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
,
62 compat_urllib_parse_urlencode
,
63 compat_urllib_parse_urlparse
,
64 compat_urllib_parse_urlunparse
,
65 compat_urllib_parse_quote
,
66 compat_urllib_parse_quote_plus
,
67 compat_urllib_parse_unquote_plus
,
68 compat_urllib_request
,
79 def register_socks_protocols():
80 # "Register" SOCKS protocols
81 # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
82 # URLs with protocols not in urlparse.uses_netloc are not handled correctly
83 for scheme
in ('socks', 'socks4', 'socks4a', 'socks5'):
84 if scheme
not in compat_urlparse
.uses_netloc
:
85 compat_urlparse
.uses_netloc
.append(scheme
)
88 # This is not clearly defined otherwise
89 compiled_regex_type
= type(re
.compile(''))
92 def random_user_agent():
93 _USER_AGENT_TPL
= 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
1672 return _USER_AGENT_TPL
% random
.choice(_CHROME_VERSIONS
)
1676 'User-Agent': random_user_agent(),
1677 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
1678 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
1679 'Accept-Encoding': 'gzip, deflate',
1680 'Accept-Language': 'en-us,en;q=0.5',
1685 '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',
1689 NO_DEFAULT
= object()
1691 ENGLISH_MONTH_NAMES
= [
1692 'January', 'February', 'March', 'April', 'May', 'June',
1693 'July', 'August', 'September', 'October', 'November', 'December']
1696 'en': ENGLISH_MONTH_NAMES
,
1698 'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
1699 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
1702 KNOWN_EXTENSIONS
= (
1703 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
1704 'flv', 'f4v', 'f4a', 'f4b',
1705 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
1706 'mkv', 'mka', 'mk3d',
1709 'asf', 'wmv', 'wma',
1715 'f4f', 'f4m', 'm3u8', 'smil')
1717 # needed for sanitizing filenames in restricted mode
1718 ACCENT_CHARS
= dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
1719 itertools
.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1720 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1743 '%Y/%m/%d %H:%M:%S',
1745 '%Y-%m-%d %H:%M:%S',
1746 '%Y-%m-%d %H:%M:%S.%f',
1749 '%Y-%m-%dT%H:%M:%SZ',
1750 '%Y-%m-%dT%H:%M:%S.%fZ',
1751 '%Y-%m-%dT%H:%M:%S.%f0Z',
1752 '%Y-%m-%dT%H:%M:%S',
1753 '%Y-%m-%dT%H:%M:%S.%f',
1755 '%b %d %Y at %H:%M',
1756 '%b %d %Y at %H:%M:%S',
1757 '%B %d %Y at %H:%M',
1758 '%B %d %Y at %H:%M:%S',
1761 DATE_FORMATS_DAY_FIRST
= list(DATE_FORMATS
)
1762 DATE_FORMATS_DAY_FIRST
.extend([
1768 '%d/%m/%Y %H:%M:%S',
1771 DATE_FORMATS_MONTH_FIRST
= list(DATE_FORMATS
)
1772 DATE_FORMATS_MONTH_FIRST
.extend([
1777 '%m/%d/%Y %H:%M:%S',
1780 PACKED_CODES_RE
= r
"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1781 JSON_LD_RE
= r
'(?is)<script[^>]+type=(["\']?
)application
/ld\
+json\
1[^
>]*>(?P
<json_ld
>.+?
)</script
>'
1784 def preferredencoding():
1785 """Get preferred encoding.
1787 Returns the best encoding scheme for the system, based on
1788 locale.getpreferredencoding() and some further tweaks.
1791 pref = locale.getpreferredencoding()
1799 def write_json_file(obj, fn):
1800 """ Encode obj as JSON and write it to fn, atomically if possible """
1802 fn = encodeFilename(fn)
1803 if sys.version_info < (3, 0) and sys.platform != 'win32
':
1804 encoding = get_filesystem_encoding()
1805 # os.path.basename returns a bytes object, but NamedTemporaryFile
1806 # will fail if the filename contains non ascii characters unless we
1807 # use a unicode object
1808 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1809 # the same for os.path.dirname
1810 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1812 path_basename = os.path.basename
1813 path_dirname = os.path.dirname
1817 'prefix
': path_basename(fn) + '.',
1818 'dir': path_dirname(fn),
1822 # In Python 2.x, json.dump expects a bytestream.
1823 # In Python 3.x, it writes to a character stream
1824 if sys.version_info < (3, 0):
1829 'encoding
': 'utf
-8',
1832 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1837 if sys.platform == 'win32
':
1838 # Need to remove existing file on Windows, else os.rename raises
1839 # WindowsError or FileExistsError.
1847 os.chmod(tf.name, 0o666 & ~mask)
1850 os.rename(tf.name, fn)
1859 if sys.version_info >= (2, 7):
1860 def find_xpath_attr(node, xpath, key, val=None):
1861 """ Find the xpath xpath[@key=val] """
1862 assert re.match(r'^
[a
-zA
-Z_
-]+$
', key)
1863 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1864 return node.find(expr)
1866 def find_xpath_attr(node, xpath, key, val=None):
1867 for f in node.findall(compat_xpath(xpath)):
1868 if key not in f.attrib:
1870 if val is None or f.attrib.get(key) == val:
1874 # On python2.6 the xml.etree.ElementTree.Element methods don't support
1875 # the namespace parameter
1878 def xpath_with_ns(path
, ns_map
):
1879 components
= [c
.split(':') for c
in path
.split('/')]
1881 for c
in components
:
1883 replaced
.append(c
[0])
1886 replaced
.append('{%s}%s' % (ns_map
[ns
], tag
))
1887 return '/'.join(replaced
)
1890 def xpath_element(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1891 def _find_xpath(xpath
):
1892 return node
.find(compat_xpath(xpath
))
1894 if isinstance(xpath
, (str, compat_str
)):
1895 n
= _find_xpath(xpath
)
1903 if default
is not NO_DEFAULT
:
1906 name
= xpath
if name
is None else name
1907 raise ExtractorError('Could not find XML element %s' % name
)
1913 def xpath_text(node
, xpath
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1914 n
= xpath_element(node
, xpath
, name
, fatal
=fatal
, default
=default
)
1915 if n
is None or n
== default
:
1918 if default
is not NO_DEFAULT
:
1921 name
= xpath
if name
is None else name
1922 raise ExtractorError('Could not find XML element\'s text %s' % name
)
1928 def xpath_attr(node
, xpath
, key
, name
=None, fatal
=False, default
=NO_DEFAULT
):
1929 n
= find_xpath_attr(node
, xpath
, key
)
1931 if default
is not NO_DEFAULT
:
1934 name
= '%s[@%s]' % (xpath
, key
) if name
is None else name
1935 raise ExtractorError('Could not find XML attribute %s' % name
)
1938 return n
.attrib
[key
]
1941 def get_element_by_id(id, html
):
1942 """Return the content of the tag with the specified ID in the passed HTML document"""
1943 return get_element_by_attribute('id', id, html
)
1946 def get_element_by_class(class_name
, html
):
1947 """Return the content of the first tag with the specified class in the passed HTML document"""
1948 retval
= get_elements_by_class(class_name
, html
)
1949 return retval
[0] if retval
else None
1952 def get_element_by_attribute(attribute
, value
, html
, escape_value
=True):
1953 retval
= get_elements_by_attribute(attribute
, value
, html
, escape_value
)
1954 return retval
[0] if retval
else None
1957 def get_elements_by_class(class_name
, html
):
1958 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1959 return get_elements_by_attribute(
1960 'class', r
'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1961 html, escape_value=False)
1964 def get_elements_by_attribute(attribute, value, html, escape_value=True):
1965 """Return the content of the tag with the specified attribute in the passed HTML document"""
1967 value = re.escape(value) if escape_value else value
1970 for m in re.finditer(r'''(?xs)
1972 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^
']*'|
))*?
1974 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^
"]*"|
='[^']*'|))*?
1978 ''' % (re.escape(attribute), value), html):
1979 res = m.group('content
')
1981 if res.startswith('"') or res.startswith("'"):
1984 retlist.append(unescapeHTML(res))
1989 class HTMLAttributeParser(compat_HTMLParser):
1990 """Trivial HTML parser to gather the attributes for a single element"""
1994 compat_HTMLParser.__init__(self)
1996 def handle_starttag(self, tag, attrs):
1997 self.attrs = dict(attrs)
2000 def extract_attributes(html_element):
2001 """Given a string for an HTML element such as
2003 a="foo" B="bar" c="&98;az" d=boz
2004 empty= noval entity="&"
2007 Decode and return a dictionary of attributes.
2009 'a
': 'foo
', 'b
': 'bar
', c: 'baz
', d: 'boz
',
2010 'empty
': '', 'noval
': None, 'entity
': '&',
2011 'sq
': '"', 'dq': '\''
2013 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2014 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2016 parser = HTMLAttributeParser()
2018 parser.feed(html_element)
2020 # Older Python may throw HTMLParseError in case of malformed HTML
2021 except compat_HTMLParseError:
2026 def clean_html(html):
2027 """Clean an HTML snippet into a readable string"""
2029 if html is None: # Convenience for sanitizing descriptions etc.
2033 html = html.replace('\n', ' ')
2034 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2035 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2037 html = re.sub('<.*?>', '', html)
2038 # Replace html entities
2039 html = unescapeHTML(html)
2043 def sanitize_open(filename, open_mode):
2044 """Try to open the given filename, and slightly tweak it if this fails.
2046 Attempts to open the given filename. If this fails, it tries to change
2047 the filename slightly, step by step, until it's either able to open it
2048 or it fails and raises a final exception, like the standard open()
2051 It returns the tuple (stream, definitive_file_name).
2055 if sys.platform == 'win32':
2057 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2058 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2059 stream = open(encodeFilename(filename), open_mode)
2060 return (stream, filename)
2061 except (IOError, OSError) as err:
2062 if err.errno in (errno.EACCES,):
2065 # In case of error, try to remove win32 forbidden chars
2066 alt_filename = sanitize_path(filename)
2067 if alt_filename == filename:
2070 # An exception here should be caught in the caller
2071 stream = open(encodeFilename(alt_filename), open_mode)
2072 return (stream, alt_filename)
2075 def timeconvert(timestr):
2076 """Convert RFC 2822 defined time string into system timestamp"""
2078 timetuple = email.utils.parsedate_tz(timestr)
2079 if timetuple is not None:
2080 timestamp = email.utils.mktime_tz(timetuple)
2084 def sanitize_filename(s, restricted=False, is_id=False):
2085 """Sanitizes a string so it could be used as part of a filename.
2086 If restricted is set, use a stricter subset of allowed characters.
2087 Set is_id if this is not an arbitrary string, but an ID that should be kept
2090 def replace_insane(char):
2091 if restricted and char in ACCENT_CHARS:
2092 return ACCENT_CHARS[char]
2093 if char == '?' or ord(char) < 32 or ord(char) == 127:
2096 return '' if restricted else '\''
2098 return '_
-' if restricted else ' -'
2099 elif char in '\\/|
*<>':
2101 if restricted and (char in '!&\'()[]{}$
;`^
,#' or char.isspace()):
2103 if restricted
and ord(char
) > 127:
2108 s
= re
.sub(r
'[0-9]+(?::[0-9]+)+', lambda m
: m
.group(0).replace(':', '_'), s
)
2109 result
= ''.join(map(replace_insane
, s
))
2111 while '__' in result
:
2112 result
= result
.replace('__', '_')
2113 result
= result
.strip('_')
2114 # Common case of "Foreign band name - English song title"
2115 if restricted
and result
.startswith('-_'):
2117 if result
.startswith('-'):
2118 result
= '_' + result
[len('-'):]
2119 result
= result
.lstrip('.')
2125 def sanitize_path(s
):
2126 """Sanitizes and normalizes path on Windows"""
2127 if sys
.platform
!= 'win32':
2129 drive_or_unc
, _
= os
.path
.splitdrive(s
)
2130 if sys
.version_info
< (2, 7) and not drive_or_unc
:
2131 drive_or_unc
, _
= os
.path
.splitunc(s
)
2132 norm_path
= os
.path
.normpath(remove_start(s
, drive_or_unc
)).split(os
.path
.sep
)
2136 path_part
if path_part
in ['.', '..'] else re
.sub(r
'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part
)
2137 for path_part
in norm_path
]
2139 sanitized_path
.insert(0, drive_or_unc
+ os
.path
.sep
)
2140 return os
.path
.join(*sanitized_path
)
2143 def sanitize_url(url
):
2144 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2145 # the number of unwanted failures due to missing protocol
2146 if url
.startswith('//'):
2147 return 'http:%s' % url
2148 # Fix some common typos seen so far
2150 # https://github.com/ytdl-org/youtube-dl/issues/15649
2151 (r
'^httpss://', r
'https://'),
2152 # https://bx1.be/lives/direct-tv/
2153 (r
'^rmtp([es]?)://', r
'rtmp\1://'),
2155 for mistake
, fixup
in COMMON_TYPOS
:
2156 if re
.match(mistake
, url
):
2157 return re
.sub(mistake
, fixup
, url
)
2161 def sanitized_Request(url
, *args
, **kwargs
):
2162 return compat_urllib_request
.Request(sanitize_url(url
), *args
, **kwargs
)
2166 """Expand shell variables and ~"""
2167 return os
.path
.expandvars(compat_expanduser(s
))
2170 def orderedSet(iterable
):
2171 """ Remove all duplicates from the input iterable """
2179 def _htmlentity_transform(entity_with_semicolon
):
2180 """Transforms an HTML entity to a character."""
2181 entity
= entity_with_semicolon
[:-1]
2183 # Known non-numeric HTML entity
2184 if entity
in compat_html_entities
.name2codepoint
:
2185 return compat_chr(compat_html_entities
.name2codepoint
[entity
])
2187 # TODO: HTML5 allows entities without a semicolon. For example,
2188 # 'Éric' should be decoded as 'Éric'.
