X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/47d80ec0b18245caeb97018d4c1af18d0b5b972b..0f0afa2a182178f8c70d50daf99f864372fe71f3:/youtube_dl/utils.py
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 9595bcf..ead9bd8 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -11,6 +11,7 @@ import contextlib
import ctypes
import datetime
import email.utils
+import email.header
import errno
import functools
import gzip
@@ -21,8 +22,8 @@ import locale
import math
import operator
import os
-import pipes
import platform
+import random
import re
import socket
import ssl
@@ -34,10 +35,14 @@ import xml.etree.ElementTree
import zlib
from .compat import (
+ compat_HTMLParseError,
compat_HTMLParser,
compat_basestring,
compat_chr,
+ compat_cookiejar,
+ compat_ctypes_WINFUNCTYPE,
compat_etree_fromstring,
+ compat_expanduser,
compat_html_entities,
compat_html_entities_html5,
compat_http_client,
@@ -45,7 +50,6 @@ from .compat import (
compat_os_name,
compat_parse_qs,
compat_shlex_quote,
- compat_socket_create_connection,
compat_str,
compat_struct_pack,
compat_struct_unpack,
@@ -78,7 +82,7 @@ def register_socks_protocols():
compiled_regex_type = type(re.compile(''))
std_headers = {
- 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/47.0 (Chrome)',
+ 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Encoding': 'gzip, deflate',
@@ -86,6 +90,11 @@ std_headers = {
}
+USER_AGENTS = {
+ 'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27',
+}
+
+
NO_DEFAULT = object()
ENGLISH_MONTH_NAMES = [
@@ -116,14 +125,20 @@ KNOWN_EXTENSIONS = (
# needed for sanitizing filenames in restricted mode
ACCENT_CHARS = dict(zip('ÃÃÃÃÃÃ
ÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÅÃÅÃÃÃÃÅ°ÃÃÃà áâãäåæçèéêëìÃîïðñòóôõöÅøÅùúûüűýþÿ',
- itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'],
- 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy')))
+ itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
+ 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
DATE_FORMATS = (
'%d %B %Y',
'%d %b %Y',
'%B %d %Y',
+ '%B %dst %Y',
+ '%B %dnd %Y',
+ '%B %dth %Y',
'%b %d %Y',
+ '%b %dst %Y',
+ '%b %dnd %Y',
+ '%b %dth %Y',
'%b %dst %Y %I:%M',
'%b %dnd %Y %I:%M',
'%b %dth %Y %I:%M',
@@ -132,6 +147,7 @@ DATE_FORMATS = (
'%Y/%m/%d',
'%Y/%m/%d %H:%M',
'%Y/%m/%d %H:%M:%S',
+ '%Y-%m-%d %H:%M',
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
'%d.%m.%Y %H:%M',
@@ -144,6 +160,8 @@ DATE_FORMATS = (
'%Y-%m-%dT%H:%M',
'%b %d %Y at %H:%M',
'%b %d %Y at %H:%M:%S',
+ '%B %d %Y at %H:%M',
+ '%B %d %Y at %H:%M:%S',
)
DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
@@ -166,6 +184,7 @@ DATE_FORMATS_MONTH_FIRST.extend([
])
PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
+JSON_LD_RE = r'(?is)'
def preferredencoding():
@@ -325,34 +344,46 @@ def get_element_by_id(id, html):
def get_element_by_class(class_name, html):
- return get_element_by_attribute(
+ """Return the content of the first tag with the specified class in the passed HTML document"""
+ retval = get_elements_by_class(class_name, html)
+ return retval[0] if retval else None
+
+
+def get_element_by_attribute(attribute, value, html, escape_value=True):
+ retval = get_elements_by_attribute(attribute, value, html, escape_value)
+ return retval[0] if retval else None
+
+
+def get_elements_by_class(class_name, html):
+ """Return the content of all tags with the specified class in the passed HTML document as a list"""
+ return get_elements_by_attribute(
'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
html, escape_value=False)
-def get_element_by_attribute(attribute, value, html, escape_value=True):
+def get_elements_by_attribute(attribute, value, html, escape_value=True):
"""Return the content of the tag with the specified attribute in the passed HTML document"""
value = re.escape(value) if escape_value else value
- m = re.search(r'''(?xs)
+ retlist = []
+ for m in re.finditer(r'''(?xs)
<([a-zA-Z0-9:._-]+)
- (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+ (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s+%s=['"]?%s['"]?
- (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+ (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s*>
(?P.*?)
