mime_type
):
4306 return 'data:%s;base64,%s' % (mime_type
, base64
.b64encode(data
).decode('ascii'))
4309 def age_restricted(content_limit
, age_limit
):
4310 """ Returns True iff the content should be blocked """
4312 if age_limit
is None: # No limit set
4314 if content_limit
is None:
4315 return False # Content available for everyone
4316 return age_limit
< content_limit
4319 def is_html(first_bytes
):
4320 """ Detect whether a file contains HTML by examining its first bytes. """
4323 (b
'\xef\xbb\xbf', 'utf-8'),
4324 (b
'\x00\x00\xfe\xff', 'utf-32-be'),
4325 (b
'\xff\xfe\x00\x00', 'utf-32-le'),
4326 (b
'\xff\xfe', 'utf-16-le'),
4327 (b
'\xfe\xff', 'utf-16-be'),
4329 for bom
, enc
in BOMS
:
4330 if first_bytes
.startswith(bom
):
4331 s
= first_bytes
[len(bom
):].decode(enc
, 'replace')
4334 s
= first_bytes
.decode('utf-8', 'replace')
4336 return re
.match(r
'^\s*<', s
)
4339 def determine_protocol(info_dict
):
4340 protocol
= info_dict
.get('protocol')
4341 if protocol
is not None:
4344 url
= info_dict
['url']
4345 if url
.startswith('rtmp'):
4347 elif url
.startswith('mms'):
4349 elif url
.startswith('rtsp'):
4352 ext
= determine_ext(url
)
4358 return compat_urllib_parse_urlparse(url
).scheme
4361 def render_table(header_row
, data
, delim
=False, extraGap
=0, hideEmpty
=False):
4362 """ Render a list of rows, each as a list of values """
4364 def get_max_lens(table
):
4365 return [max(len(compat_str(v
)) for v
in col
) for col
in zip(*table
)]
4367 def filter_using_list(row
, filterArray
):
4368 return [col
for (take
, col
) in zip(filterArray
, row
) if take
]
4371 max_lens
= get_max_lens(data
)
4372 header_row
= filter_using_list(header_row
, max_lens
)
4373 data
= [filter_using_list(row
, max_lens
) for row
in data
]
4375 table
= [header_row
] + data
4376 max_lens
= get_max_lens(table
)
4378 table
= [header_row
] + [['-' * ml
for ml
in max_lens
]] + data
4379 format_str
= ' '.join('%-' + compat_str(ml
+ extraGap
) + 's' for ml
in max_lens
[:-1]) + ' %s'
4380 return '\n'.join(format_str
% tuple(row
) for row
in table
)
4383 def _match_one(filter_part
, dct
):
4384 COMPARISON_OPERATORS
= {
4392 operator_rex
= re
.compile(r
'''(?x)\s*
4394 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4396 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4397 (?P<quote>["\'])(?P
<quotedstrval
>(?
:\\.|
(?
!(?P
=quote
)|
\\).)+?
)(?P
=quote
)|
4398 (?P
<strval
>(?
![0-9.])[a
-z0
-9A
-Z
]*)
4401 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4402 m = operator_rex.search(filter_part)
4404 op = COMPARISON_OPERATORS[m.group('op')]
4405 actual_value = dct.get(m.group('key'))
4406 if (m.group('quotedstrval') is not None
4407 or m.group('strval') is not None
4408 # If the original field is a string and matching comparisonvalue is
4409 # a number we should respect the origin of the original field
4410 # and process comparison value as a string (see
4411 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4412 or actual_value is not None and m.group('intval') is not None
4413 and isinstance(actual_value, compat_str)):
4414 if m.group('op') not in ('=', '!='):
4416 'Operator %s does not support string values!' % m.group('op'))
4417 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4418 quote = m.group('quote')
4419 if quote is not None:
4420 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4423 comparison_value = int(m.group('intval'))
4425 comparison_value = parse_filesize(m.group('intval'))
4426 if comparison_value is None:
4427 comparison_value = parse_filesize(m.group('intval') + 'B')
4428 if comparison_value is None:
4430 'Invalid integer value %r in filter part %r' % (
4431 m.group('intval'), filter_part))
4432 if actual_value is None:
4433 return m.group('none_inclusive')
4434 return op(actual_value, comparison_value)
4437 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4438 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4440 operator_rex = re.compile(r'''(?x
)\s
*
4441 (?P
<op
>%s)\s
*(?P
<key
>[a
-z_
]+)
4443 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4444 m = operator_rex.search(filter_part)
4446 op = UNARY_OPERATORS[m.group('op')]
4447 actual_value = dct.get(m.group('key'))
4448 return op(actual_value)