2189 if entity_with_semicolon
in compat_html_entities_html5
:
2190 return compat_html_entities_html5
[entity_with_semicolon
]
2192 mobj
= re
.match(r
'#(x[0-9a-fA-F]+|[0-9]+)', entity
)
2193 if mobj
is not None:
2194 numstr
= mobj
.group(1)
2195 if numstr
.startswith('x'):
2197 numstr
= '0%s' % numstr
2200 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2202 return compat_chr(int(numstr
, base
))
2206 # Unknown entity in name, return its literal representation
2207 return '&%s;' % entity
2210 def unescapeHTML(s
):
2213 assert type(s
) == compat_str
2216 r
'&([^&;]+;)', lambda m
: _htmlentity_transform(m
.group(1)), s
)
2219 def process_communicate_or_kill(p
, *args
, **kwargs
):
2221 return p
.communicate(*args
, **kwargs
)
2222 except BaseException
: # Including KeyboardInterrupt
2228 def get_subprocess_encoding():
2229 if sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2230 # For subprocess calls, encode with locale encoding
2231 # Refer to http://stackoverflow.com/a/9951851/35070
2232 encoding
= preferredencoding()
2234 encoding
= sys
.getfilesystemencoding()
2235 if encoding
is None:
2240 def encodeFilename(s
, for_subprocess
=False):
2242 @param s The name of the file
2245 assert type(s
) == compat_str
2247 # Python 3 has a Unicode API
2248 if sys
.version_info
>= (3, 0):
2251 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2252 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2253 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2254 if not for_subprocess
and sys
.platform
== 'win32' and sys
.getwindowsversion()[0] >= 5:
2257 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2258 if sys
.platform
.startswith('java'):
2261 return s
.encode(get_subprocess_encoding(), 'ignore')
2264 def decodeFilename(b
, for_subprocess
=False):
2266 if sys
.version_info
>= (3, 0):
2269 if not isinstance(b
, bytes):
2272 return b
.decode(get_subprocess_encoding(), 'ignore')
2275 def encodeArgument(s
):
2276 if not isinstance(s
, compat_str
):
2277 # Legacy code that uses byte strings
2278 # Uncomment the following line after fixing all post processors
2279 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2280 s
= s
.decode('ascii')
2281 return encodeFilename(s
, True)
2284 def decodeArgument(b
):
2285 return decodeFilename(b
, True)
2288 def decodeOption(optval
):
2291 if isinstance(optval
, bytes):
2292 optval
= optval
.decode(preferredencoding())
2294 assert isinstance(optval
, compat_str
)
2298 def formatSeconds(secs
, delim
=':'):
2300 return '%d%s%02d%s%02d' % (secs
// 3600, delim
, (secs
% 3600) // 60, delim
, secs
% 60)
2302 return '%d%s%02d' % (secs
// 60, delim
, secs
% 60)
2307 def make_HTTPS_handler(params
, **kwargs
):
2308 opts_no_check_certificate
= params
.get('nocheckcertificate', False)
2309 if hasattr(ssl
, 'create_default_context'): # Python >= 3.4 or 2.7.9
2310 context
= ssl
.create_default_context(ssl
.Purpose
.SERVER_AUTH
)
2311 if opts_no_check_certificate
:
2312 context
.check_hostname
= False
2313 context
.verify_mode
= ssl
.CERT_NONE
2315 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2318 # (create_default_context present but HTTPSHandler has no context=)
2321 if sys
.version_info
< (3, 2):
2322 return YoutubeDLHTTPSHandler(params
, **kwargs
)
2323 else: # Python < 3.4
2324 context
= ssl
.SSLContext(ssl
.PROTOCOL_TLSv1
)
2325 context
.verify_mode
= (ssl
.CERT_NONE
2326 if opts_no_check_certificate
2327 else ssl
.CERT_REQUIRED
)
2328 context
.set_default_verify_paths()
2329 return YoutubeDLHTTPSHandler(params
, context
=context
, **kwargs
)
2332 def bug_reports_message():
2333 if ytdl_is_updateable():
2334 update_cmd
= 'type youtube-dlc -U to update'
2336 update_cmd
= 'see https://github.com/pukkandan/yt-dlp on how to update'
2337 msg
= '; please report this issue on https://github.com/pukkandan/yt-dlp .'
2338 msg
+= ' Make sure you are using the latest version; %s.' % update_cmd
2339 msg
+= ' Be sure to call youtube-dlc with the --verbose flag and include its complete output.'
2343 class YoutubeDLError(Exception):
2344 """Base exception for YoutubeDL errors."""
2348 class ExtractorError(YoutubeDLError
):
2349 """Error during info extraction."""
2351 def __init__(self
, msg
, tb
=None, expected
=False, cause
=None, video_id
=None):
2352 """ tb, if given, is the original traceback (so that it can be printed out).
2353 If expected is set, this is a normal error message and most likely not a bug in youtube-dlc.
2356 if sys
.exc_info()[0] in (compat_urllib_error
.URLError
, socket
.timeout
, UnavailableVideoError
):
2358 if video_id
is not None:
2359 msg
= video_id
+ ': ' + msg
2361 msg
+= ' (caused by %r)' % cause
2363 msg
+= bug_reports_message()
2364 super(ExtractorError
, self
).__init
__(msg
)
2367 self
.exc_info
= sys
.exc_info() # preserve original exception
2369 self
.video_id
= video_id
2371 def format_traceback(self
):
2372 if self
.traceback
is None:
2374 return ''.join(traceback
.format_tb(self
.traceback
))
2377 class UnsupportedError(ExtractorError
):
2378 def __init__(self
, url
):
2379 super(UnsupportedError
, self
).__init
__(
2380 'Unsupported URL: %s' % url
, expected
=True)
2384 class RegexNotFoundError(ExtractorError
):
2385 """Error when a regex didn't match"""
2389 class GeoRestrictedError(ExtractorError
):
2390 """Geographic restriction Error exception.
2392 This exception may be thrown when a video is not available from your
2393 geographic location due to geographic restrictions imposed by a website.
2396 def __init__(self
, msg
, countries
=None):
2397 super(GeoRestrictedError
, self
).__init
__(msg
, expected
=True)
2399 self
.countries
= countries
2402 class DownloadError(YoutubeDLError
):
2403 """Download Error exception.
2405 This exception may be thrown by FileDownloader objects if they are not
2406 configured to continue on errors. They will contain the appropriate
2410 def __init__(self
, msg
, exc_info
=None):
2411 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2412 super(DownloadError
, self
).__init
__(msg
)
2413 self
.exc_info
= exc_info
2416 class SameFileError(YoutubeDLError
):
2417 """Same File exception.
2419 This exception will be thrown by FileDownloader objects if they detect
2420 multiple files would have to be downloaded to the same file on disk.
2425 class PostProcessingError(YoutubeDLError
):
2426 """Post Processing exception.
2428 This exception may be raised by PostProcessor's .run() method to
2429 indicate an error in the postprocessing task.
2432 def __init__(self
, msg
):
2433 super(PostProcessingError
, self
).__init
__(msg
)
2437 class ExistingVideoReached(YoutubeDLError
):
2438 """ --max-downloads limit has been reached. """
2442 class RejectedVideoReached(YoutubeDLError
):
2443 """ --max-downloads limit has been reached. """
2447 class MaxDownloadsReached(YoutubeDLError
):
2448 """ --max-downloads limit has been reached. """
2452 class UnavailableVideoError(YoutubeDLError
):
2453 """Unavailable Format exception.
2455 This exception will be thrown when a video is requested
2456 in a format that is not available for that video.
2461 class ContentTooShortError(YoutubeDLError
):
2462 """Content Too Short exception.
2464 This exception may be raised by FileDownloader objects when a file they
2465 download is too small for what the server announced first, indicating
2466 the connection was probably interrupted.
2469 def __init__(self
, downloaded
, expected
):
2470 super(ContentTooShortError
, self
).__init
__(
2471 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded
, expected
)
2474 self
.downloaded
= downloaded
2475 self
.expected
= expected
2478 class XAttrMetadataError(YoutubeDLError
):
2479 def __init__(self
, code
=None, msg
='Unknown error'):
2480 super(XAttrMetadataError
, self
).__init
__(msg
)
2484 # Parsing code and msg
2485 if (self
.code
in (errno
.ENOSPC
, errno
.EDQUOT
)
2486 or 'No space left' in self
.msg
or 'Disk quota exceeded' in self
.msg
):
2487 self
.reason
= 'NO_SPACE'
2488 elif self
.code
== errno
.E2BIG
or 'Argument list too long' in self
.msg
:
2489 self
.reason
= 'VALUE_TOO_LONG'
2491 self
.reason
= 'NOT_SUPPORTED'
2494 class XAttrUnavailableError(YoutubeDLError
):
2498 def _create_http_connection(ydl_handler
, http_class
, is_https
, *args
, **kwargs
):
2499 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2500 # expected HTTP responses to meet HTTP/1.0 or later (see also
2501 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2502 if sys
.version_info
< (3, 0):
2503 kwargs
['strict'] = True
2504 hc
= http_class(*args
, **compat_kwargs(kwargs
))
2505 source_address
= ydl_handler
._params
.get('source_address')
2507 if source_address
is not None:
2508 # This is to workaround _create_connection() from socket where it will try all
2509 # address data from getaddrinfo() including IPv6. This filters the result from
2510 # getaddrinfo() based on the source_address value.
2511 # This is based on the cpython socket.create_connection() function.
2512 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2513 def _create_connection(address
, timeout
=socket
._GLOBAL
_DEFAULT
_TIMEOUT
, source_address
=None):
2514 host
, port
= address
2516 addrs
= socket
.getaddrinfo(host
, port
, 0, socket
.SOCK_STREAM
)
2517 af
= socket
.AF_INET
if '.' in source_address
[0] else socket
.AF_INET6
2518 ip_addrs
= [addr
for addr
in addrs
if addr
[0] == af
]
2519 if addrs
and not ip_addrs
:
2520 ip_version
= 'v4' if af
== socket
.AF_INET
else 'v6'
2522 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2523 % (ip_version
, source_address
[0]))
2524 for res
in ip_addrs
:
2525 af
, socktype
, proto
, canonname
, sa
= res
2528 sock
= socket
.socket(af
, socktype
, proto
)
2529 if timeout
is not socket
._GLOBAL
_DEFAULT
_TIMEOUT
:
2530 sock
.settimeout(timeout
)
2531 sock
.bind(source_address
)
2533 err
= None # Explicitly break reference cycle
2535 except socket
.error
as _
:
2537 if sock
is not None:
2542 raise socket
.error('getaddrinfo returns an empty list')
2543 if hasattr(hc
, '_create_connection'):
2544 hc
._create
_connection
= _create_connection
2545 sa
= (source_address
, 0)
2546 if hasattr(hc
, 'source_address'): # Python 2.7+
2547 hc
.source_address
= sa
2549 def _hc_connect(self
, *args
, **kwargs
):
2550 sock
= _create_connection(
2551 (self
.host
, self
.port
), self
.timeout
, sa
)
2553 self
.sock
= ssl
.wrap_socket(
2554 sock
, self
.key_file
, self
.cert_file
,
2555 ssl_version
=ssl
.PROTOCOL_TLSv1
)
2558 hc
.connect
= functools
.partial(_hc_connect
, hc
)
2563 def handle_youtubedl_headers(headers
):
2564 filtered_headers
= headers
2566 if 'Youtubedl-no-compression' in filtered_headers
:
2567 filtered_headers
= dict((k
, v
) for k
, v
in filtered_headers
.items() if k
.lower() != 'accept-encoding')
2568 del filtered_headers
['Youtubedl-no-compression']
2570 return filtered_headers
2573 class YoutubeDLHandler(compat_urllib_request
.HTTPHandler
):
2574 """Handler for HTTP requests and responses.
2576 This class, when installed with an OpenerDirector, automatically adds
2577 the standard headers to every HTTP request and handles gzipped and
2578 deflated responses from web servers. If compression is to be avoided in
2579 a particular request, the original request in the program code only has
2580 to include the HTTP header "Youtubedl-no-compression", which will be
2581 removed before making the real request.