\1>
- ''' % (re.escape(attribute), value), html)
+ ''' % (re.escape(attribute), value), html):
+ res = m.group('content')
- if not m:
- return None
- res = m.group('content')
+ if res.startswith('"') or res.startswith("'"):
+ res = res[1:-1]
- if res.startswith('"') or res.startswith("'"):
- res = res[1:-1]
+ retlist.append(unescapeHTML(res))
- return unescapeHTML(res)
+ return retlist
class HTMLAttributeParser(compat_HTMLParser):
@@ -382,8 +413,12 @@ def extract_attributes(html_element):
but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
"""
parser = HTMLAttributeParser()
- parser.feed(html_element)
- parser.close()
+ try:
+ parser.feed(html_element)
+ parser.close()
+ # Older Python may throw HTMLParseError in case of malformed HTML
+ except compat_HTMLParseError:
+ pass
return parser.attrs
@@ -395,8 +430,8 @@ def clean_html(html):
# Newline vs
html = html.replace('\n', ' ')
- html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html)
- html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
+ html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
+ html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
# Strip html tags
html = re.sub('<.*?>', '', html)
# Replace html entities
@@ -448,7 +483,8 @@ def timeconvert(timestr):
def sanitize_filename(s, restricted=False, is_id=False):
"""Sanitizes a string so it could be used as part of a filename.
If restricted is set, use a stricter subset of allowed characters.
- Set is_id if this is not an arbitrary string, but an ID that should be kept if possible
+ Set is_id if this is not an arbitrary string, but an ID that should be kept
+ if possible.
"""
def replace_insane(char):
if restricted and char in ACCENT_CHARS:
@@ -496,23 +532,40 @@ def sanitize_path(s):
if drive_or_unc:
norm_path.pop(0)
sanitized_path = [
- path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|[\s.]$)', '#', path_part)
+ path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
for path_part in norm_path]
if drive_or_unc:
sanitized_path.insert(0, drive_or_unc + os.path.sep)
return os.path.join(*sanitized_path)
-# Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of
-# unwanted failures due to missing protocol
def sanitize_url(url):
- return 'http:%s' % url if url.startswith('//') else url
+ # Prepend protocol-less URLs with `http:` scheme in order to mitigate
+ # the number of unwanted failures due to missing protocol
+ if url.startswith('//'):
+ return 'http:%s' % url
+ # Fix some common typos seen so far
+ COMMON_TYPOS = (
+ # https://github.com/ytdl-org/youtube-dl/issues/15649
+ (r'^httpss://', r'https://'),
+ # https://bx1.be/lives/direct-tv/
+ (r'^rmtp([es]?)://', r'rtmp\1://'),
+ )
+ for mistake, fixup in COMMON_TYPOS:
+ if re.match(mistake, url):
+ return re.sub(mistake, fixup, url)
+ return url
def sanitized_Request(url, *args, **kwargs):
return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs)
+def expand_path(s):
+ """Expand shell variables and ~"""
+ return os.path.expandvars(compat_expanduser(s))
+
+
def orderedSet(iterable):
""" Remove all duplicates from the input iterable """
res = []
@@ -543,7 +596,7 @@ def _htmlentity_transform(entity_with_semicolon):
numstr = '0%s' % numstr
else:
base = 10
- # See https://github.com/rg3/youtube-dl/issues/7518
+ # See https://github.com/ytdl-org/youtube-dl/issues/7518
try:
return compat_chr(int(numstr, base))
except ValueError:
@@ -559,7 +612,7 @@ def unescapeHTML(s):
assert type(s) == compat_str
return re.sub(
- r'&([^;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
+ r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
def get_subprocess_encoding():
@@ -677,7 +730,12 @@ def bug_reports_message():
return msg
-class ExtractorError(Exception):
+class YoutubeDLError(Exception):
+ """Base exception for YoutubeDL errors."""
+ pass
+
+
+class ExtractorError(YoutubeDLError):
"""Error during info extraction."""
def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
@@ -718,7 +776,19 @@ class RegexNotFoundError(ExtractorError):
pass
-class DownloadError(Exception):
+class GeoRestrictedError(ExtractorError):
+ """Geographic restriction Error exception.
+
+ This exception may be thrown when a video is not available from your
+ geographic location due to geographic restrictions imposed by a website.
+ """
+ def __init__(self, msg, countries=None):
+ super(GeoRestrictedError, self).__init__(msg, expected=True)
+ self.msg = msg
+ self.countries = countries
+
+
+class DownloadError(YoutubeDLError):
"""Download Error exception.