4450 raise ValueError('Invalid filter part %r' % filter_part)
4453 def match_str(filter_str, dct):
4454 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
4457 _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
4460 def match_filter_func(filter_str):
4461 def _match_func(info_dict):
4462 if match_str(filter_str, info_dict):
4465 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4466 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4470 def parse_dfxp_time_expr(time_expr):
4474 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4476 return float(mobj.group('time_offset'))
4478 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4480 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4483 def srt_subtitles_timecode(seconds):
4484 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4487 def dfxp2srt(dfxp_data):
4489 @param dfxp_data A
bytes-like
object containing DFXP data
4490 @returns A
unicode object containing converted SRT data
4492 LEGACY_NAMESPACES = (
4493 (b'http://www.w3.org/ns/ttml', [
4494 b'http://www.w3.org/2004/11/ttaf1',
4495 b'http://www.w3.org/2006/04/ttaf1',
4496 b'http://www.w3.org/2006/10/ttaf1',
4498 (b'http://www.w3.org/ns/ttml#styling', [
4499 b'http://www.w3.org/ns/ttml#style',
4503 SUPPORTED_STYLING = [
4512 _x = functools.partial(xpath_with_ns, ns_map={
4513 'xml': 'http://www.w3.org/XML/1998/namespace',
4514 'ttml': 'http://www.w3.org/ns/ttml',
4515 'tts': 'http://www.w3.org/ns/ttml#styling',
4521 class TTMLPElementParser(object):
4523 _unclosed_elements = []
4524 _applied_styles = []
4526 def start(self, tag, attrib):
4527 if tag in (_x('ttml:br'), 'br'):
4530 unclosed_elements = []
4532 element_style_id = attrib.get('style')
4534 style.update(default_style)
4535 if element_style_id:
4536 style.update(styles.get(element_style_id, {}))
4537 for prop in SUPPORTED_STYLING:
4538 prop_val = attrib.get(_x('tts:' + prop))
4540 style[prop] = prop_val
4543 for k, v in sorted(style.items()):
4544 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4547 font += ' color="%s"' % v
4548 elif k == 'fontSize':
4549 font += ' size="%s"' % v
4550 elif k == 'fontFamily':
4551 font += ' face="%s"' % v
4552 elif k == 'fontWeight' and v == 'bold':
4554 unclosed_elements.append('b')
4555 elif k == 'fontStyle' and v == 'italic':
4557 unclosed_elements.append('i')
4558 elif k == 'textDecoration' and v == 'underline':
4560 unclosed_elements.append('u')
4562 self._out += '<font' + font + '>'
4563 unclosed_elements.append('font')
4565 if self._applied_styles:
4566 applied_style.update(self._applied_styles[-1])
4567 applied_style.update(style)
4568 self._applied_styles.append(applied_style)
4569 self._unclosed_elements.append(unclosed_elements)
4572 if tag not in (_x('ttml:br'), 'br'):
4573 unclosed_elements = self._unclosed_elements.pop()
4574 for element in reversed(unclosed_elements):
4575 self._out += '</%s>' % element
4576 if unclosed_elements and self._applied_styles:
4577 self._applied_styles.pop()
4579 def data(self, data):
4583 return self._out.strip()
4585 def parse_node(node):
4586 target = TTMLPElementParser()
4587 parser = xml.etree.ElementTree.XMLParser(target=target)
4588 parser.feed(xml.etree.ElementTree.tostring(node))
4589 return parser.close()
4591 for k, v in LEGACY_NAMESPACES:
4593 dfxp_data = dfxp_data.replace(ns, k)
4595 dfxp = compat_etree_fromstring(dfxp_data)
4597 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4600 raise ValueError('Invalid dfxp/TTML subtitle')
4604 for style in dfxp.findall(_x('.//ttml:style')):
4605 style_id = style.get('id') or style.get(_x('xml:id'))
4608 parent_style_id = style.get('style')
4610 if parent_style_id not in styles:
4613 styles[style_id] = styles[parent_style_id].copy()
4614 for prop in SUPPORTED_STYLING:
4615 prop_val = style.get(_x('tts:' + prop))
4617 styles.setdefault(style_id, {})[prop] = prop_val
4623 for p in ('body', 'div'):
4624 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4627 style = styles.get(ele.get('style'))
4630 default_style.update(style)
4632 for para, index in zip(paras, itertools.count(1)):
4633 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
4634 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
4635 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