2583 Part of this code was copied from:
2585 http://techknack.net/python-urllib2-handlers/
2587 Andrew Rowls, the author of that code, agreed to release it to the
2591 def __init__(self
, params
, *args
, **kwargs
):
2592 compat_urllib_request
.HTTPHandler
.__init
__(self
, *args
, **kwargs
)
2593 self
._params
= params
2595 def http_open(self
, req
):
2596 conn_class
= compat_http_client
.HTTPConnection
2598 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2600 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2601 del req
.headers
['Ytdl-socks-proxy']
2603 return self
.do_open(functools
.partial(
2604 _create_http_connection
, self
, conn_class
, False),
2610 return zlib
.decompress(data
, -zlib
.MAX_WBITS
)
2612 return zlib
.decompress(data
)
2614 def http_request(self
, req
):
2615 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2616 # always respected by websites, some tend to give out URLs with non percent-encoded
2617 # non-ASCII characters (see telemb.py, ard.py [#3412])
2618 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2619 # To work around aforementioned issue we will replace request's original URL with
2620 # percent-encoded one
2621 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2622 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2623 url
= req
.get_full_url()
2624 url_escaped
= escape_url(url
)
2626 # Substitute URL if any change after escaping
2627 if url
!= url_escaped
:
2628 req
= update_Request(req
, url
=url_escaped
)
2630 for h
, v
in std_headers
.items():
2631 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2632 # The dict keys are capitalized because of this bug by urllib
2633 if h
.capitalize() not in req
.headers
:
2634 req
.add_header(h
, v
)
2636 req
.headers
= handle_youtubedl_headers(req
.headers
)
2638 if sys
.version_info
< (2, 7) and '#' in req
.get_full_url():
2639 # Python 2.6 is brain-dead when it comes to fragments
2640 req
._Request
__original
= req
._Request
__original
.partition('#')[0]
2641 req
._Request
__r
_type
= req
._Request
__r
_type
.partition('#')[0]
2645 def http_response(self
, req
, resp
):
2648 if resp
.headers
.get('Content-encoding', '') == 'gzip':
2649 content
= resp
.read()
2650 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
), mode
='rb')
2652 uncompressed
= io
.BytesIO(gz
.read())
2653 except IOError as original_ioerror
:
2654 # There may be junk add the end of the file
2655 # See http://stackoverflow.com/q/4928560/35070 for details
2656 for i
in range(1, 1024):
2658 gz
= gzip
.GzipFile(fileobj
=io
.BytesIO(content
[:-i
]), mode
='rb')
2659 uncompressed
= io
.BytesIO(gz
.read())
2664 raise original_ioerror
2665 resp
= compat_urllib_request
.addinfourl(uncompressed
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2666 resp
.msg
= old_resp
.msg
2667 del resp
.headers
['Content-encoding']
2669 if resp
.headers
.get('Content-encoding', '') == 'deflate':
2670 gz
= io
.BytesIO(self
.deflate(resp
.read()))
2671 resp
= compat_urllib_request
.addinfourl(gz
, old_resp
.headers
, old_resp
.url
, old_resp
.code
)
2672 resp
.msg
= old_resp
.msg
2673 del resp
.headers
['Content-encoding']
2674 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2675 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2676 if 300 <= resp
.code
< 400:
2677 location
= resp
.headers
.get('Location')
2679 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2680 if sys
.version_info
>= (3, 0):
2681 location
= location
.encode('iso-8859-1').decode('utf-8')
2683 location
= location
.decode('utf-8')
2684 location_escaped
= escape_url(location
)
2685 if location
!= location_escaped
:
2686 del resp
.headers
['Location']
2687 if sys
.version_info
< (3, 0):
2688 location_escaped
= location_escaped
.encode('utf-8')
2689 resp
.headers
['Location'] = location_escaped
2692 https_request
= http_request
2693 https_response
= http_response
2696 def make_socks_conn_class(base_class
, socks_proxy
):
2697 assert issubclass(base_class
, (
2698 compat_http_client
.HTTPConnection
, compat_http_client
.HTTPSConnection
))
2700 url_components
= compat_urlparse
.urlparse(socks_proxy
)
2701 if url_components
.scheme
.lower() == 'socks5':
2702 socks_type
= ProxyType
.SOCKS5
2703 elif url_components
.scheme
.lower() in ('socks', 'socks4'):
2704 socks_type
= ProxyType
.SOCKS4
2705 elif url_components
.scheme
.lower() == 'socks4a':
2706 socks_type
= ProxyType
.SOCKS4A
2708 def unquote_if_non_empty(s
):
2711 return compat_urllib_parse_unquote_plus(s
)
2715 url_components
.hostname
, url_components
.port
or 1080,
2717 unquote_if_non_empty(url_components
.username
),
2718 unquote_if_non_empty(url_components
.password
),
2721 class SocksConnection(base_class
):
2723 self
.sock
= sockssocket()
2724 self
.sock
.setproxy(*proxy_args
)
2725 if type(self
.timeout
) in (int, float):
2726 self
.sock
.settimeout(self
.timeout
)
2727 self
.sock
.connect((self
.host
, self
.port
))
2729 if isinstance(self
, compat_http_client
.HTTPSConnection
):
2730 if hasattr(self
, '_context'): # Python > 2.6
2731 self
.sock
= self
._context
.wrap_socket(
2732 self
.sock
, server_hostname
=self
.host
)
2734 self
.sock
= ssl
.wrap_socket(self
.sock
)
2736 return SocksConnection
2739 class YoutubeDLHTTPSHandler(compat_urllib_request
.HTTPSHandler
):
2740 def __init__(self
, params
, https_conn_class
=None, *args
, **kwargs
):
2741 compat_urllib_request
.HTTPSHandler
.__init
__(self
, *args
, **kwargs
)
2742 self
._https
_conn
_class
= https_conn_class
or compat_http_client
.HTTPSConnection
2743 self
._params
= params
2745 def https_open(self
, req
):
2747 conn_class
= self
._https
_conn
_class
2749 if hasattr(self
, '_context'): # python > 2.6
2750 kwargs
['context'] = self
._context
2751 if hasattr(self
, '_check_hostname'): # python 3.x
2752 kwargs
['check_hostname'] = self
._check
_hostname
2754 socks_proxy
= req
.headers
.get('Ytdl-socks-proxy')
2756 conn_class
= make_socks_conn_class(conn_class
, socks_proxy
)
2757 del req
.headers
['Ytdl-socks-proxy']
2759 return self
.do_open(functools
.partial(
2760 _create_http_connection
, self
, conn_class
, True),
2764 class YoutubeDLCookieJar(compat_cookiejar
.MozillaCookieJar
):
2766 See [1] for cookie file format.
2768 1. https://curl.haxx.se/docs/http-cookies.html
2770 _HTTPONLY_PREFIX
= '#HttpOnly_'
2772 _HEADER
= '''# Netscape HTTP Cookie File
2773 # This file is generated by youtube-dlc. Do not edit.
2776 _CookieFileEntry
= collections
.namedtuple(
2778 ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
2780 def save(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2782 Save cookies to a file.
2784 Most of the code is taken from CPython 3.8 and slightly adapted
2785 to support cookie files with UTF-8 in both python 2 and 3.
2787 if filename
is None:
2788 if self
.filename
is not None:
2789 filename
= self
.filename
2791 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2793 # Store session cookies with `expires` set to 0 instead of an empty
2796 if cookie
.expires
is None:
2799 with io
.open(filename
, 'w', encoding
='utf-8') as f
:
2800 f
.write(self
._HEADER
)
2803 if not ignore_discard
and cookie
.discard
:
2805 if not ignore_expires
and cookie
.is_expired(now
):
2811 if cookie
.domain
.startswith('.'):
2812 initial_dot
= 'TRUE'
2814 initial_dot
= 'FALSE'
2815 if cookie
.expires
is not None:
2816 expires
= compat_str(cookie
.expires
)
2819 if cookie
.value
is None:
2820 # cookies.txt regards 'Set-Cookie: foo' as a cookie
2821 # with no name, whereas http.cookiejar regards it as a
2822 # cookie with no value.
2827 value
= cookie
.value
2829 '\t'.join([cookie
.domain
, initial_dot
, cookie
.path
,
2830 secure
, expires
, name
, value
]) + '\n')
2832 def load(self
, filename
=None, ignore_discard
=False, ignore_expires
=False):
2833 """Load cookies from a file."""
2834 if filename
is None:
2835 if self
.filename
is not None:
2836 filename
= self
.filename
2838 raise ValueError(compat_cookiejar
.MISSING_FILENAME_TEXT
)
2840 def prepare_line(line
):
2841 if line
.startswith(self
._HTTPONLY
_PREFIX
):
2842 line
= line
[len(self
._HTTPONLY
_PREFIX
):]
2843 # comments and empty lines are fine
2844 if line
.startswith('#') or not line
.strip():
2846 cookie_list
= line
.split('\t')
2847 if len(cookie_list
) != self
._ENTRY
_LEN
:
2848 raise compat_cookiejar
.LoadError('invalid length %d' % len(cookie_list
))
2849 cookie
= self
._CookieFileEntry
(*cookie_list
)
2850 if cookie
.expires_at
and not cookie
.expires_at
.isdigit():
2851 raise compat_cookiejar
.LoadError('invalid expires at %s' % cookie
.expires_at
)
2855 with io
.open(filename
, encoding
='utf-8') as f
:
2858 cf
.write(prepare_line(line
))
2859 except compat_cookiejar
.LoadError
as e
:
2861 'WARNING: skipping cookie file entry due to %s: %r\n'
2862 % (e
, line
), sys
.stderr
)
2865 self
._really
_load
(cf
, filename
, ignore_discard
, ignore_expires
)
2866 # Session cookies are denoted by either `expires` field set to
2867 # an empty string or 0. MozillaCookieJar only recognizes the former
2868 # (see [1]). So we need force the latter to be recognized as session
2869 # cookies on our own.
2870 # Session cookies may be important for cookies-based authentication,
2871 # e.g. usually, when user does not check 'Remember me' check box while
2872 # logging in on a site, some important cookies are stored as session
2873 # cookies so that not recognizing them will result in failed login.
2874 # 1. https://bugs.python.org/issue17164
2876 # Treat `expires=0` cookies as session cookies
2877 if cookie
.expires
== 0:
2878 cookie
.expires
= None
2879 cookie
.discard
= True
2882 class YoutubeDLCookieProcessor(compat_urllib_request
.HTTPCookieProcessor
):
2883 def __init__(self
, cookiejar
=None):
2884 compat_urllib_request
.HTTPCookieProcessor
.__init
__(self
, cookiejar
)
2886 def http_response(self
, request
, response
):
2887 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2888 # characters in Set-Cookie HTTP header of last response (see
2889 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2890 # In order to at least prevent crashing we will percent encode Set-Cookie
2891 # header before HTTPCookieProcessor starts processing it.
2892 # if sys.version_info < (3, 0) and response.headers:
2893 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2894 # set_cookie = response.headers.get(set_cookie_header)
2896 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2897 # if set_cookie != set_cookie_escaped:
2898 # del response.headers[set_cookie_header]
2899 # response.headers[set_cookie_header] = set_cookie_escaped
2900 return compat_urllib_request
.HTTPCookieProcessor
.http_response(self
, request
, response
)
2902 https_request
= compat_urllib_request
.HTTPCookieProcessor
.http_request
2903 https_response
= http_response
2906 class YoutubeDLRedirectHandler(compat_urllib_request
.HTTPRedirectHandler
):
2907 if sys
.version_info
[0] < 3:
2908 def redirect_request(self
, req
, fp
, code
, msg
, headers
, newurl
):
2909 # On python 2 urlh.geturl() may sometimes return redirect URL
2910 # as byte string instead of unicode. This workaround allows
2911 # to force it always return unicode.
2912 return compat_urllib_request
.HTTPRedirectHandler
.redirect_request(self
, req
, fp
, code
, msg
, headers
, compat_str(newurl
))
2915 def extract_timezone(date_str
):
2917 r
'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
2920 timezone
= datetime
.timedelta()
2922 date_str
= date_str
[:-len(m
.group('tz'))]
2923 if not m
.group('sign'):
2924 timezone
= datetime
.timedelta()
2926 sign
= 1 if m
.group('sign') == '+' else -1
2927 timezone
= datetime
.timedelta(
2928 hours
=sign
* int(m
.group('hours')),
2929 minutes
=sign
* int(m
.group('minutes')))
2930 return timezone
, date_str
2933 def parse_iso8601(date_str
, delimiter
='T', timezone
=None):
2934 """ Return a UNIX timestamp from the given date """
2936 if date_str
is None:
2939 date_str
= re
.sub(r
'\.[0-9]+', '', date_str
)
2941 if timezone
is None:
2942 timezone
, date_str
= extract_timezone(date_str
)
2945 date_format
= '%Y-%m-%d{0}%H:%M:%S'.format(delimiter
)
2946 dt
= datetime
.datetime
.strptime(date_str
, date_format
) - timezone
2947 return calendar
.timegm(dt
.timetuple())
2952 def date_formats(day_first
=True):
2953 return DATE_FORMATS_DAY_FIRST
if day_first
else DATE_FORMATS_MONTH_FIRST
2956 def unified_strdate(date_str
, day_first
=True):
2957 """Return a string with the date in the format YYYYMMDD"""
2959 if date_str
is None:
2963 date_str
= date_str
.replace(',', ' ')
2964 # Remove AM/PM + timezone
2965 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
2966 _
, date_str
= extract_timezone(date_str
)
2968 for expression
in date_formats(day_first
):
2970 upload_date
= datetime
.datetime
.strptime(date_str
, expression
).strftime('%Y%m%d')
2973 if upload_date
is None:
2974 timetuple
= email
.utils
.parsedate_tz(date_str
)
2977 upload_date
= datetime
.datetime(*timetuple
[:6]).strftime('%Y%m%d')
2980 if upload_date
is not None:
2981 return compat_str(upload_date
)
2984 def unified_timestamp(date_str
, day_first
=True):
2985 if date_str
is None:
2988 date_str
= re
.sub(r
'[,|]', '', date_str
)
2990 pm_delta
= 12 if re
.search(r
'(?i)PM', date_str
) else 0
2991 timezone
, date_str
= extract_timezone(date_str
)
2993 # Remove AM/PM + timezone
2994 date_str
= re
.sub(r
'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str
)
2996 # Remove unrecognized timezones from ISO 8601 alike timestamps
2997 m
= re
.search(r
'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str
)
2999 date_str
= date_str
[:-len(m
.group('tz'))]
3001 # Python only supports microseconds, so remove nanoseconds
3002 m
= re
.search(r
'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str
)
3004 date_str
= m
.group(1)
3006 for expression
in date_formats(day_first
):
3008 dt
= datetime
.datetime
.strptime(date_str
, expression
) - timezone
+ datetime
.timedelta(hours
=pm_delta
)
3009 return calendar
.timegm(dt
.timetuple())
3012 timetuple
= email
.utils
.parsedate_tz(date_str
)
3014 return calendar
.timegm(timetuple
) + pm_delta
* 3600
3017 def determine_ext(url
, default_ext
='unknown_video'):
3018 if url
is None or '.' not in url
:
3020 guess
= url
.partition('?')[0].rpartition('.')[2]
3021 if re
.match(r
'^[A-Za-z0-9]+$', guess
):
3023 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
3024 elif guess
.rstrip('/') in KNOWN_EXTENSIONS
:
3025 return guess
.rstrip('/')
3030 def subtitles_filename(filename
, sub_lang
, sub_format
, expected_real_ext
=None):
3031 return replace_extension(filename
, sub_lang
+ '.' + sub_format
, expected_real_ext
)
3034 def date_from_str(date_str
):
3036 Return a datetime object from a string in the format YYYYMMDD or
3037 (now|today)[+-][0-9](day|week|month|year)(s)?"""
3038 today
= datetime
.date
.today()
3039 if date_str
in ('now', 'today'):
3041 if date_str
== 'yesterday':
3042 return today
- datetime
.timedelta(days
=1)
3043 match
= re
.match(r
'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str
)
3044 if match
is not None:
3045 sign
= match
.group('sign')
3046 time
= int(match
.group('time'))
3049 unit
= match
.group('unit')
3050 # A bad approximation?
3054 elif unit
== 'year':
3058 delta
= datetime
.timedelta(**{unit: time}
)
3059 return today
+ delta
3060 return datetime
.datetime
.strptime(date_str
, '%Y%m%d').date()
3063 def hyphenate_date(date_str
):
3065 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
3066 match
= re
.match(r
'^(\d\d\d\d)(\d\d)(\d\d)$', date_str
)
3067 if match
is not None:
3068 return '-'.join(match
.groups())
3073 class DateRange(object):
3074 """Represents a time interval between two dates"""
3076 def __init__(self
, start
=None, end
=None):
3077 """start and end must be strings in the format accepted by date"""
3078 if start
is not None:
3079 self
.start
= date_from_str(start
)
3081 self
.start
= datetime
.datetime
.min.date()
3083 self
.end
= date_from_str(end
)
3085 self
.end
= datetime
.datetime
.max.date()
3086 if self
.start
> self
.end
:
3087 raise ValueError('Date range: "%s" , the start date must be before the end date' % self
)
3091 """Returns a range that only contains the given day"""
3092 return cls(day
, day
)
3094 def __contains__(self
, date
):
3095 """Check if the date is in the range"""
3096 if not isinstance(date
, datetime
.date
):
3097 date
= date_from_str(date
)
3098 return self
.start
<= date
<= self
.end
3101 return '%s - %s' % (self
.start
.isoformat(), self
.end
.isoformat())
3104 def platform_name():
3105 """ Returns the platform name as a compat_str """
3106 res
= platform
.platform()
3107 if isinstance(res
, bytes):
3108 res
= res
.decode(preferredencoding())
3110 assert isinstance(res
, compat_str
)
3114 def _windows_write_string(s
, out
):
3115 """ Returns True if the string was written using special methods,
3116 False if it has yet to be written out."""
3117 # Adapted from http://stackoverflow.com/a/3259271/35070
3120 import ctypes
.wintypes
3128 fileno
= out
.fileno()
3129 except AttributeError:
3130 # If the output stream doesn't have a fileno, it's virtual
3132 except io
.UnsupportedOperation
:
3133 # Some strange Windows pseudo files?