4636 if begin_time is None:
4641 end_time = begin_time + dur
4642 out.append('%d\n%s --> %s\n%s\n\n' % (
4644 srt_subtitles_timecode(begin_time),
4645 srt_subtitles_timecode(end_time),
4651 def cli_option(params, command_option, param):
4652 param = params.get(param)
4654 param = compat_str(param)
4655 return [command_option, param] if param is not None else []
4658 def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
4659 param = params.get(param)
4662 assert isinstance(param, bool)
4664 return [command_option + separator + (true_value if param else false_value)]
4665 return [command_option, true_value if param else false_value]
4668 def cli_valueless_option(params, command_option, param, expected_value=True):
4669 param = params.get(param)
4670 return [command_option] if param == expected_value else []
4673 def cli_configuration_args(params, arg_name, key, default=[], exe=None): # returns arg, for_compat
4674 argdict = params.get(arg_name, {})
4675 if isinstance(argdict, (list, tuple)): # for backward compatibility
4676 return argdict, True
4679 return default, False
4680 assert isinstance(argdict, dict)
4682 assert isinstance(key, compat_str)
4685 args = exe_args = None
4687 assert isinstance(exe, compat_str)
4689 args = argdict.get('%s+%s' % (key, exe))
4691 exe_args = argdict.get(exe)
4694 args = argdict.get(key) if key != exe else None
4695 if args is None and exe_args is None:
4696 args = argdict.get('default', default)
4698 args, exe_args = args or [], exe_args or []
4699 assert isinstance(args, (list, tuple))
4700 assert isinstance(exe_args, (list, tuple))
4701 return args + exe_args, False
4704 class ISO639Utils(object):
4705 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
4764 'iw': 'heb', # Replaced by he in 1989 revision
4774 'in': 'ind', # Replaced by id in 1989 revision
4889 'ji': 'yid', # Replaced by yi in 1989 revision
4897 def short2long(cls, code):
4898 """Convert language code from ISO 639-1 to ISO 639-2/T"""
4899 return cls._lang_map.get(code[:2])
4902 def long2short(cls, code):
4903 """Convert language code from ISO 639-2/T to ISO 639-1"""
4904 for short_name, long_name in cls._lang_map.items():
4905 if long_name == code:
4909 class ISO3166Utils(object):
4910 # From http://data.okfn.org/data/core/country-list
4912 'AF': 'Afghanistan',
4913 'AX': 'Åland Islands',
4916 'AS': 'American Samoa',
4921 'AG': 'Antigua and Barbuda',
4938 'BO': 'Bolivia, Plurinational State of',
4939 'BQ': 'Bonaire, Sint Eustatius and Saba',
4940 'BA': 'Bosnia and Herzegovina',
4942 'BV': 'Bouvet Island',
4944 'IO': 'British Indian Ocean Territory',
4945 'BN': 'Brunei Darussalam',
4947 'BF': 'Burkina Faso',
4953 'KY': 'Cayman Islands',
4954 'CF': 'Central African Republic',
4958 'CX': 'Christmas Island',
4959 'CC': 'Cocos (Keeling) Islands',
4963 'CD': 'Congo, the Democratic Republic of the',
4964 'CK': 'Cook Islands',
4966 'CI': 'Côte d\'Ivoire',
4971 'CZ': 'Czech Republic',
4975 'DO': 'Dominican Republic',
4978 'SV': 'El Salvador',
4979 'GQ': 'Equatorial Guinea',
4983 'FK': 'Falkland Islands (Malvinas)',
4984 'FO': 'Faroe Islands',
4988 'GF': 'French Guiana',
4989 'PF': 'French Polynesia',
4990 'TF': 'French Southern Territories',
5005 'GW': 'Guinea-Bissau',
5008 'HM': 'Heard Island and McDonald Islands',
5009 'VA': 'Holy See (Vatican City State)',
5016 'IR': 'Iran, Islamic Republic of',
5019 'IM': 'Isle of Man',
5029 'KP': 'Korea, Democratic People\'s Republic of',
5030 'KR': 'Korea, Republic of',
5033 'LA': 'Lao People\'s Democratic Republic',
5039 'LI': 'Liechtenstein',
5043 'MK': 'Macedonia, the Former Yugoslav Republic of',
5050 'MH': 'Marshall Islands',
5056 'FM': 'Micronesia, Federated States of',
5057 'MD': 'Moldova, Republic of',
5068 'NL': 'Netherlands',
5069 'NC': 'New Caledonia',
5070 'NZ': 'New Zealand',
5075 'NF': 'Norfolk Island',
5076 'MP': 'Northern Mariana Islands',
5081 'PS': 'Palestine, State of',
5083 'PG': 'Papua New Guinea',
5086 'PH': 'Philippines',
5090 'PR': 'Puerto Rico',
5094 'RU': 'Russian Federation',
5096 'BL': 'Saint Barthélemy',
5097 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
5098 'KN': 'Saint Kitts and Nevis',
5099 'LC': 'Saint Lucia',
5100 'MF': 'Saint Martin (French part)',
5101 'PM': 'Saint Pierre and Miquelon',
5102 'VC': 'Saint Vincent and the Grenadines',
5105 'ST': 'Sao Tome and Principe',
5106 'SA': 'Saudi Arabia',
5110 'SL': 'Sierra Leone',
5112 'SX': 'Sint Maarten (Dutch part)',
5115 'SB': 'Solomon Islands',
5117 'ZA': 'South Africa',
5118 'GS': 'South Georgia and the South Sandwich Islands',
5119 'SS': 'South Sudan',
5124 'SJ': 'Svalbard and Jan Mayen',
5127 'CH': 'Switzerland',
5128 'SY': 'Syrian Arab Republic',
5129 'TW': 'Taiwan, Province of China',
5131 'TZ': 'Tanzania, United Republic of',
5133 'TL': 'Timor-Leste',
5137 'TT': 'Trinidad and Tobago',
5140 'TM': 'Turkmenistan',
5141 'TC': 'Turks and Caicos Islands',
5145 'AE': 'United Arab Emirates',
5146 'GB': 'United Kingdom',
5147 'US': 'United States',
5148 'UM': 'United States Minor Outlying Islands',
5152 'VE': 'Venezuela, Bolivarian Republic of',
5154 'VG': 'Virgin Islands, British',
5155 'VI': 'Virgin Islands, U.S.',
5156 'WF': 'Wallis and Futuna',
5157 'EH': 'Western Sahara',
5164 def short2full(cls, code):
5165 """Convert an ISO 3166-2 country code to the corresponding full name"""
5166 return cls._country_map.get(code.upper())
5169 class GeoUtils(object):
5170 # Major IPv4 address blocks per country
5172 'AD': '46.172.224.0/19',
5173 'AE': '94.200.0.0/13',
5174 'AF': '149.54.0.0/17',
5175 'AG': '209.59.64.0/18',
5176 'AI': '204.14.248.0/21',
5177 'AL': '46.99.0.0/16',
5178 'AM': '46.70.0.0/15',
5179 'AO': '105.168.0.0/13',
5180 'AP': '182.50.184.0/21',
5181 'AQ': '23.154.160.0/24',
5182 'AR': '181.0.0.0/12',
5183 'AS': '202.70.112.0/20',
5184 'AT': '77.116.0.0/14',
5185 'AU': '1.128.0.0/11',
5186 'AW': '181.41.0.0/18',
5187 'AX': '185.217.4.0/22',
5188 'AZ': '5.197.0.0/16',
5189 'BA': '31.176.128.0/17',
5190 'BB': '65.48.128.0/17',
5191 'BD': '114.130.0.0/16',
5193 'BF': '102.178.0.0/15',
5194 'BG': '95.42.0.0/15',
5195 'BH': '37.131.0.0/17',
5196 'BI': '154.117.192.0/18',
5197 'BJ': '137.255.0.0/16',
5198 'BL': '185.212.72.0/23',
5199 'BM': '196.12.64.0/18',
5200 'BN': '156.31.0.0/16',
5201 'BO': '161.56.0.0/16',
5202 'BQ': '161.0.80.0/20',
5203 'BR': '191.128.0.0/12',
5204 'BS': '24.51.64.0/18',
5205 'BT': '119.2.96.0/19',
5206 'BW': '168.167.0.0/16',
5207 'BY': '178.120.0.0/13',
5208 'BZ': '179.42.192.0/18',
5209 'CA': '99.224.0.0/11',
5210 'CD': '41.243.0.0/16',
5211 'CF': '197.242.176.0/21',
5212 'CG': '160.113.0.0/16',
5213 'CH': '85.0.0.0/13',
5214 'CI': '102.136.0.0/14',
5215 'CK': '202.65.32.0/19',
5216 'CL': '152.172.0.0/14',
5217 'CM': '102.244.0.0/14',
5218 'CN': '36.128.0.0/10',
5219 'CO': '181.240.0.0/12',
5220 'CR': '201.192.0.0/12',
5221 'CU': '152.206.0.0/15',
5222 'CV': '165.90.96.0/19',
5223 'CW': '190.88.128.0/17',
5224 'CY': '31.153.0.0/16',
5225 'CZ': '88.100.0.0/14',
5227 'DJ': '197.241.0.0/17',
5228 'DK': '87.48.0.0/12',
5229 'DM': '192.243.48.0/20',
5230 'DO': '152.166.0.0/15',
5231 'DZ': '41.96.0.0/12',
5232 'EC': '186.68.0.0/15',
5233 'EE': '90.190.0.0/15',
5234 'EG': '156.160.0.0/11',
5235 'ER': '196.200.96.0/20',
5236 'ES': '88.0.0.0/11',
5237 'ET': '196.188.0.0/14',
5238 'EU': '2.16.0.0/13',
5239 'FI': '91.152.0.0/13',
5240 'FJ': '144.120.0.0/16',
5241 'FK': '80.73.208.0/21',
5242 'FM': '119.252.112.0/20',
5243 'FO': '88.85.32.0/19',
5245 'GA': '41.158.0.0/15',
5247 'GD': '74.122.88.0/21',
5248 'GE': '31.146.0.0/16',
5249 'GF': '161.22.64.0/18',
5250 'GG': '62.68.160.0/19',
5251 'GH': '154.160.0.0/12',
5252 'GI': '95.164.0.0/16',
5253 'GL': '88.83.0.0/19',
5254 'GM': '160.182.0.0/15',
5255 'GN': '197.149.192.0/18',
5256 'GP': '104.250.0.0/19',
5257 'GQ': '105.235.224.0/20',
5258 'GR': '94.64.0.0/13',
5259 'GT': '168.234.0.0/16',
5260 'GU': '168.123.0.0/16',
5261 'GW': '197.214.80.0/20',
5262 'GY': '181.41.64.0/18',
5263 'HK': '113.252.0.0/14',
5264 'HN': '181.210.0.0/16',
5265 'HR': '93.136.0.0/13',
5266 'HT': '148.102.128.0/17',
5267 'HU': '84.0.0.0/14',
5268 'ID': '39.192.0.0/10',
5269 'IE': '87.32.0.0/12',
5270 'IL': '79.176.0.0/13',
5271 'IM': '5.62.80.0/20',
5272 'IN': '117.192.0.0/10',
5273 'IO': '203.83.48.0/21',
5274 'IQ': '37.236.0.0/14',
5275 'IR': '2.176.0.0/12',
5276 'IS': '82.221.0.0/16',
5277 'IT': '79.0.0.0/10',
5278 'JE': '87.244.64.0/18',
5279 'JM': '72.27.0.0/17',
5280 'JO': '176.29.0.0/16',
5281 'JP': '133.0.0.0/8',
5282 'KE': '105.48.0.0/12',