3135 if fileno
not in WIN_OUTPUT_IDS
:
3138 GetStdHandle
= compat_ctypes_WINFUNCTYPE(
3139 ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.DWORD
)(
3140 ('GetStdHandle', ctypes
.windll
.kernel32
))
3141 h
= GetStdHandle(WIN_OUTPUT_IDS
[fileno
])
3143 WriteConsoleW
= compat_ctypes_WINFUNCTYPE(
3144 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
, ctypes
.wintypes
.LPWSTR
,
3145 ctypes
.wintypes
.DWORD
, ctypes
.POINTER(ctypes
.wintypes
.DWORD
),
3146 ctypes
.wintypes
.LPVOID
)(('WriteConsoleW', ctypes
.windll
.kernel32
))
3147 written
= ctypes
.wintypes
.DWORD(0)
3149 GetFileType
= compat_ctypes_WINFUNCTYPE(ctypes
.wintypes
.DWORD
, ctypes
.wintypes
.DWORD
)(('GetFileType', ctypes
.windll
.kernel32
))
3150 FILE_TYPE_CHAR
= 0x0002
3151 FILE_TYPE_REMOTE
= 0x8000
3152 GetConsoleMode
= compat_ctypes_WINFUNCTYPE(
3153 ctypes
.wintypes
.BOOL
, ctypes
.wintypes
.HANDLE
,
3154 ctypes
.POINTER(ctypes
.wintypes
.DWORD
))(
3155 ('GetConsoleMode', ctypes
.windll
.kernel32
))
3156 INVALID_HANDLE_VALUE
= ctypes
.wintypes
.DWORD(-1).value
3158 def not_a_console(handle
):
3159 if handle
== INVALID_HANDLE_VALUE
or handle
is None:
3161 return ((GetFileType(handle
) & ~FILE_TYPE_REMOTE
) != FILE_TYPE_CHAR
3162 or GetConsoleMode(handle
, ctypes
.byref(ctypes
.wintypes
.DWORD())) == 0)
3164 if not_a_console(h
):
3167 def next_nonbmp_pos(s
):
3169 return next(i
for i
, c
in enumerate(s
) if ord(c
) > 0xffff)
3170 except StopIteration:
3174 count
= min(next_nonbmp_pos(s
), 1024)
3176 ret
= WriteConsoleW(
3177 h
, s
, count
if count
else 2, ctypes
.byref(written
), None)
3179 raise OSError('Failed to write string')
3180 if not count
: # We just wrote a non-BMP character
3181 assert written
.value
== 2
3184 assert written
.value
> 0
3185 s
= s
[written
.value
:]
3189 def write_string(s
, out
=None, encoding
=None):
3192 assert type(s
) == compat_str
3194 if sys
.platform
== 'win32' and encoding
is None and hasattr(out
, 'fileno'):
3195 if _windows_write_string(s
, out
):
3198 if ('b' in getattr(out
, 'mode', '')
3199 or sys
.version_info
[0] < 3): # Python 2 lies about mode of sys.stderr
3200 byt
= s
.encode(encoding
or preferredencoding(), 'ignore')
3202 elif hasattr(out
, 'buffer'):
3203 enc
= encoding
or getattr(out
, 'encoding', None) or preferredencoding()
3204 byt
= s
.encode(enc
, 'ignore')
3205 out
.buffer.write(byt
)
3211 def bytes_to_intlist(bs
):
3214 if isinstance(bs
[0], int): # Python 3
3217 return [ord(c
) for c
in bs
]
3220 def intlist_to_bytes(xs
):
3223 return compat_struct_pack('%dB' % len(xs
), *xs
)
3226 # Cross-platform file locking
3227 if sys
.platform
== 'win32':
3228 import ctypes
.wintypes
3231 class OVERLAPPED(ctypes
.Structure
):
3233 ('Internal', ctypes
.wintypes
.LPVOID
),
3234 ('InternalHigh', ctypes
.wintypes
.LPVOID
),
3235 ('Offset', ctypes
.wintypes
.DWORD
),
3236 ('OffsetHigh', ctypes
.wintypes
.DWORD
),
3237 ('hEvent', ctypes
.wintypes
.HANDLE
),
3240 kernel32
= ctypes
.windll
.kernel32
3241 LockFileEx
= kernel32
.LockFileEx
3242 LockFileEx
.argtypes
= [
3243 ctypes
.wintypes
.HANDLE
, # hFile
3244 ctypes
.wintypes
.DWORD
, # dwFlags
3245 ctypes
.wintypes
.DWORD
, # dwReserved
3246 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3247 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3248 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3250 LockFileEx
.restype
= ctypes
.wintypes
.BOOL
3251 UnlockFileEx
= kernel32
.UnlockFileEx
3252 UnlockFileEx
.argtypes
= [
3253 ctypes
.wintypes
.HANDLE
, # hFile
3254 ctypes
.wintypes
.DWORD
, # dwReserved
3255 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockLow
3256 ctypes
.wintypes
.DWORD
, # nNumberOfBytesToLockHigh
3257 ctypes
.POINTER(OVERLAPPED
) # Overlapped
3259 UnlockFileEx
.restype
= ctypes
.wintypes
.BOOL
3260 whole_low
= 0xffffffff
3261 whole_high
= 0x7fffffff
3263 def _lock_file(f
, exclusive
):
3264 overlapped
= OVERLAPPED()
3265 overlapped
.Offset
= 0
3266 overlapped
.OffsetHigh
= 0
3267 overlapped
.hEvent
= 0
3268 f
._lock
_file
_overlapped
_p
= ctypes
.pointer(overlapped
)
3269 handle
= msvcrt
.get_osfhandle(f
.fileno())
3270 if not LockFileEx(handle
, 0x2 if exclusive
else 0x0, 0,
3271 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3272 raise OSError('Locking file failed: %r' % ctypes
.FormatError())
3274 def _unlock_file(f
):
3275 assert f
._lock
_file
_overlapped
_p
3276 handle
= msvcrt
.get_osfhandle(f
.fileno())
3277 if not UnlockFileEx(handle
, 0,
3278 whole_low
, whole_high
, f
._lock
_file
_overlapped
_p
):
3279 raise OSError('Unlocking file failed: %r' % ctypes
.FormatError())
3282 # Some platforms, such as Jython, is missing fcntl
3286 def _lock_file(f
, exclusive
):
3287 fcntl
.flock(f
, fcntl
.LOCK_EX
if exclusive
else fcntl
.LOCK_SH
)
3289 def _unlock_file(f
):
3290 fcntl
.flock(f
, fcntl
.LOCK_UN
)
3292 UNSUPPORTED_MSG
= 'file locking is not supported on this platform'
3294 def _lock_file(f
, exclusive
):
3295 raise IOError(UNSUPPORTED_MSG
)
3297 def _unlock_file(f
):
3298 raise IOError(UNSUPPORTED_MSG
)
3301 class locked_file(object):
3302 def __init__(self
, filename
, mode
, encoding
=None):
3303 assert mode
in ['r', 'a', 'w']
3304 self
.f
= io
.open(filename
, mode
, encoding
=encoding
)
3307 def __enter__(self
):
3308 exclusive
= self
.mode
!= 'r'
3310 _lock_file(self
.f
, exclusive
)
3316 def __exit__(self
, etype
, value
, traceback
):
3318 _unlock_file(self
.f
)
3325 def write(self
, *args
):
3326 return self
.f
.write(*args
)
3328 def read(self
, *args
):
3329 return self
.f
.read(*args
)
3332 def get_filesystem_encoding():
3333 encoding
= sys
.getfilesystemencoding()
3334 return encoding
if encoding
is not None else 'utf-8'
3337 def shell_quote(args
):
3339 encoding
= get_filesystem_encoding()
3341 if isinstance(a
, bytes):
3342 # We may get a filename encoded with 'encodeFilename'
3343 a
= a
.decode(encoding
)
3344 quoted_args
.append(compat_shlex_quote(a
))
3345 return ' '.join(quoted_args
)
3348 def smuggle_url(url
, data
):
3349 """ Pass additional data in a URL for internal use. """
3351 url
, idata
= unsmuggle_url(url
, {})
3353 sdata
= compat_urllib_parse_urlencode(
3354 {'__youtubedl_smuggle': json.dumps(data)}
)
3355 return url
+ '#' + sdata
3358 def unsmuggle_url(smug_url
, default
=None):
3359 if '#__youtubedl_smuggle' not in smug_url
:
3360 return smug_url
, default
3361 url
, _
, sdata
= smug_url
.rpartition('#')
3362 jsond
= compat_parse_qs(sdata
)['__youtubedl_smuggle'][0]
3363 data
= json
.loads(jsond
)
3367 def format_bytes(bytes):
3370 if type(bytes) is str:
3371 bytes = float(bytes)
3375 exponent
= int(math
.log(bytes, 1024.0))
3376 suffix
= ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent
]
3377 converted
= float(bytes) / float(1024 ** exponent
)
3378 return '%.2f%s' % (converted
, suffix
)
3381 def lookup_unit_table(unit_table
, s
):
3382 units_re
= '|'.join(re
.escape(u
) for u
in unit_table
)
3384 r
'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re
, s
)
3387 num_str
= m
.group('num').replace(',', '.')
3388 mult
= unit_table
[m
.group('unit')]
3389 return int(float(num_str
) * mult
)
3392 def parse_filesize(s
):
3396 # The lower-case forms are of course incorrect and unofficial,
3397 # but we support those too
3414 'megabytes': 1000 ** 2,
3415 'mebibytes': 1024 ** 2,
3421 'gigabytes': 1000 ** 3,
3422 'gibibytes': 1024 ** 3,
3428 'terabytes': 1000 ** 4,
3429 'tebibytes': 1024 ** 4,
3435 'petabytes': 1000 ** 5,
3436 'pebibytes': 1024 ** 5,
3442 'exabytes': 1000 ** 6,
3443 'exbibytes': 1024 ** 6,
3449 'zettabytes': 1000 ** 7,
3450 'zebibytes': 1024 ** 7,
3456 'yottabytes': 1000 ** 8,
3457 'yobibytes': 1024 ** 8,
3460 return lookup_unit_table(_UNIT_TABLE
, s
)
3469 if re
.match(r
'^[\d,.]+$', s
):
3470 return str_to_int(s
)
3481 return lookup_unit_table(_UNIT_TABLE
, s
)
3484 def parse_resolution(s
):
3488 mobj
= re
.search(r
'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s
)
3491 'width': int(mobj
.group('w')),
3492 'height': int(mobj
.group('h')),
3495 mobj
= re
.search(r
'\b(\d+)[pPiI]\b', s
)
3497 return {'height': int(mobj.group(1))}
3499 mobj
= re
.search(r
'\b([48])[kK]\b', s
)
3501 return {'height': int(mobj.group(1)) * 540}
3506 def parse_bitrate(s
):
3507 if not isinstance(s
, compat_str
):
3509 mobj
= re
.search(r
'\b(\d+)\s*kbps', s
)
3511 return int(mobj
.group(1))
3514 def month_by_name(name
, lang
='en'):
3515 """ Return the number of a month by (locale-independently) English name """
3517 month_names
= MONTH_NAMES
.get(lang
, MONTH_NAMES
['en'])
3520 return month_names
.index(name
) + 1
3525 def month_by_abbreviation(abbrev
):
3526 """ Return the number of a month by (locale-independently) English
3530 return [s
[:3] for s
in ENGLISH_MONTH_NAMES
].index(abbrev
) + 1
3535 def fix_xml_ampersands(xml_str
):
3536 """Replace all the '&' by '&' in XML"""
3538 r
'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3543 def setproctitle(title
):
3544 assert isinstance(title
, compat_str
)
3546 # ctypes in Jython is not complete
3547 # http://bugs.jython.org/issue2148
3548 if sys
.platform
.startswith('java'):
3552 libc
= ctypes
.cdll
.LoadLibrary('libc.so.6')
3556 # LoadLibrary in Windows Python 2.7.13 only expects
3557 # a bytestring, but since unicode_literals turns
3558 # every string into a unicode string, it fails.