5283 'KG': '158.181.128.0/17',
5284 'KH': '36.37.128.0/17',
5285 'KI': '103.25.140.0/22',
5286 'KM': '197.255.224.0/20',
5287 'KN': '198.167.192.0/19',
5288 'KP': '175.45.176.0/22',
5289 'KR': '175.192.0.0/10',
5290 'KW': '37.36.0.0/14',
5291 'KY': '64.96.0.0/15',
5292 'KZ': '2.72.0.0/13',
5293 'LA': '115.84.64.0/18',
5294 'LB': '178.135.0.0/16',
5295 'LC': '24.92.144.0/20',
5296 'LI': '82.117.0.0/19',
5297 'LK': '112.134.0.0/15',
5298 'LR': '102.183.0.0/16',
5299 'LS': '129.232.0.0/17',
5300 'LT': '78.56.0.0/13',
5301 'LU': '188.42.0.0/16',
5302 'LV': '46.109.0.0/16',
5303 'LY': '41.252.0.0/14',
5304 'MA': '105.128.0.0/11',
5305 'MC': '88.209.64.0/18',
5306 'MD': '37.246.0.0/16',
5307 'ME': '178.175.0.0/17',
5308 'MF': '74.112.232.0/21',
5309 'MG': '154.126.0.0/17',
5310 'MH': '117.103.88.0/21',
5311 'MK': '77.28.0.0/15',
5312 'ML': '154.118.128.0/18',
5313 'MM': '37.111.0.0/17',
5314 'MN': '49.0.128.0/17',
5315 'MO': '60.246.0.0/16',
5316 'MP': '202.88.64.0/20',
5317 'MQ': '109.203.224.0/19',
5318 'MR': '41.188.64.0/18',
5319 'MS': '208.90.112.0/22',
5320 'MT': '46.11.0.0/16',
5321 'MU': '105.16.0.0/12',
5322 'MV': '27.114.128.0/18',
5323 'MW': '102.70.0.0/15',
5324 'MX': '187.192.0.0/11',
5325 'MY': '175.136.0.0/13',
5326 'MZ': '197.218.0.0/15',
5327 'NA': '41.182.0.0/16',
5328 'NC': '101.101.0.0/18',
5329 'NE': '197.214.0.0/18',
5330 'NF': '203.17.240.0/22',
5331 'NG': '105.112.0.0/12',
5332 'NI': '186.76.0.0/15',
5333 'NL': '145.96.0.0/11',
5334 'NO': '84.208.0.0/13',
5335 'NP': '36.252.0.0/15',
5336 'NR': '203.98.224.0/19',
5337 'NU': '49.156.48.0/22',
5338 'NZ': '49.224.0.0/14',
5339 'OM': '5.36.0.0/15',
5340 'PA': '186.72.0.0/15',
5341 'PE': '186.160.0.0/14',
5342 'PF': '123.50.64.0/18',
5343 'PG': '124.240.192.0/19',
5344 'PH': '49.144.0.0/13',
5345 'PK': '39.32.0.0/11',
5346 'PL': '83.0.0.0/11',
5347 'PM': '70.36.0.0/20',
5348 'PR': '66.50.0.0/16',
5349 'PS': '188.161.0.0/16',
5350 'PT': '85.240.0.0/13',
5351 'PW': '202.124.224.0/20',
5352 'PY': '181.120.0.0/14',
5353 'QA': '37.210.0.0/15',
5354 'RE': '102.35.0.0/16',
5355 'RO': '79.112.0.0/13',
5356 'RS': '93.86.0.0/15',
5357 'RU': '5.136.0.0/13',
5358 'RW': '41.186.0.0/16',
5359 'SA': '188.48.0.0/13',
5360 'SB': '202.1.160.0/19',
5361 'SC': '154.192.0.0/11',
5362 'SD': '102.120.0.0/13',
5363 'SE': '78.64.0.0/12',
5364 'SG': '8.128.0.0/10',
5365 'SI': '188.196.0.0/14',
5366 'SK': '78.98.0.0/15',
5367 'SL': '102.143.0.0/17',
5368 'SM': '89.186.32.0/19',
5369 'SN': '41.82.0.0/15',
5370 'SO': '154.115.192.0/18',
5371 'SR': '186.179.128.0/17',
5372 'SS': '105.235.208.0/21',
5373 'ST': '197.159.160.0/19',
5374 'SV': '168.243.0.0/16',
5375 'SX': '190.102.0.0/20',
5377 'SZ': '41.84.224.0/19',
5378 'TC': '65.255.48.0/20',
5379 'TD': '154.68.128.0/19',
5380 'TG': '196.168.0.0/14',
5381 'TH': '171.96.0.0/13',
5382 'TJ': '85.9.128.0/18',
5383 'TK': '27.96.24.0/21',
5384 'TL': '180.189.160.0/20',
5385 'TM': '95.85.96.0/19',
5386 'TN': '197.0.0.0/11',
5387 'TO': '175.176.144.0/21',
5388 'TR': '78.160.0.0/11',
5389 'TT': '186.44.0.0/15',
5390 'TV': '202.2.96.0/19',
5391 'TW': '120.96.0.0/11',
5392 'TZ': '156.156.0.0/14',
5393 'UA': '37.52.0.0/14',
5394 'UG': '102.80.0.0/13',
5396 'UY': '167.56.0.0/13',
5397 'UZ': '84.54.64.0/18',
5398 'VA': '212.77.0.0/19',
5399 'VC': '207.191.240.0/21',
5400 'VE': '186.88.0.0/13',
5401 'VG': '66.81.192.0/20',
5402 'VI': '146.226.0.0/16',
5403 'VN': '14.160.0.0/11',
5404 'VU': '202.80.32.0/20',
5405 'WF': '117.20.32.0/21',
5406 'WS': '202.4.32.0/19',
5407 'YE': '134.35.0.0/16',
5408 'YT': '41.242.116.0/22',
5409 'ZA': '41.0.0.0/11',
5410 'ZM': '102.144.0.0/13',
5411 'ZW': '102.177.192.0/18',
5415 def random_ipv4(cls, code_or_block):
5416 if len(code_or_block) == 2:
5417 block = cls._country_ip_map.get(code_or_block.upper())
5421 block = code_or_block
5422 addr, preflen = block.split('/')
5423 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5424 addr_max = addr_min | (0xffffffff >> int(preflen))
5425 return compat_str(socket.inet_ntoa(
5426 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5429 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5430 def __init__(self, proxies=None):
5431 # Set default handlers
5432 for type in ('http', 'https'):
5433 setattr(self, '%s_open' % type,
5434 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5435 meth(r, proxy, type))
5436 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5438 def proxy_open(self, req, proxy, type):
5439 req_proxy = req.headers.get('Ytdl-request-proxy')
5440 if req_proxy is not None:
5442 del req.headers['Ytdl-request-proxy']
5444 if proxy == '__noproxy__':
5445 return None # No Proxy
5446 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5447 req.add_header('Ytdl-socks-proxy', proxy)
5448 # youtube-dlc's http/https handlers do wrapping the socket with socks
5450 return compat_urllib_request.ProxyHandler.proxy_open(
5451 self, req, proxy, type)
5454 # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5455 # released into Public Domain
5456 # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5458 def long_to_bytes(n, blocksize=0):
5459 """long_to_bytes(n:long, blocksize:int) : string
5460 Convert a long integer to a byte string.
5462 If optional blocksize is given and greater than zero, pad the front of the
5463 byte string with binary zeros so that the length is a multiple of
5466 # after much testing, this algorithm was deemed to be the fastest
5470 s = compat_struct_pack('>I', n & 0xffffffff) + s
5472 # strip off leading zeros
5473 for i in range(len(s)):
5474 if s[i] != b'\000'[0]:
5477 # only happens when n == 0
5481 # add back some pad bytes. this could be done more efficiently w.r.t. the
5482 # de-padding being done above, but sigh...
5483 if blocksize > 0 and len(s) % blocksize:
5484 s = (blocksize - len(s) % blocksize) * b'\000' + s
5488 def bytes_to_long(s):
5489 """bytes_to_long(string) : long
5490 Convert a byte string to a long integer.
5492 This is (essentially) the inverse of long_to_bytes().
5497 extra = (4 - length % 4)
5498 s = b'\000' * extra + s
5499 length = length + extra
5500 for i in range(0, length, 4):
5501 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5505 def ohdave_rsa_encrypt(data, exponent, modulus):
5507 Implement OHDave
's RSA algorithm. See http://www.ohdave.com/rsa/
5510 data: data to encrypt, bytes-like object
5511 exponent, modulus: parameter e and N of RSA algorithm, both integer
5512 Output: hex string of encrypted data
5514 Limitation: supports one block encryption only
5517 payload = int(binascii.hexlify(data[::-1]), 16)
5518 encrypted = pow(payload, exponent, modulus)
5519 return '%x' % encrypted
5522 def pkcs1pad(data, length):
5524 Padding input data with PKCS#1 scheme
5526 @param {int[]} data input data
5527 @param {int} length target length
5528 @returns {int[]} padded data
5530 if len(data) > length - 11:
5531 raise ValueError('Input data too
long for PKCS
#1 padding')
5533 pseudo_random
= [random
.randint(0, 254) for _
in range(length
- len(data
) - 3)]
5534 return [0, 2] + pseudo_random
+ [0] + data
5537 def encode_base_n(num
, n
, table
=None):
5538 FULL_TABLE
= '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5540 table
= FULL_TABLE
[:n
]
5543 raise ValueError('base %d exceeds table length %d' % (n
, len(table
)))
5550 ret
= table
[num
% n
] + ret
5555 def decode_packed_codes(code
):
5556 mobj
= re
.search(PACKED_CODES_RE
, code
)
5557 obfuscated_code
, base
, count
, symbols
= mobj
.groups()
5560 symbols
= symbols
.split('|')
5565 base_n_count
= encode_base_n(count
, base
)
5566 symbol_table
[base_n_count
] = symbols
[count
] or base_n_count
5569 r
'\b(\w+)\b', lambda mobj
: symbol_table
[mobj
.group(0)],
5573 def caesar(s
, alphabet
, shift
):
5578 alphabet
[(alphabet
.index(c
) + shift
) % l
] if c
in alphabet
else c
5583 return caesar(s
, r
'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5586 def parse_m3u8_attributes(attrib
):
5588 for (key
, val
) in re
.findall(r
'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib
):
5589 if val
.startswith('"'):
5595 def urshift(val
, n
):
5596 return val
>> n
if val
>= 0 else (val
+ 0x100000000) >> n
5599 # Based on png2str() written by @gdkchan and improved by @yokrysty
5600 # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5601 def decode_png(png_data
):
5602 # Reference: https://www.w3.org/TR/PNG/
5603 header
= png_data
[8:]
5605 if png_data
[:8] != b
'\x89PNG\x0d\x0a\x1a\x0a' or header
[4:8] != b
'IHDR':
5606 raise IOError('Not a valid PNG file.')