3560 title_bytes
= title
.encode('utf-8')
3561 buf
= ctypes
.create_string_buffer(len(title_bytes
))
3562 buf
.value
= title_bytes
3564 libc
.prctl(15, buf
, 0, 0, 0)
3565 except AttributeError:
3566 return # Strange libc, just skip this
3569 def remove_start(s
, start
):
3570 return s
[len(start
):] if s
is not None and s
.startswith(start
) else s
3573 def remove_end(s
, end
):
3574 return s
[:-len(end
)] if s
is not None and s
.endswith(end
) else s
3577 def remove_quotes(s
):
3578 if s
is None or len(s
) < 2:
3580 for quote
in ('"', "'", ):
3581 if s
[0] == quote
and s
[-1] == quote
:
3586 def get_domain(url
):
3587 domain
= re
.match(r
'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url
)
3588 return domain
.group('domain') if domain
else None
3591 def url_basename(url
):
3592 path
= compat_urlparse
.urlparse(url
).path
3593 return path
.strip('/').split('/')[-1]
3597 return re
.match(r
'https?://[^?#&]+/', url
).group()
3600 def urljoin(base
, path
):
3601 if isinstance(path
, bytes):
3602 path
= path
.decode('utf-8')
3603 if not isinstance(path
, compat_str
) or not path
:
3605 if re
.match(r
'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path
):
3607 if isinstance(base
, bytes):
3608 base
= base
.decode('utf-8')
3609 if not isinstance(base
, compat_str
) or not re
.match(
3610 r
'^(?:https?:)?//', base
):
3612 return compat_urlparse
.urljoin(base
, path
)
3615 class HEADRequest(compat_urllib_request
.Request
):
3616 def get_method(self
):
3620 class PUTRequest(compat_urllib_request
.Request
):
3621 def get_method(self
):
3625 def int_or_none(v
, scale
=1, default
=None, get_attr
=None, invscale
=1):
3628 v
= getattr(v
, get_attr
, None)
3634 return int(v
) * invscale
// scale
3635 except (ValueError, TypeError):
3639 def str_or_none(v
, default
=None):
3640 return default
if v
is None else compat_str(v
)
3643 def str_to_int(int_str
):
3644 """ A more relaxed version of int_or_none """
3645 if isinstance(int_str
, compat_integer_types
):
3647 elif isinstance(int_str
, compat_str
):
3648 int_str
= re
.sub(r
'[,\.\+]', '', int_str
)
3649 return int_or_none(int_str
)
3652 def float_or_none(v
, scale
=1, invscale
=1, default
=None):
3656 return float(v
) * invscale
/ scale
3657 except (ValueError, TypeError):
3661 def bool_or_none(v
, default
=None):
3662 return v
if isinstance(v
, bool) else default
3665 def strip_or_none(v
, default
=None):
3666 return v
.strip() if isinstance(v
, compat_str
) else default
3669 def url_or_none(url
):
3670 if not url
or not isinstance(url
, compat_str
):
3673 return url
if re
.match(r
'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url
) else None
3676 def parse_duration(s
):
3677 if not isinstance(s
, compat_basestring
):
3682 days
, hours
, mins
, secs
, ms
= [None] * 5
3683 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
)
3685 days
, hours
, mins
, secs
, ms
= m
.groups()
3690 [0-9]+\s*y(?:ears?)?\s*
3693 [0-9]+\s*m(?:onths?)?\s*
3696 [0-9]+\s*w(?:eeks?)?\s*
3699 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3703 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3706 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3709 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3712 days
, hours
, mins
, secs
, ms
= m
.groups()
3714 m
= re
.match(r
'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s
)
3716 hours
, mins
= m
.groups()
3722 duration
+= float(secs
)
3724 duration
+= float(mins
) * 60
3726 duration
+= float(hours
) * 60 * 60
3728 duration
+= float(days
) * 24 * 60 * 60
3730 duration
+= float(ms
)
3734 def prepend_extension(filename
, ext
, expected_real_ext
=None):
3735 name
, real_ext
= os
.path
.splitext(filename
)
3737 '{0}.{1}{2}'.format(name
, ext
, real_ext
)
3738 if not expected_real_ext
or real_ext
[1:] == expected_real_ext
3739 else '{0}.{1}'.format(filename
, ext
))
3742 def replace_extension(filename
, ext
, expected_real_ext
=None):
3743 name
, real_ext
= os
.path
.splitext(filename
)
3744 return '{0}.{1}'.format(
3745 name
if not expected_real_ext
or real_ext
[1:] == expected_real_ext
else filename
,
3749 def check_executable(exe
, args
=[]):
3750 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3751 args can be a list of arguments for a short output (like -version) """
3753 process_communicate_or_kill(subprocess
.Popen(
3754 [exe
] + args
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
))
3760 def get_exe_version(exe
, args
=['--version'],
3761 version_re
=None, unrecognized
='present'):
3762 """ Returns the version of the specified executable,
3763 or False if the executable is not present """
3765 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3766 # SIGTTOU if youtube-dlc is run in the background.
3767 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3768 out
, _
= process_communicate_or_kill(subprocess
.Popen(
3769 [encodeArgument(exe
)] + args
,
3770 stdin
=subprocess
.PIPE
,
3771 stdout
=subprocess
.PIPE
, stderr
=subprocess
.STDOUT
))
3774 if isinstance(out
, bytes): # Python 2.x
3775 out
= out
.decode('ascii', 'ignore')
3776 return detect_exe_version(out
, version_re
, unrecognized
)
3779 def detect_exe_version(output
, version_re
=None, unrecognized
='present'):
3780 assert isinstance(output
, compat_str
)
3781 if version_re
is None:
3782 version_re
= r
'version\s+([-0-9._a-zA-Z]+)'
3783 m
= re
.search(version_re
, output
)
3790 class PagedList(object):
3792 # This is only useful for tests
3793 return len(self
.getslice())
3796 class OnDemandPagedList(PagedList
):
3797 def __init__(self
, pagefunc
, pagesize
, use_cache
=True):
3798 self
._pagefunc
= pagefunc
3799 self
._pagesize
= pagesize
3800 self
._use
_cache
= use_cache
3804 def getslice(self
, start
=0, end
=None):
3806 for pagenum
in itertools
.count(start
// self
._pagesize
):
3807 firstid
= pagenum
* self
._pagesize
3808 nextfirstid
= pagenum
* self
._pagesize
+ self
._pagesize
3809 if start
>= nextfirstid
:
3814 page_results
= self
._cache
.get(pagenum
)
3815 if page_results
is None:
3816 page_results
= list(self
._pagefunc
(pagenum
))
3818 self
._cache
[pagenum
] = page_results
3821 start
% self
._pagesize
3822 if firstid
<= start
< nextfirstid
3826 ((end
- 1) % self
._pagesize
) + 1
3827 if (end
is not None and firstid
<= end
<= nextfirstid
)
3830 if startv
!= 0 or endv
is not None:
3831 page_results
= page_results
[startv
:endv
]
3832 res
.extend(page_results
)
3834 # A little optimization - if current page is not "full", ie. does
3835 # not contain page_size videos then we can assume that this page
3836 # is the last one - there are no more ids on further pages -
3837 # i.e. no need to query again.
3838 if len(page_results
) + startv
< self
._pagesize
:
3841 # If we got the whole page, but the next page is not interesting,
3842 # break out early as well
3843 if end
== nextfirstid
:
3848 class InAdvancePagedList(PagedList
):
3849 def __init__(self
, pagefunc
, pagecount
, pagesize
):
3850 self
._pagefunc
= pagefunc
3851 self
._pagecount
= pagecount
3852 self
._pagesize
= pagesize
3854 def getslice(self
, start
=0, end
=None):
3856 start_page
= start
// self
._pagesize
3858 self
._pagecount
if end
is None else (end
// self
._pagesize
+ 1))
3859 skip_elems
= start
- start_page
* self
._pagesize
3860 only_more
= None if end
is None else end
- start
3861 for pagenum
in range(start_page
, end_page
):
3862 page
= list(self
._pagefunc
(pagenum
))
3864 page
= page
[skip_elems
:]
3866 if only_more
is not None:
3867 if len(page
) < only_more
:
3868 only_more
-= len(page
)
3870 page
= page
[:only_more
]
3877 def uppercase_escape(s
):
3878 unicode_escape
= codecs
.getdecoder('unicode_escape')
3880 r
'\\U[0-9a-fA-F]{8}',
3881 lambda m
: unicode_escape(m
.group(0))[0],
3885 def lowercase_escape(s
):
3886 unicode_escape
= codecs
.getdecoder('unicode_escape')
3888 r
'\\u[0-9a-fA-F]{4}',
3889 lambda m
: unicode_escape(m
.group(0))[0],
3893 def escape_rfc3986(s
):
3894 """Escape non-ASCII characters as suggested by RFC 3986"""
3895 if sys
.version_info
< (3, 0) and isinstance(s
, compat_str
):
3896 s
= s
.encode('utf-8')
3897 return compat_urllib_parse
.quote(s
, b
"%/;:@&=+$,!~*'()?#[]")
3900 def escape_url(url
):
3901 """Escape URL as suggested by RFC 3986"""
3902 url_parsed
= compat_urllib_parse_urlparse(url
)
3903 return url_parsed
._replace
(
3904 netloc
=url_parsed
.netloc
.encode('idna').decode('ascii'),
3905 path
=escape_rfc3986(url_parsed
.path
),
3906 params
=escape_rfc3986(url_parsed
.params
),
3907 query
=escape_rfc3986(url_parsed
.query
),
3908 fragment
=escape_rfc3986(url_parsed
.fragment
)
3912 def read_batch_urls(batch_fd
):
3914 if not isinstance(url
, compat_str
):
3915 url
= url
.decode('utf-8', 'replace')
3916 BOM_UTF8
= ('\xef\xbb\xbf', '\ufeff')
3917 for bom
in BOM_UTF8
:
3918 if url
.startswith(bom
):
3919 url
= url
[len(bom
):]
3921 if not url
or url
.startswith(('#', ';', ']')):
3923 # "#" cannot be stripped out since it is part of the URI
3924 # However, it can be safely stipped out if follwing a whitespace
3925 return re
.split(r
'\s#', url
, 1)[0].rstrip()
3927 with contextlib
.closing(batch_fd
) as fd
:
3928 return [url
for url
in map(fixup
, fd
) if url
]
3931 def urlencode_postdata(*args
, **kargs
):
3932 return compat_urllib_parse_urlencode(*args
, **kargs
).encode('ascii')
3935 def update_url_query(url
, query
):
3938 parsed_url
= compat_urlparse
.urlparse(url
)
3939 qs
= compat_parse_qs(parsed_url
.query
)
3941 return compat_urlparse
.urlunparse(parsed_url
._replace
(
3942 query
=compat_urllib_parse_urlencode(qs
, True)))
3945 def update_Request(req
, url
=None, data
=None, headers
={}, query={}
):
3946 req_headers
= req
.headers
.copy()
3947 req_headers
.update(headers
)
3948 req_data
= data
or req
.data
3949 req_url
= update_url_query(url
or req
.get_full_url(), query
)
3950 req_get_method
= req
.get_method()
3951 if req_get_method
== 'HEAD':
3952 req_type
= HEADRequest
3953 elif req_get_method
== 'PUT':
3954 req_type
= PUTRequest
3956 req_type
= compat_urllib_request
.Request
3958 req_url
, data
=req_data
, headers
=req_headers
,
3959 origin_req_host
=req
.origin_req_host
, unverifiable
=req
.unverifiable
)
3960 if hasattr(req
, 'timeout'):
3961 new_req
.timeout
= req
.timeout
3965 def _multipart_encode_impl(data
, boundary
):
3966 content_type
= 'multipart/form-data; boundary=%s' % boundary
3969 for k
, v
in data
.items():
3970 out
+= b
'--' + boundary
.encode('ascii') + b
'\r\n'
3971 if isinstance(k
, compat_str
):
3972 k
= k
.encode('utf-8')
3973 if isinstance(v
, compat_str
):
3974 v
= v
.encode('utf-8')
3975 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
3976 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
3977 content
= b
'Content-Disposition: form-data; name="' + k
+ b
'"\r\n\r\n' + v
+ b
'\r\n'
3978 if boundary
.encode('ascii') in content
:
3979 raise ValueError('Boundary overlaps with data')
3982 out
+= b
'--' + boundary
.encode('ascii') + b
'--\r\n'
3984 return out
, content_type
3987 def multipart_encode(data
, boundary
=None):
3989 Encode a dict to RFC 7578-compliant form-data
3992 A dict where keys and values can be either Unicode or bytes-like
3995 If specified a Unicode object, it's used as the boundary. Otherwise
3996 a random boundary is generated.
3998 Reference: https://tools.ietf.org/html/rfc7578
4000 has_specified_boundary
= boundary
is not None
4003 if boundary
is None:
4004 boundary
= '---------------' + str(random
.randrange(0x0fffffff, 0xffffffff))
4007 out
, content_type
= _multipart_encode_impl(data
, boundary
)
4010 if has_specified_boundary
:
4014 return out
, content_type
4017 def dict_get(d
, key_or_keys
, default
=None, skip_false_values
=True):
4018 if isinstance(key_or_keys
, (list, tuple)):
4019 for key
in key_or_keys
:
4020 if key
not in d
or d
[key
] is None or skip_false_values
and not d
[key
]:
4024 return d
.get(key_or_keys
, default
)
4027 def try_get(src
, getter
, expected_type
=None):
4028 if not isinstance(getter
, (list, tuple)):
4033 except (AttributeError, KeyError, TypeError, IndexError):
4036 if expected_type
is None or isinstance(v
, expected_type
):
4040 def merge_dicts(*dicts
):
4042 for a_dict
in dicts
:
4043 for k
, v
in a_dict
.items():
4047 or (isinstance(v
, compat_str
) and v
4048 and isinstance(merged
[k
], compat_str
)
4049 and not merged
[k
])):
4054 def encode_compat_str(string
, encoding
=preferredencoding(), errors
='strict'):
4055 return string
if isinstance(string
, compat_str
) else compat_str(string
, encoding
, errors
)
4067 TV_PARENTAL_GUIDELINES
= {
4077 def parse_age_limit(s
):
4079 return s
if 0 <= s
<= 21 else None
4080 if not isinstance(s
, compat_basestring
):
4082 m
= re
.match(r
'^(?P<age>\d{1,2})\+?$', s
)
4084 return int(m
.group('age'))
4086 return US_RATINGS
[s
]
4087 m
= re
.match(r
'^TV[_-]?(%s)$' % '|'.join(k
[3:] for k
in TV_PARENTAL_GUIDELINES
), s
)
4089 return TV_PARENTAL_GUIDELINES
['TV-' + m
.group(1)]
4093 def strip_jsonp(code
):
4096 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
4097 (?:\s*&&\s*(?P=func_name))?
4098 \s*\(\s*(?P<callback_data>.*)\);?