5608 int_map
= {1: '>B', 2: '>H', 4: '>I'}
5609 unpack_integer
= lambda x
: compat_struct_unpack(int_map
[len(x
)], x
)[0]
5614 length
= unpack_integer(header
[:4])
5617 chunk_type
= header
[:4]
5620 chunk_data
= header
[:length
]
5621 header
= header
[length
:]
5623 header
= header
[4:] # Skip CRC
5631 ihdr
= chunks
[0]['data']
5633 width
= unpack_integer(ihdr
[:4])
5634 height
= unpack_integer(ihdr
[4:8])
5638 for chunk
in chunks
:
5639 if chunk
['type'] == b
'IDAT':
5640 idat
+= chunk
['data']
5643 raise IOError('Unable to read PNG data.')
5645 decompressed_data
= bytearray(zlib
.decompress(idat
))
5650 def _get_pixel(idx
):
5655 for y
in range(height
):
5656 basePos
= y
* (1 + stride
)
5657 filter_type
= decompressed_data
[basePos
]
5661 pixels
.append(current_row
)
5663 for x
in range(stride
):
5664 color
= decompressed_data
[1 + basePos
+ x
]
5665 basex
= y
* stride
+ x
5670 left
= _get_pixel(basex
- 3)
5672 up
= _get_pixel(basex
- stride
)
5674 if filter_type
== 1: # Sub
5675 color
= (color
+ left
) & 0xff
5676 elif filter_type
== 2: # Up
5677 color
= (color
+ up
) & 0xff
5678 elif filter_type
== 3: # Average
5679 color
= (color
+ ((left
+ up
) >> 1)) & 0xff
5680 elif filter_type
== 4: # Paeth
5686 c
= _get_pixel(basex
- stride
- 3)
5694 if pa
<= pb
and pa
<= pc
:
5695 color
= (color
+ a
) & 0xff
5697 color
= (color
+ b
) & 0xff
5699 color
= (color
+ c
) & 0xff
5701 current_row
.append(color
)
5703 return width
, height
, pixels
5706 def write_xattr(path
, key
, value
):
5707 # This mess below finds the best xattr tool for the job
5709 # try the pyxattr module...
5712 if hasattr(xattr
, 'set'): # pyxattr
5713 # Unicode arguments are not supported in python-pyxattr until
5715 # See https://github.com/ytdl-org/youtube-dl/issues/5498
5716 pyxattr_required_version
= '0.5.0'
5717 if version_tuple(xattr
.__version
__) < version_tuple(pyxattr_required_version
):
5718 # TODO: fallback to CLI tools
5719 raise XAttrUnavailableError(
5720 'python-pyxattr is detected but is too old. '
5721 'youtube-dlc requires %s or above while your version is %s. '
5722 'Falling back to other xattr implementations' % (
5723 pyxattr_required_version
, xattr
.__version
__))
5725 setxattr
= xattr
.set
5727 setxattr
= xattr
.setxattr
5730 setxattr(path
, key
, value
)
5731 except EnvironmentError as e
:
5732 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5735 if compat_os_name
== 'nt':
5736 # Write xattrs to NTFS Alternate Data Streams:
5737 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
5738 assert ':' not in key
5739 assert os
.path
.exists(path
)
5741 ads_fn
= path
+ ':' + key
5743 with open(ads_fn
, 'wb') as f
:
5745 except EnvironmentError as e
:
5746 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5748 user_has_setfattr
= check_executable('setfattr', ['--version'])
5749 user_has_xattr
= check_executable('xattr', ['-h'])
5751 if user_has_setfattr
or user_has_xattr
:
5753 value
= value
.decode('utf-8')
5754 if user_has_setfattr
:
5755 executable
= 'setfattr'
5756 opts
= ['-n', key
, '-v', value
]
5757 elif user_has_xattr
:
5758 executable
= 'xattr'
5759 opts
= ['-w', key
, value
]
5761 cmd
= ([encodeFilename(executable
, True)]
5762 + [encodeArgument(o
) for o
in opts
]
5763 + [encodeFilename(path
, True)])
5766 p
= subprocess
.Popen(
5767 cmd
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
, stdin
=subprocess
.PIPE
)
5768 except EnvironmentError as e
:
5769 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5770 stdout
, stderr
= process_communicate_or_kill(p
)
5771 stderr
= stderr
.decode('utf-8', 'replace')
5772 if p
.returncode
!= 0:
5773 raise XAttrMetadataError(p
.returncode
, stderr
)
5776 # On Unix, and can't find pyxattr, setfattr, or xattr.
5777 if sys
.platform
.startswith('linux'):
5778 raise XAttrUnavailableError(
5779 "Couldn't find a tool to set the xattrs. "
5780 "Install either the python 'pyxattr' or 'xattr' "
5781 "modules, or the GNU 'attr' package "
5782 "(which contains the 'setfattr' tool).")