4099 \s*?(?://[^\n]*)*$''',
4100 r
'\g<callback_data>', code
)
4103 def js_to_json(code
, vars={}):
4104 # vars is a dict of var, val pairs to substitute
4105 COMMENT_RE
= r
'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
4106 SKIP_RE
= r
'\s*(?:{comment})?\s*'.format(comment
=COMMENT_RE
)
4108 (r
'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip
=SKIP_RE
), 16),
4109 (r
'(?s)^(0+[0-7]+){skip}:?$'.format(skip
=SKIP_RE
), 8),
4114 if v
in ('true', 'false', 'null'):
4116 elif v
.startswith('/*') or v
.startswith('//') or v
.startswith('!') or v
== ',':
4119 if v
[0] in ("'", '"'):
4120 v
= re
.sub(r
'(?s)\\.|"', lambda m
: {
4125 }.get(m
.group(0), m
.group(0)), v
[1:-1])
4127 for regex
, base
in INTEGER_TABLE
:
4128 im
= re
.match(regex
, v
)
4130 i
= int(im
.group(1), base
)
4131 return '"%d":' % i
if v
.endswith(':') else '%d' % i
4138 return re
.sub(r
'''(?sx)
4139 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4140 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4141 {comment}|,(?={skip}[\]}}])|
4142 (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
4143 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4146 '''.format(comment
=COMMENT_RE
, skip
=SKIP_RE
), fix_kv
, code
)
4149 def qualities(quality_ids
):
4150 """ Get a numeric quality value out of a list of possible values """
4153 return quality_ids
.index(qid
)
4159 DEFAULT_OUTTMPL
= '%(title)s [%(id)s].%(ext)s'
4162 def limit_length(s
, length
):
4163 """ Add ellipses to overly long strings """
4168 return s
[:length
- len(ELLIPSES
)] + ELLIPSES
4172 def version_tuple(v
):
4173 return tuple(int(e
) for e
in re
.split(r
'[-.]', v
))
4176 def is_outdated_version(version
, limit
, assume_new
=True):
4178 return not assume_new
4180 return version_tuple(version
) < version_tuple(limit
)
4182 return not assume_new
4185 def ytdl_is_updateable():
4186 """ Returns if youtube-dlc can be updated with -U """
4189 from zipimport
import zipimporter
4191 return isinstance(globals().get('__loader__'), zipimporter
) or hasattr(sys
, 'frozen')
4194 def args_to_str(args
):
4195 # Get a short string representation for a subprocess command
4196 return ' '.join(compat_shlex_quote(a
) for a
in args
)
4199 def error_to_compat_str(err
):
4201 # On python 2 error byte string must be decoded with proper
4202 # encoding rather than ascii
4203 if sys
.version_info
[0] < 3:
4204 err_str
= err_str
.decode(preferredencoding())
4208 def mimetype2ext(mt
):
4214 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4215 # it's the most popular one
4216 'audio/mpeg': 'mp3',
4217 'audio/x-wav': 'wav',
4222 _
, _
, res
= mt
.rpartition('/')
4223 res
= res
.split(';')[0].strip().lower()
4227 'smptett+xml': 'tt',
4231 'x-mp4-fragmented': 'mp4',
4232 'x-ms-sami': 'sami',
4235 'x-mpegurl': 'm3u8',
4236 'vnd.apple.mpegurl': 'm3u8',
4240 'vnd.ms-sstr+xml': 'ism',
4247 def parse_codecs(codecs_str
):
4248 # http://tools.ietf.org/html/rfc6381
4251 split_codecs
= list(filter(None, map(
4252 lambda str: str.strip(), codecs_str
.strip().strip(',').split(','))))
4253 vcodec
, acodec
= None, None
4254 for full_codec
in split_codecs
:
4255 codec
= full_codec
.split('.')[0]
4256 if codec
in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4259 elif codec
in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4263 write_string('WARNING: Unknown codec %s\n' % full_codec
, sys
.stderr
)
4264 if not vcodec
and not acodec
:
4265 if len(split_codecs
) == 2:
4267 'vcodec': split_codecs
[0],
4268 'acodec': split_codecs
[1],
4272 'vcodec': vcodec
or 'none',
4273 'acodec': acodec
or 'none',
4278 def urlhandle_detect_ext(url_handle
):
4279 getheader
= url_handle
.headers
.get
4281 cd
= getheader('Content-Disposition')
4283 m
= re
.match(r
'attachment;\s*filename="(?P<filename>[^"]+)"', cd
)
4285 e
= determine_ext(m
.group('filename'), default_ext
=None)
4289 return mimetype2ext(getheader('Content-Type'))
4292 def encode_data_uri(data
, mime_type
):
4293 return 'data:%s;base64,%s' % (mime_type
, base64
.b64encode(data
).decode('ascii'))
4296 def age_restricted(content_limit
, age_limit
):
4297 """ Returns True iff the content should be blocked """
4299 if age_limit
is None: # No limit set
4301 if content_limit
is None:
4302 return False # Content available for everyone
4303 return age_limit
< content_limit
4306 def is_html(first_bytes
):
4307 """ Detect whether a file contains HTML by examining its first bytes. """
4310 (b
'\xef\xbb\xbf', 'utf-8'),
4311 (b
'\x00\x00\xfe\xff', 'utf-32-be'),
4312 (b
'\xff\xfe\x00\x00', 'utf-32-le'),
4313 (b
'\xff\xfe', 'utf-16-le'),
4314 (b
'\xfe\xff', 'utf-16-be'),
4316 for bom
, enc
in BOMS
:
4317 if first_bytes
.startswith(bom
):
4318 s
= first_bytes
[len(bom
):].decode(enc
, 'replace')
4321 s
= first_bytes
.decode('utf-8', 'replace')
4323 return re
.match(r
'^\s*<', s
)
4326 def determine_protocol(info_dict
):
4327 protocol
= info_dict
.get('protocol')
4328 if protocol
is not None:
4331 url
= info_dict
['url']
4332 if url
.startswith('rtmp'):
4334 elif url
.startswith('mms'):
4336 elif url
.startswith('rtsp'):
4339 ext
= determine_ext(url
)
4345 return compat_urllib_parse_urlparse(url
).scheme
4348 def render_table(header_row
, data
, delim
=False, extraGap
=0, hideEmpty
=False):
4349 """ Render a list of rows, each as a list of values """
4351 def get_max_lens(table
):
4352 return [max(len(compat_str(v
)) for v
in col
) for col
in zip(*table
)]
4354 def filter_using_list(row
, filterArray
):
4355 return [col
for (take
, col
) in zip(filterArray
, row
) if take
]
4358 max_lens
= get_max_lens(data
)
4359 header_row
= filter_using_list(header_row
, max_lens
)
4360 data
= [filter_using_list(row
, max_lens
) for row
in data
]
4362 table
= [header_row
] + data
4363 max_lens
= get_max_lens(table
)
4365 table
= [header_row
] + [['-' * ml
for ml
in max_lens
]] + data
4366 format_str
= ' '.join('%-' + compat_str(ml
+ extraGap
) + 's' for ml
in max_lens
[:-1]) + ' %s'
4367 return '\n'.join(format_str
% tuple(row
) for row
in table
)
4370 def _match_one(filter_part
, dct
):
4371 COMPARISON_OPERATORS
= {
4379 operator_rex
= re
.compile(r
'''(?x)\s*
4381 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4383 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4384 (?P<quote>["\'])(?P
<quotedstrval
>(?
:\\.|
(?
!(?P
=quote
)|
\\).)+?
)(?P
=quote
)|
4385 (?P
<strval
>(?
![0-9.])[a
-z0
-9A
-Z
]*)
4388 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4389 m = operator_rex.search(filter_part)
4391 op = COMPARISON_OPERATORS[m.group('op')]
4392 actual_value = dct.get(m.group('key'))
4393 if (m.group('quotedstrval') is not None
4394 or m.group('strval') is not None
4395 # If the original field is a string and matching comparisonvalue is
4396 # a number we should respect the origin of the original field
4397 # and process comparison value as a string (see
4398 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4399 or actual_value is not None and m.group('intval') is not None
4400 and isinstance(actual_value, compat_str)):
4401 if m.group('op') not in ('=', '!='):
4403 'Operator %s does not support string values!' % m.group('op'))
4404 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4405 quote = m.group('quote')
4406 if quote is not None:
4407 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4410 comparison_value = int(m.group('intval'))
4412 comparison_value = parse_filesize(m.group('intval'))
4413 if comparison_value is None:
4414 comparison_value = parse_filesize(m.group('intval') + 'B')
4415 if comparison_value is None:
4417 'Invalid integer value %r in filter part %r' % (
4418 m.group('intval'), filter_part))
4419 if actual_value is None:
4420 return m.group('none_inclusive')
4421 return op(actual_value, comparison_value)
4424 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4425 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4427 operator_rex = re.compile(r'''(?x
)\s
*
4428 (?P
<op
>%s)\s
*(?P
<key
>[a
-z_
]+)
4430 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4431 m = operator_rex.search(filter_part)
4433 op = UNARY_OPERATORS[m.group('op')]
4434 actual_value = dct.get(m.group('key'))
4435 return op(actual_value)
4437 raise ValueError('Invalid filter part %r' % filter_part)
4440 def match_str(filter_str, dct):
4441 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
4444 _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
4447 def match_filter_func(filter_str):
4448 def _match_func(info_dict):
4449 if match_str(filter_str, info_dict):
4452 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4453 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4457 def parse_dfxp_time_expr(time_expr):
4461 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4463 return float(mobj.group('time_offset'))
4465 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4467 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4470 def srt_subtitles_timecode(seconds):
4471 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4474 def dfxp2srt(dfxp_data):
4476 @param dfxp_data A
bytes-like
object containing DFXP data
4477 @returns A
unicode object containing converted SRT data
4479 LEGACY_NAMESPACES = (
4480 (b'http://www.w3.org/ns/ttml', [
4481 b'http://www.w3.org/2004/11/ttaf1',
4482 b'http://www.w3.org/2006/04/ttaf1',
4483 b'http://www.w3.org/2006/10/ttaf1',
4485 (b'http://www.w3.org/ns/ttml#styling', [
4486 b'http://www.w3.org/ns/ttml#style',
4490 SUPPORTED_STYLING = [
4499 _x = functools.partial(xpath_with_ns, ns_map={
4500 'xml': 'http://www.w3.org/XML/1998/namespace',
4501 'ttml': 'http://www.w3.org/ns/ttml',
4502 'tts': 'http://www.w3.org/ns/ttml#styling',
4508 class TTMLPElementParser(object):
4510 _unclosed_elements = []
4511 _applied_styles = []
4513 def start(self, tag, attrib):
4514 if tag in (_x('ttml:br'), 'br'):
4517 unclosed_elements = []
4519 element_style_id = attrib.get('style')
4521 style.update(default_style)
4522 if element_style_id:
4523 style.update(styles.get(element_style_id, {}))
4524 for prop in SUPPORTED_STYLING:
4525 prop_val = attrib.get(_x('tts:' + prop))
4527 style[prop] = prop_val
4530 for k, v in sorted(style.items()):
4531 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4534 font += ' color="%s"' % v
4535 elif k == 'fontSize':
4536 font += ' size="%s"' % v
4537 elif k == 'fontFamily':
4538 font += ' face="%s"' % v
4539 elif k == 'fontWeight' and v == 'bold':
4541 unclosed_elements.append('b')
4542 elif k == 'fontStyle' and v == 'italic':
4544 unclosed_elements.append('i')
4545 elif k == 'textDecoration' and v == 'underline':
4547 unclosed_elements.append('u')
4549 self._out += '<font' + font + '>'
4550 unclosed_elements.append('font')
4552 if self._applied_styles:
4553 applied_style.update(self._applied_styles[-1])
4554 applied_style.update(style)
4555 self._applied_styles.append(applied_style)
4556 self._unclosed_elements.append(unclosed_elements)
4559 if tag not in (_x('ttml:br'), 'br'):
4560 unclosed_elements = self._unclosed_elements.pop()
4561 for element in reversed(unclosed_elements):
4562 self._out += '</%s>' % element
4563 if unclosed_elements and self._applied_styles:
4564 self._applied_styles.pop()
4566 def data(self, data):
4570 return self._out.strip()
4572 def parse_node(node):
4573 target = TTMLPElementParser()
4574 parser = xml.etree.ElementTree.XMLParser(target=target)
4575 parser.feed(xml.etree.ElementTree.tostring(node))
4576 return parser.close()
4578 for k, v in LEGACY_NAMESPACES:
4580 dfxp_data = dfxp_data.replace(ns, k)
4582 dfxp = compat_etree_fromstring(dfxp_data)
4584 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4587 raise ValueError('Invalid dfxp/TTML subtitle')
4591 for style in dfxp.findall(_x('.//ttml:style')):
4592 style_id = style.get('id') or style.get(_x('xml:id'))
4595 parent_style_id = style.get('style')
4597 if parent_style_id not in styles:
4600 styles[style_id] = styles[parent_style_id].copy()
4601 for prop in SUPPORTED_STYLING:
4602 prop_val = style.get(_x('tts:' + prop))
4604 styles.setdefault(style_id, {})[prop] = prop_val
4610 for p in ('body', 'div'):
4611 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4614 style = styles.get(ele.get('style'))
4617 default_style.update(style)
4619 for para, index in zip(paras, itertools.count(1)):
4620 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
4621 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
4622 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
4623 if begin_time is None:
4628 end_time = begin_time + dur
4629 out.append('%d\n%s --> %s\n%s\n\n' % (
4631 srt_subtitles_timecode(begin_time),
4632 srt_subtitles_timecode(end_time),
4638 def cli_option(params, command_option, param):
4639 param = params.get(param)
4641 param = compat_str(param)
4642 return [command_option, param] if param is not None else []
4645 def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
4646 param = params.get(param)
4649 assert isinstance(param, bool)
4651 return [command_option + separator + (true_value if param else false_value)]
4652 return [command_option, true_value if param else false_value]
4655 def cli_valueless_option(params, command_option, param, expected_value=True):
4656 param = params.get(param)
4657 return [command_option] if param == expected_value else []
4660 def cli_configuration_args(params, arg_name, key, default=[], exe=None): # returns arg, for_compat
4661 argdict = params.get(arg_name, {})
4662 if isinstance(argdict, (list, tuple)): # for backward compatibility
4663 return argdict, True
4666 return default, False
4667 assert isinstance(argdict, dict)
4669 assert isinstance(key, compat_str)
4672 args = exe_args = None
4674 assert isinstance(exe, compat_str)
4676 args = argdict.get('%s+%s' % (key, exe))
4678 exe_args = argdict.get(exe)
4681 args = argdict.get(key) if key != exe else None
4682 if args is None and exe_args is None:
4683 args = argdict.get('default', default)
4685 args, exe_args = args or [], exe_args or []
4686 assert isinstance(args, (list, tuple))
4687 assert isinstance(exe_args, (list, tuple))
4688 return args + exe_args, False
4691 class ISO639Utils(object):
4692 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
4751 'iw': 'heb', # Replaced by he in 1989 revision
4761 'in': 'ind', # Replaced by id in 1989 revision
4876 'ji': 'yid', # Replaced by yi in 1989 revision
4884 def short2long(cls, code):
4885 """Convert language code from ISO 639-1 to ISO 639-2/T"""
4886 return cls._lang_map.get(code[:2])
4889 def long2short(cls, code):
4890 """Convert language code from ISO 639-2/T to ISO 639-1"""
4891 for short_name, long_name in cls._lang_map.items():
4892 if long_name == code:
4896 class ISO3166Utils(object):
4897 # From http://data.okfn.org/data/core/country-list
4899 'AF': 'Afghanistan',
4900 'AX': 'Åland Islands',
4903 'AS': 'American Samoa',
4908 'AG': 'Antigua and Barbuda',
4925 'BO': 'Bolivia, Plurinational State of',
4926 'BQ': 'Bonaire, Sint Eustatius and Saba',
4927 'BA': 'Bosnia and Herzegovina',
4929 'BV': 'Bouvet Island',
4931 'IO': 'British Indian Ocean Territory',
4932 'BN': 'Brunei Darussalam',
4934 'BF': 'Burkina Faso',
4940 'KY': 'Cayman Islands',
4941 'CF': 'Central African Republic',
4945 'CX': 'Christmas Island',
4946 'CC': 'Cocos (Keeling) Islands',
4950 'CD': 'Congo, the Democratic Republic of the',
4951 'CK': 'Cook Islands',
4953 'CI': 'Côte d\'Ivoire',
4958 'CZ': 'Czech Republic',
4962 'DO': 'Dominican Republic',
4965 'SV': 'El Salvador',
4966 'GQ': 'Equatorial Guinea',
4970 'FK': 'Falkland Islands (Malvinas)',
4971 'FO': 'Faroe Islands',
4975 'GF': 'French Guiana',
4976 'PF': 'French Polynesia',
4977 'TF': 'French Southern Territories',
4992 'GW': 'Guinea-Bissau',
4995 'HM': 'Heard Island and McDonald Islands',
4996 'VA': 'Holy See (Vatican City State)',
5003 'IR': 'Iran, Islamic Republic of',
5006 'IM': 'Isle of Man',
5016 'KP': 'Korea, Democratic People\'s Republic of',
5017 'KR': 'Korea, Republic of',
5020 'LA': 'Lao People\'s Democratic Republic',
5026 'LI': 'Liechtenstein',
5030 'MK': 'Macedonia, the Former Yugoslav Republic of',
5037 'MH': 'Marshall Islands',
5043 'FM': 'Micronesia, Federated States of',
5044 'MD': 'Moldova, Republic of',
5055 'NL': 'Netherlands',
5056 'NC': 'New Caledonia',
5057 'NZ': 'New Zealand',
5062 'NF': 'Norfolk Island',
5063 'MP': 'Northern Mariana Islands',
5068 'PS': 'Palestine, State of',
5070 'PG': 'Papua New Guinea',
5073 'PH': 'Philippines',
5077 'PR': 'Puerto Rico',
5081 'RU': 'Russian Federation',
5083 'BL': 'Saint Barthélemy',
5084 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
5085 'KN': 'Saint Kitts and Nevis',
5086 'LC': 'Saint Lucia',
5087 'MF': 'Saint Martin (French part)',
5088 'PM': 'Saint Pierre and Miquelon',
5089 'VC': 'Saint Vincent and the Grenadines',
5092 'ST': 'Sao Tome and Principe',
5093 'SA': 'Saudi Arabia',
5097 'SL': 'Sierra Leone',
5099 'SX': 'Sint Maarten (Dutch part)',
5102 'SB': 'Solomon Islands',
5104 'ZA': 'South Africa',
5105 'GS': 'South Georgia and the South Sandwich Islands',
5106 'SS': 'South Sudan',
5111 'SJ': 'Svalbard and Jan Mayen',
5114 'CH': 'Switzerland',
5115 'SY': 'Syrian Arab Republic',
5116 'TW': 'Taiwan, Province of China',
5118 'TZ': 'Tanzania, United Republic of',
5120 'TL': 'Timor-Leste',
5124 'TT': 'Trinidad and Tobago',
5127 'TM': 'Turkmenistan',
5128 'TC': 'Turks and Caicos Islands',
5132 'AE': 'United Arab Emirates',
5133 'GB': 'United Kingdom',
5134 'US': 'United States',
5135 'UM': 'United States Minor Outlying Islands',
5139 'VE': 'Venezuela, Bolivarian Republic of',
5141 'VG': 'Virgin Islands, British',
5142 'VI': 'Virgin Islands, U.S.',
5143 'WF': 'Wallis and Futuna',
5144 'EH': 'Western Sahara',
5151 def short2full(cls, code):
5152 """Convert an ISO 3166-2 country code to the corresponding full name"""
5153 return cls._country_map.get(code.upper())
5156 class GeoUtils(object):
5157 # Major IPv4 address blocks per country
5159 'AD': '46.172.224.0/19',
5160 'AE': '94.200.0.0/13',
5161 'AF': '149.54.0.0/17',
5162 'AG': '209.59.64.0/18',
5163 'AI': '204.14.248.0/21',
5164 'AL': '46.99.0.0/16',
5165 'AM': '46.70.0.0/15',
5166 'AO': '105.168.0.0/13',
5167 'AP': '182.50.184.0/21',
5168 'AQ': '23.154.160.0/24',
5169 'AR': '181.0.0.0/12',
5170 'AS': '202.70.112.0/20',
5171 'AT': '77.116.0.0/14',
5172 'AU': '1.128.0.0/11',
5173 'AW': '181.41.0.0/18',
5174 'AX': '185.217.4.0/22',
5175 'AZ': '5.197.0.0/16',
5176 'BA': '31.176.128.0/17',
5177 'BB': '65.48.128.0/17',
5178 'BD': '114.130.0.0/16',
5180 'BF': '102.178.0.0/15',
5181 'BG': '95.42.0.0/15',
5182 'BH': '37.131.0.0/17',
5183 'BI': '154.117.192.0/18',
5184 'BJ': '137.255.0.0/16',
5185 'BL': '185.212.72.0/23',
5186 'BM': '196.12.64.0/18',
5187 'BN': '156.31.0.0/16',
5188 'BO': '161.56.0.0/16',
5189 'BQ': '161.0.80.0/20',
5190 'BR': '191.128.0.0/12',
5191 'BS': '24.51.64.0/18',
5192 'BT': '119.2.96.0/19',
5193 'BW': '168.167.0.0/16',
5194 'BY': '178.120.0.0/13',
5195 'BZ': '179.42.192.0/18',
5196 'CA': '99.224.0.0/11',
5197 'CD': '41.243.0.0/16',
5198 'CF': '197.242.176.0/21',
5199 'CG': '160.113.0.0/16',
5200 'CH': '85.0.0.0/13',
5201 'CI': '102.136.0.0/14',
5202 'CK': '202.65.32.0/19',
5203 'CL': '152.172.0.0/14',
5204 'CM': '102.244.0.0/14',
5205 'CN': '36.128.0.0/10',
5206 'CO': '181.240.0.0/12',
5207 'CR': '201.192.0.0/12',
5208 'CU': '152.206.0.0/15',
5209 'CV': '165.90.96.0/19',
5210 'CW': '190.88.128.0/17',
5211 'CY': '31.153.0.0/16',
5212 'CZ': '88.100.0.0/14',
5214 'DJ': '197.241.0.0/17',
5215 'DK': '87.48.0.0/12',
5216 'DM': '192.243.48.0/20',
5217 'DO': '152.166.0.0/15',
5218 'DZ': '41.96.0.0/12',
5219 'EC': '186.68.0.0/15',
5220 'EE': '90.190.0.0/15',
5221 'EG': '156.160.0.0/11',
5222 'ER': '196.200.96.0/20',
5223 'ES': '88.0.0.0/11',
5224 'ET': '196.188.0.0/14',
5225 'EU': '2.16.0.0/13',
5226 'FI': '91.152.0.0/13',
5227 'FJ': '144.120.0.0/16',
5228 'FK': '80.73.208.0/21',
5229 'FM': '119.252.112.0/20',
5230 'FO': '88.85.32.0/19',
5232 'GA': '41.158.0.0/15',
5234 'GD': '74.122.88.0/21',
5235 'GE': '31.146.0.0/16',
5236 'GF': '161.22.64.0/18',
5237 'GG': '62.68.160.0/19',
5238 'GH': '154.160.0.0/12',
5239 'GI': '95.164.0.0/16',
5240 'GL': '88.83.0.0/19',
5241 'GM': '160.182.0.0/15',
5242 'GN': '197.149.192.0/18',
5243 'GP': '104.250.0.0/19',
5244 'GQ': '105.235.224.0/20',
5245 'GR': '94.64.0.0/13',
5246 'GT': '168.234.0.0/16',
5247 'GU': '168.123.0.0/16',
5248 'GW': '197.214.80.0/20',
5249 'GY': '181.41.64.0/18',
5250 'HK': '113.252.0.0/14',
5251 'HN': '181.210.0.0/16',
5252 'HR': '93.136.0.0/13',
5253 'HT': '148.102.128.0/17',
5254 'HU': '84.0.0.0/14',
5255 'ID': '39.192.0.0/10',
5256 'IE': '87.32.0.0/12',
5257 'IL': '79.176.0.0/13',
5258 'IM': '5.62.80.0/20',
5259 'IN': '117.192.0.0/10',
5260 'IO': '203.83.48.0/21',
5261 'IQ': '37.236.0.0/14',
5262 'IR': '2.176.0.0/12',
5263 'IS': '82.221.0.0/16',
5264 'IT': '79.0.0.0/10',
5265 'JE': '87.244.64.0/18',
5266 'JM': '72.27.0.0/17',
5267 'JO': '176.29.0.0/16',
5268 'JP': '133.0.0.0/8',
5269 'KE': '105.48.0.0/12',
5270 'KG': '158.181.128.0/17',
5271 'KH': '36.37.128.0/17',
5272 'KI': '103.25.140.0/22',
5273 'KM': '197.255.224.0/20',
5274 'KN': '198.167.192.0/19',
5275 'KP': '175.45.176.0/22',
5276 'KR': '175.192.0.0/10',
5277 'KW': '37.36.0.0/14',
5278 'KY': '64.96.0.0/15',
5279 'KZ': '2.72.0.0/13',
5280 'LA': '115.84.64.0/18',
5281 'LB': '178.135.0.0/16',
5282 'LC': '24.92.144.0/20',
5283 'LI': '82.117.0.0/19',
5284 'LK': '112.134.0.0/15',
5285 'LR': '102.183.0.0/16',
5286 'LS': '129.232.0.0/17',
5287 'LT': '78.56.0.0/13',
5288 'LU': '188.42.0.0/16',
5289 'LV': '46.109.0.0/16',
5290 'LY': '41.252.0.0/14',
5291 'MA': '105.128.0.0/11',
5292 'MC': '88.209.64.0/18',
5293 'MD': '37.246.0.0/16',
5294 'ME': '178.175.0.0/17',
5295 'MF': '74.112.232.0/21',
5296 'MG': '154.126.0.0/17',
5297 'MH': '117.103.88.0/21',
5298 'MK': '77.28.0.0/15',
5299 'ML': '154.118.128.0/18',
5300 'MM': '37.111.0.0/17',
5301 'MN': '49.0.128.0/17',
5302 'MO': '60.246.0.0/16',
5303 'MP': '202.88.64.0/20',
5304 'MQ': '109.203.224.0/19',
5305 'MR': '41.188.64.0/18',
5306 'MS': '208.90.112.0/22',
5307 'MT': '46.11.0.0/16',
5308 'MU': '105.16.0.0/12',
5309 'MV': '27.114.128.0/18',
5310 'MW': '102.70.0.0/15',
5311 'MX': '187.192.0.0/11',
5312 'MY': '175.136.0.0/13',
5313 'MZ': '197.218.0.0/15',
5314 'NA': '41.182.0.0/16',
5315 'NC': '101.101.0.0/18',
5316 'NE': '197.214.0.0/18',
5317 'NF': '203.17.240.0/22',
5318 'NG': '105.112.0.0/12',
5319 'NI': '186.76.0.0/15',
5320 'NL': '145.96.0.0/11',
5321 'NO': '84.208.0.0/13',
5322 'NP': '36.252.0.0/15',
5323 'NR': '203.98.224.0/19',
5324 'NU': '49.156.48.0/22',
5325 'NZ': '49.224.0.0/14',
5326 'OM': '5.36.0.0/15',
5327 'PA': '186.72.0.0/15',
5328 'PE': '186.160.0.0/14',
5329 'PF': '123.50.64.0/18',
5330 'PG': '124.240.192.0/19',
5331 'PH': '49.144.0.0/13',
5332 'PK': '39.32.0.0/11',
5333 'PL': '83.0.0.0/11',
5334 'PM': '70.36.0.0/20',
5335 'PR': '66.50.0.0/16',
5336 'PS': '188.161.0.0/16',
5337 'PT': '85.240.0.0/13',
5338 'PW': '202.124.224.0/20',
5339 'PY': '181.120.0.0/14',
5340 'QA': '37.210.0.0/15',
5341 'RE': '102.35.0.0/16',
5342 'RO': '79.112.0.0/13',
5343 'RS': '93.86.0.0/15',
5344 'RU': '5.136.0.0/13',
5345 'RW': '41.186.0.0/16',
5346 'SA': '188.48.0.0/13',
5347 'SB': '202.1.160.0/19',
5348 'SC': '154.192.0.0/11',
5349 'SD': '102.120.0.0/13',
5350 'SE': '78.64.0.0/12',
5351 'SG': '8.128.0.0/10',
5352 'SI': '188.196.0.0/14',
5353 'SK': '78.98.0.0/15',
5354 'SL': '102.143.0.0/17',
5355 'SM': '89.186.32.0/19',
5356 'SN': '41.82.0.0/15',
5357 'SO': '154.115.192.0/18',
5358 'SR': '186.179.128.0/17',
5359 'SS': '105.235.208.0/21',
5360 'ST': '197.159.160.0/19',
5361 'SV': '168.243.0.0/16',
5362 'SX': '190.102.0.0/20',
5364 'SZ': '41.84.224.0/19',
5365 'TC': '65.255.48.0/20',
5366 'TD': '154.68.128.0/19',
5367 'TG': '196.168.0.0/14',
5368 'TH': '171.96.0.0/13',
5369 'TJ': '85.9.128.0/18',
5370 'TK': '27.96.24.0/21',
5371 'TL': '180.189.160.0/20',
5372 'TM': '95.85.96.0/19',
5373 'TN': '197.0.0.0/11',
5374 'TO': '175.176.144.0/21',
5375 'TR': '78.160.0.0/11',
5376 'TT': '186.44.0.0/15',
5377 'TV': '202.2.96.0/19',
5378 'TW': '120.96.0.0/11',
5379 'TZ': '156.156.0.0/14',
5380 'UA': '37.52.0.0/14',
5381 'UG': '102.80.0.0/13',
5383 'UY': '167.56.0.0/13',
5384 'UZ': '84.54.64.0/18',
5385 'VA': '212.77.0.0/19',
5386 'VC': '207.191.240.0/21',
5387 'VE': '186.88.0.0/13',
5388 'VG': '66.81.192.0/20',
5389 'VI': '146.226.0.0/16',
5390 'VN': '14.160.0.0/11',
5391 'VU': '202.80.32.0/20',
5392 'WF': '117.20.32.0/21',
5393 'WS': '202.4.32.0/19',
5394 'YE': '134.35.0.0/16',
5395 'YT': '41.242.116.0/22',
5396 'ZA': '41.0.0.0/11',
5397 'ZM': '102.144.0.0/13',
5398 'ZW': '102.177.192.0/18',
5402 def random_ipv4(cls, code_or_block):
5403 if len(code_or_block) == 2:
5404 block = cls._country_ip_map.get(code_or_block.upper())
5408 block = code_or_block
5409 addr, preflen = block.split('/')
5410 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5411 addr_max = addr_min | (0xffffffff >> int(preflen))
5412 return compat_str(socket.inet_ntoa(
5413 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5416 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5417 def __init__(self, proxies=None):
5418 # Set default handlers
5419 for type in ('http', 'https'):
5420 setattr(self, '%s_open' % type,
5421 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5422 meth(r, proxy, type))
5423 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5425 def proxy_open(self, req, proxy, type):
5426 req_proxy = req.headers.get('Ytdl-request-proxy')
5427 if req_proxy is not None:
5429 del req.headers['Ytdl-request-proxy']
5431 if proxy == '__noproxy__':
5432 return None # No Proxy
5433 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5434 req.add_header('Ytdl-socks-proxy', proxy)
5435 # youtube-dlc's http/https handlers do wrapping the socket with socks
5437 return compat_urllib_request.ProxyHandler.proxy_open(
5438 self, req, proxy, type)
5441 # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5442 # released into Public Domain
5443 # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5445 def long_to_bytes(n, blocksize=0):
5446 """long_to_bytes(n:long, blocksize:int) : string
5447 Convert a long integer to a byte string.
5449 If optional blocksize is given and greater than zero, pad the front of the
5450 byte string with binary zeros so that the length is a multiple of
5453 # after much testing, this algorithm was deemed to be the fastest
5457 s = compat_struct_pack('>I', n & 0xffffffff) + s
5459 # strip off leading zeros
5460 for i in range(len(s)):
5461 if s[i] != b'\000'[0]:
5464 # only happens when n == 0
5468 # add back some pad bytes. this could be done more efficiently w.r.t. the
5469 # de-padding being done above, but sigh...