5784 raise XAttrUnavailableError(
5785 "Couldn't find a tool to set the xattrs. "
5786 "Install either the python 'xattr' module, "
5787 "or the 'xattr' binary.")
5790 def random_birthday(year_field
, month_field
, day_field
):
5791 start_date
= datetime
.date(1950, 1, 1)
5792 end_date
= datetime
.date(1995, 12, 31)
5793 offset
= random
.randint(0, (end_date
- start_date
).days
)
5794 random_date
= start_date
+ datetime
.timedelta(offset
)
5796 year_field
: str(random_date
.year
),
5797 month_field
: str(random_date
.month
),
5798 day_field
: str(random_date
.day
),
5802 # Templates for internet shortcut files, which are plain text files.
5803 DOT_URL_LINK_TEMPLATE
= '''
5808 DOT_WEBLOC_LINK_TEMPLATE
= '''
5809 <?xml version="1.0" encoding="UTF-8"?>
5810 <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
5811 <plist version="1.0">
5814 \t<string>%(url)s</string>
5819 DOT_DESKTOP_LINK_TEMPLATE
= '''
5829 def iri_to_uri(iri
):
5831 Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
5833 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.
5836 iri_parts
= compat_urllib_parse_urlparse(iri
)
5838 if '[' in iri_parts
.netloc
:
5839 raise ValueError('IPv6 URIs are not, yet, supported.')
5840 # Querying `.netloc`, when there's only one bracket, also raises a ValueError.
5842 # 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.
5845 if iri_parts
.username
:
5846 net_location
+= compat_urllib_parse_quote(iri_parts
.username
, safe
=r
"!$%&'()*+,~")
5847 if iri_parts
.password
is not None:
5848 net_location
+= ':' + compat_urllib_parse_quote(iri_parts
.password
, safe
=r
"!$%&'()*+,~")
5851 net_location
+= iri_parts
.hostname
.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
5852 # The 'idna' encoding produces ASCII text.
5853 if iri_parts
.port
is not None and iri_parts
.port
!= 80:
5854 net_location
+= ':' + str(iri_parts
.port
)
5856 return compat_urllib_parse_urlunparse(
5860 compat_urllib_parse_quote_plus(iri_parts
.path
, safe
=r
"!$%&'()*+,/:;=@|~"),
5862 # Unsure about the `safe` argument, since this is a legacy way of handling parameters.
5863 compat_urllib_parse_quote_plus(iri_parts
.params
, safe
=r
"!$%&'()*+,/:;=@|~"),
5865 # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
5866 compat_urllib_parse_quote_plus(iri_parts
.query
, safe
=r
"!$%&'()*+,/:;=?@{|}~"),
5868 compat_urllib_parse_quote_plus(iri_parts
.fragment
, safe
=r
"!#$%&'()*+,/:;=?@{|}~")))
5870 # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
5873 def to_high_limit_path(path
):
5874 if sys
.platform
in ['win32', 'cygwin']:
5875 # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
5876 return r
'\\?\ '.rstrip() + os
.path
.abspath(path
)
5881 def format_field(obj
, field
, template
='%s', ignore
=(None, ''), default
='', func
=None):
5882 val
= obj
.get(field
, default
)
5883 if func
and val
not in ignore
:
5885 return template
% val
if val
not in ignore
else default
5888 def clean_podcast_url(url
):
5889 return re
.sub(r
'''(?x)
5893 media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
5896 (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
5899 cn\.co| # https://podcorn.com/analytics-prefix/
5900 st\.fm # https://podsights.com/docs/
5905 _HEX_TABLE
= '0123456789abcdef'
5908 def random_uuidv4():
5909 return re
.sub(r
'[xy]', lambda x
: _HEX_TABLE
[random
.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
5912 def make_dir(path
, to_screen
=None):
5914 dn
= os
.path
.dirname(path
)
5915 if dn
and not os
.path
.exists(dn
):
5918 except (OSError, IOError) as err
:
5919 if callable(to_screen
) is not None:
5920 to_screen('unable to create directory ' + error_to_compat_str(err
))
5924 def get_executable_path():
5925 path
= os
.path
.dirname(sys
.argv
[0])
5926 if os
.path
.abspath(sys
.argv
[0]) != os
.path
.abspath(sys
.executable
): # Not packaged
5927 path
= os
.path
.join(path
, '..')
5928 return os
.path
.abspath(path
)
5931 def load_plugins(name
, type, namespace
):
5932 plugin_info
= [None]
5935 plugin_info
= imp
.find_module(
5936 name
, [os
.path
.join(get_executable_path(), 'ytdlp_plugins')])
5937 plugins
= imp
.load_module(name
, *plugin_info
)
5938 for name
in dir(plugins
):
5939 if not name
.endswith(type):
5941 klass
= getattr(plugins
, name
)
5942 classes
.append(klass
)
5943 namespace
[name
] = klass
5947 if plugin_info
[0] is not None:
5948 plugin_info
[0].close()
5952 def traverse_dict(dictn
, keys
, casesense
=True):
5953 if not isinstance(dictn
, dict):
5957 dictn
= {key.lower(): val for key, val in dictn.items()}
5958 first_key
= first_key
.lower()
5959 value
= dictn
.get(first_key
, None)
5960 return value
if len(keys
) < 2 else traverse_dict(value
, keys
[1:], casesense
)