5470 if blocksize > 0 and len(s) % blocksize:
5471 s = (blocksize - len(s) % blocksize) * b'\000' + s
5475 def bytes_to_long(s):
5476 """bytes_to_long(string) : long
5477 Convert a byte string to a long integer.
5479 This is (essentially) the inverse of long_to_bytes().
5484 extra = (4 - length % 4)
5485 s = b'\000' * extra + s
5486 length = length + extra
5487 for i in range(0, length, 4):
5488 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5492 def ohdave_rsa_encrypt(data, exponent, modulus):
5494 Implement OHDave
's RSA algorithm. See http://www.ohdave.com/rsa/
5497 data: data to encrypt, bytes-like object
5498 exponent, modulus: parameter e and N of RSA algorithm, both integer
5499 Output: hex string of encrypted data
5501 Limitation: supports one block encryption only
5504 payload = int(binascii.hexlify(data[::-1]), 16)
5505 encrypted = pow(payload, exponent, modulus)
5506 return '%x' % encrypted
5509 def pkcs1pad(data, length):
5511 Padding input data with PKCS#1 scheme
5513 @param {int[]} data input data
5514 @param {int} length target length
5515 @returns {int[]} padded data
5517 if len(data) > length - 11:
5518 raise ValueError('Input data too
long for PKCS
#1 padding')
5520 pseudo_random
= [random
.randint(0, 254) for _
in range(length
- len(data
) - 3)]
5521 return [0, 2] + pseudo_random
+ [0] + data
5524 def encode_base_n(num
, n
, table
=None):
5525 FULL_TABLE
= '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5527 table
= FULL_TABLE
[:n
]
5530 raise ValueError('base %d exceeds table length %d' % (n
, len(table
)))
5537 ret
= table
[num
% n
] + ret
5542 def decode_packed_codes(code
):
5543 mobj
= re
.search(PACKED_CODES_RE
, code
)
5544 obfuscated_code
, base
, count
, symbols
= mobj
.groups()
5547 symbols
= symbols
.split('|')
5552 base_n_count
= encode_base_n(count
, base
)
5553 symbol_table
[base_n_count
] = symbols
[count
] or base_n_count
5556 r
'\b(\w+)\b', lambda mobj
: symbol_table
[mobj
.group(0)],
5560 def caesar(s
, alphabet
, shift
):
5565 alphabet
[(alphabet
.index(c
) + shift
) % l
] if c
in alphabet
else c
5570 return caesar(s
, r
'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5573 def parse_m3u8_attributes(attrib
):
5575 for (key
, val
) in re
.findall(r
'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib
):
5576 if val
.startswith('"'):
5582 def urshift(val
, n
):
5583 return val
>> n
if val
>= 0 else (val
+ 0x100000000) >> n
5586 # Based on png2str() written by @gdkchan and improved by @yokrysty
5587 # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5588 def decode_png(png_data
):
5589 # Reference: https://www.w3.org/TR/PNG/
5590 header
= png_data
[8:]
5592 if png_data
[:8] != b
'\x89PNG\x0d\x0a\x1a\x0a' or header
[4:8] != b
'IHDR':
5593 raise IOError('Not a valid PNG file.')
5595 int_map
= {1: '>B', 2: '>H', 4: '>I'}
5596 unpack_integer
= lambda x
: compat_struct_unpack(int_map
[len(x
)], x
)[0]
5601 length
= unpack_integer(header
[:4])
5604 chunk_type
= header
[:4]
5607 chunk_data
= header
[:length
]
5608 header
= header
[length
:]
5610 header
= header
[4:] # Skip CRC
5618 ihdr
= chunks
[0]['data']
5620 width
= unpack_integer(ihdr
[:4])
5621 height
= unpack_integer(ihdr
[4:8])
5625 for chunk
in chunks
:
5626 if chunk
['type'] == b
'IDAT':
5627 idat
+= chunk
['data']
5630 raise IOError('Unable to read PNG data.')
5632 decompressed_data
= bytearray(zlib
.decompress(idat
))
5637 def _get_pixel(idx
):
5642 for y
in range(height
):
5643 basePos
= y
* (1 + stride
)
5644 filter_type
= decompressed_data
[basePos
]
5648 pixels
.append(current_row
)
5650 for x
in range(stride
):
5651 color
= decompressed_data
[1 + basePos
+ x
]
5652 basex
= y
* stride
+ x
5657 left
= _get_pixel(basex
- 3)
5659 up
= _get_pixel(basex
- stride
)
5661 if filter_type
== 1: # Sub
5662 color
= (color
+ left
) & 0xff
5663 elif filter_type
== 2: # Up
5664 color
= (color
+ up
) & 0xff
5665 elif filter_type
== 3: # Average
5666 color
= (color
+ ((left
+ up
) >> 1)) & 0xff
5667 elif filter_type
== 4: # Paeth
5673 c
= _get_pixel(basex
- stride
- 3)
5681 if pa
<= pb
and pa
<= pc
:
5682 color
= (color
+ a
) & 0xff
5684 color
= (color
+ b
) & 0xff
5686 color
= (color
+ c
) & 0xff
5688 current_row
.append(color
)
5690 return width
, height
, pixels
5693 def write_xattr(path
, key
, value
):
5694 # This mess below finds the best xattr tool for the job
5696 # try the pyxattr module...
5699 if hasattr(xattr
, 'set'): # pyxattr
5700 # Unicode arguments are not supported in python-pyxattr until
5702 # See https://github.com/ytdl-org/youtube-dl/issues/5498
5703 pyxattr_required_version
= '0.5.0'
5704 if version_tuple(xattr
.__version
__) < version_tuple(pyxattr_required_version
):
5705 # TODO: fallback to CLI tools
5706 raise XAttrUnavailableError(
5707 'python-pyxattr is detected but is too old. '
5708 'youtube-dlc requires %s or above while your version is %s. '
5709 'Falling back to other xattr implementations' % (
5710 pyxattr_required_version
, xattr
.__version
__))
5712 setxattr
= xattr
.set
5714 setxattr
= xattr
.setxattr
5717 setxattr(path
, key
, value
)
5718 except EnvironmentError as e
:
5719 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5722 if compat_os_name
== 'nt':
5723 # Write xattrs to NTFS Alternate Data Streams:
5724 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
5725 assert ':' not in key
5726 assert os
.path
.exists(path
)
5728 ads_fn
= path
+ ':' + key
5730 with open(ads_fn
, 'wb') as f
:
5732 except EnvironmentError as e
:
5733 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5735 user_has_setfattr
= check_executable('setfattr', ['--version'])
5736 user_has_xattr
= check_executable('xattr', ['-h'])
5738 if user_has_setfattr
or user_has_xattr
:
5740 value
= value
.decode('utf-8')
5741 if user_has_setfattr
:
5742 executable
= 'setfattr'
5743 opts
= ['-n', key
, '-v', value
]
5744 elif user_has_xattr
:
5745 executable
= 'xattr'
5746 opts
= ['-w', key
, value
]
5748 cmd
= ([encodeFilename(executable
, True)]
5749 + [encodeArgument(o
) for o
in opts
]
5750 + [encodeFilename(path
, True)])
5753 p
= subprocess
.Popen(
5754 cmd
, stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
, stdin
=subprocess
.PIPE
)
5755 except EnvironmentError as e
:
5756 raise XAttrMetadataError(e
.errno
, e
.strerror
)
5757 stdout
, stderr
= process_communicate_or_kill(p
)
5758 stderr
= stderr
.decode('utf-8', 'replace')
5759 if p
.returncode
!= 0:
5760 raise XAttrMetadataError(p
.returncode
, stderr
)
5763 # On Unix, and can't find pyxattr, setfattr, or xattr.
5764 if sys
.platform
.startswith('linux'):
5765 raise XAttrUnavailableError(
5766 "Couldn't find a tool to set the xattrs. "
5767 "Install either the python 'pyxattr' or 'xattr' "
5768 "modules, or the GNU 'attr' package "
5769 "(which contains the 'setfattr' tool).")
5771 raise XAttrUnavailableError(
5772 "Couldn't find a tool to set the xattrs. "
5773 "Install either the python 'xattr' module, "
5774 "or the 'xattr' binary.")
5777 def random_birthday(year_field
, month_field
, day_field
):
5778 start_date
= datetime
.date(1950, 1, 1)
5779 end_date
= datetime
.date(1995, 12, 31)
5780 offset
= random
.randint(0, (end_date
- start_date
).days
)
5781 random_date
= start_date
+ datetime
.timedelta(offset
)
5783 year_field
: str(random_date
.year
),
5784 month_field
: str(random_date
.month
),
5785 day_field
: str(random_date
.day
),
5789 # Templates for internet shortcut files, which are plain text files.
5790 DOT_URL_LINK_TEMPLATE
= '''
5795 DOT_WEBLOC_LINK_TEMPLATE
= '''
5796 <?xml version="1.0" encoding="UTF-8"?>
5797 <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
5798 <plist version="1.0">
5801 \t<string>%(url)s</string>
5806 DOT_DESKTOP_LINK_TEMPLATE
= '''
5816 def iri_to_uri(iri
):
5818 Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
5820 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.
5823 iri_parts
= compat_urllib_parse_urlparse(iri
)
5825 if '[' in iri_parts
.netloc
:
5826 raise ValueError('IPv6 URIs are not, yet, supported.')
5827 # Querying `.netloc`, when there's only one bracket, also raises a ValueError.
5829 # 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.
5832 if iri_parts
.username
:
5833 net_location
+= compat_urllib_parse_quote(iri_parts
.username
, safe
=r
"!$%&'()*+,~")
5834 if iri_parts
.password
is not None:
5835 net_location
+= ':' + compat_urllib_parse_quote(iri_parts
.password
, safe
=r
"!$%&'()*+,~")
5838 net_location
+= iri_parts
.hostname
.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
5839 # The 'idna' encoding produces ASCII text.
5840 if iri_parts
.port
is not None and iri_parts
.port
!= 80:
5841 net_location
+= ':' + str(iri_parts
.port
)
5843 return compat_urllib_parse_urlunparse(
5847 compat_urllib_parse_quote_plus(iri_parts
.path
, safe
=r
"!$%&'()*+,/:;=@|~"),
5849 # Unsure about the `safe` argument, since this is a legacy way of handling parameters.
5850 compat_urllib_parse_quote_plus(iri_parts
.params
, safe
=r
"!$%&'()*+,/:;=@|~"),
5852 # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
5853 compat_urllib_parse_quote_plus(iri_parts
.query
, safe
=r
"!$%&'()*+,/:;=?@{|}~"),
5855 compat_urllib_parse_quote_plus(iri_parts
.fragment
, safe
=r
"!#$%&'()*+,/:;=?@{|}~")))
5857 # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
5860 def to_high_limit_path(path
):
5861 if sys
.platform
in ['win32', 'cygwin']:
5862 # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
5863 return r
'\\?\ '.rstrip() + os
.path
.abspath(path
)
5868 def format_field(obj
, field
, template
='%s', ignore
=(None, ''), default
='', func
=None):
5869 val
= obj
.get(field
, default
)
5870 if func
and val
not in ignore
:
5872 return template
% val
if val
not in ignore
else default
5875 def clean_podcast_url(url
):
5876 return re
.sub(r
'''(?x)
5880 media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
5883 (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
5886 cn\.co| # https://podcorn.com/analytics-prefix/
5887 st\.fm # https://podsights.com/docs/
5892 _HEX_TABLE
= '0123456789abcdef'
5895 def random_uuidv4():
5896 return re
.sub(r
'[xy]', lambda x
: _HEX_TABLE
[random
.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
5899 def make_dir(path
, to_screen
=None):
5901 dn
= os
.path
.dirname(path
)
5902 if dn
and not os
.path
.exists(dn
):
5905 except (OSError, IOError) as err
:
5906 if callable(to_screen
) is not None:
5907 to_screen('unable to create directory ' + error_to_compat_str(err
))
5911 def get_executable_path():
5912 path
= os
.path
.dirname(sys
.argv
[0])
5913 if os
.path
.abspath(sys
.argv
[0]) != os
.path
.abspath(sys
.executable
): # Not packaged
5914 path
= os
.path
.join(path
, '..')
5915 return os
.path
.abspath(path
)
5918 def load_plugins(name
, type, namespace
):
5919 plugin_info
= [None]
5922 plugin_info
= imp
.find_module(
5923 name
, [os
.path
.join(get_executable_path(), 'ytdlp_plugins')])
5924 plugins
= imp
.load_module(name
, *plugin_info
)
5925 for name
in dir(plugins
):
5926 if not name
.endswith(type):
5928 klass
= getattr(plugins
, name
)
5929 classes
.append(klass
)
5930 namespace
[name
] = klass
5934 if plugin_info
[0] is not None:
5935 plugin_info
[0].close()