This exception may be thrown by FileDownloader objects if they are not
@@ -732,7 +802,7 @@ class DownloadError(Exception):
self.exc_info = exc_info
-class SameFileError(Exception):
+class SameFileError(YoutubeDLError):
"""Same File exception.
This exception will be thrown by FileDownloader objects if they detect
@@ -741,7 +811,7 @@ class SameFileError(Exception):
pass
-class PostProcessingError(Exception):
+class PostProcessingError(YoutubeDLError):
"""Post Processing exception.
This exception may be raised by PostProcessor's .run() method to
@@ -749,15 +819,16 @@ class PostProcessingError(Exception):
"""
def __init__(self, msg):
+ super(PostProcessingError, self).__init__(msg)
self.msg = msg
-class MaxDownloadsReached(Exception):
+class MaxDownloadsReached(YoutubeDLError):
""" --max-downloads limit has been reached. """
pass
-class UnavailableVideoError(Exception):
+class UnavailableVideoError(YoutubeDLError):
"""Unavailable Format exception.
This exception will be thrown when a video is requested
@@ -766,7 +837,7 @@ class UnavailableVideoError(Exception):
pass
-class ContentTooShortError(Exception):
+class ContentTooShortError(YoutubeDLError):
"""Content Too Short exception.
This exception may be raised by FileDownloader objects when a file they
@@ -775,20 +846,23 @@ class ContentTooShortError(Exception):
"""
def __init__(self, downloaded, expected):
+ super(ContentTooShortError, self).__init__(
+ 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
+ )
# Both in bytes
self.downloaded = downloaded
self.expected = expected
-class XAttrMetadataError(Exception):
+class XAttrMetadataError(YoutubeDLError):
def __init__(self, code=None, msg='Unknown error'):
super(XAttrMetadataError, self).__init__(msg)
self.code = code
self.msg = msg
# Parsing code and msg
- if (self.code in (errno.ENOSPC, errno.EDQUOT) or
- 'No space left' in self.msg or 'Disk quota excedded' in self.msg):
+ if (self.code in (errno.ENOSPC, errno.EDQUOT)
+ or 'No space left' in self.msg or 'Disk quota excedded' in self.msg):
self.reason = 'NO_SPACE'
elif self.code == errno.E2BIG or 'Argument list too long' in self.msg:
self.reason = 'VALUE_TOO_LONG'
@@ -796,25 +870,63 @@ class XAttrMetadataError(Exception):
self.reason = 'NOT_SUPPORTED'
-class XAttrUnavailableError(Exception):
+class XAttrUnavailableError(YoutubeDLError):
pass
def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
# Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
# expected HTTP responses to meet HTTP/1.0 or later (see also
- # https://github.com/rg3/youtube-dl/issues/6727)
+ # https://github.com/ytdl-org/youtube-dl/issues/6727)
if sys.version_info < (3, 0):
- kwargs[b'strict'] = True
- hc = http_class(*args, **kwargs)
+ kwargs['strict'] = True
+ hc = http_class(*args, **compat_kwargs(kwargs))
source_address = ydl_handler._params.get('source_address')
+
if source_address is not None:
+ # This is to workaround _create_connection() from socket where it will try all
+ # address data from getaddrinfo() including IPv6. This filters the result from
+ # getaddrinfo() based on the source_address value.
+ # This is based on the cpython socket.create_connection() function.
+ # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
+ def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
+ host, port = address
+ err = None
+ addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
+ af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6
+ ip_addrs = [addr for addr in addrs if addr[0] == af]
+ if addrs and not ip_addrs:
+ ip_version = 'v4' if af == socket.AF_INET else 'v6'
+ raise socket.error(
+ "No remote IP%s addresses available for connect, can't use '%s' as source address"
+ % (ip_version, source_address[0]))
+ for res in ip_addrs:
+ af, socktype, proto, canonname, sa = res
+ sock = None
+ try:
+ sock = socket.socket(af, socktype, proto)
+ if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
+ sock.settimeout(timeout)
+ sock.bind(source_address)
+ sock.connect(sa)
+ err = None # Explicitly break reference cycle
+ return sock
+ except socket.error as _:
+ err = _
+ if sock is not None:
+ sock.close()
+ if err is not None:
+ raise err
+ else:
+ raise socket.error('getaddrinfo returns an empty list')
+ if hasattr(hc, '_create_connection'):
+ hc._create_connection = _create_connection
sa = (source_address, 0)
if hasattr(hc, 'source_address'): # Python 2.7+
hc.source_address = sa
else: # Python 2.6
def _hc_connect(self, *args, **kwargs):
- sock = compat_socket_create_connection(
+ sock = _create_connection(
(self.host, self.port), self.timeout, sa)
if is_https:
self.sock = ssl.wrap_socket(
@@ -878,14 +990,6 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
except zlib.error:
return zlib.decompress(data)
- @staticmethod
- def addinfourl_wrapper(stream, headers, url, code):
- if hasattr(compat_urllib_request.addinfourl, 'getcode'):
- return compat_urllib_request.addinfourl(stream, headers, url, code)
- ret = compat_urllib_request.addinfourl(stream, headers, url)
- ret.code = code
- return ret
-
def http_request(self, req):
# According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
# always respected by websites, some tend to give out URLs with non percent-encoded
@@ -937,17 +1041,17 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
break
else:
raise original_ioerror
- resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
+ resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
del resp.headers['Content-encoding']
# deflate
if resp.headers.get('Content-encoding', '') == 'deflate':
gz = io.BytesIO(self.deflate(resp.read()))
- resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
+ resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
del resp.headers['Content-encoding']
# Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
- # https://github.com/rg3/youtube-dl/issues/6457).
+ # https://github.com/ytdl-org/youtube-dl/issues/6457).
if 300 <= resp.code < 400:
location = resp.headers.get('Location')
if location:
@@ -1036,6 +1140,49 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
req, **kwargs)
+class YoutubeDLCookieJar(compat_cookiejar.MozillaCookieJar):
+ _HTTPONLY_PREFIX = '#HttpOnly_'
+
+ def save(self, filename=None, ignore_discard=False, ignore_expires=False):
+ # Store session cookies with `expires` set to 0 instead of an empty
+ # string
+ for cookie in self:
+ if cookie.expires is None:
+ cookie.expires = 0
+ compat_cookiejar.MozillaCookieJar.save(self, filename, ignore_discard, ignore_expires)
+
+ def load(self, filename=None, ignore_discard=False, ignore_expires=False):
+ """Load cookies from a file."""
+ if filename is None:
+ if self.filename is not None:
+ filename = self.filename
+ else:
+ raise ValueError(compat_cookiejar.MISSING_FILENAME_TEXT)
+
+ cf = io.StringIO()
+ with open(filename) as f:
+ for line in f:
+ if line.startswith(self._HTTPONLY_PREFIX):
+ line = line[len(self._HTTPONLY_PREFIX):]
+ cf.write(compat_str(line))
+ cf.seek(0)
+ self._really_load(cf, filename, ignore_discard, ignore_expires)
+ # Session cookies are denoted by either `expires` field set to
+ # an empty string or 0. MozillaCookieJar only recognizes the former
+ # (see [1]). So we need force the latter to be recognized as session
+ # cookies on our own.
+ # Session cookies may be important for cookies-based authentication,
+ # e.g. usually, when user does not check 'Remember me' check box while
+ # logging in on a site, some important cookies are stored as session
+ # cookies so that not recognizing them will result in failed login.
+ # 1. https://bugs.python.org/issue17164
+ for cookie in self:
+ # Treat `expires=0` cookies as session cookies
+ if cookie.expires == 0:
+ cookie.expires = None
+ cookie.discard = True
+
+
class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
def __init__(self, cookiejar=None):
compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
@@ -1043,7 +1190,7 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
def http_response(self, request, response):
# Python 2 will choke on next HTTP request in row if there are non-ASCII
# characters in Set-Cookie HTTP header of last response (see
- # https://github.com/rg3/youtube-dl/issues/6769).
+ # https://github.com/ytdl-org/youtube-dl/issues/6769).
# In order to at least prevent crashing we will percent encode Set-Cookie
# header before HTTPCookieProcessor starts processing it.
# if sys.version_info < (3, 0) and response.headers:
@@ -1133,7 +1280,7 @@ def unified_timestamp(date_str, day_first=True):
if date_str is None:
return None
- date_str = date_str.replace(',', ' ')
+ date_str = re.sub(r'[,|]', '', date_str)
pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
timezone, date_str = extract_timezone(date_str)
@@ -1141,6 +1288,16 @@ def unified_timestamp(date_str, day_first=True):
# Remove AM/PM + timezone
date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
+ # Remove unrecognized timezones from ISO 8601 alike timestamps
+ m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P\s*[A-Z]+)$', date_str)
+ if m:
+ date_str = date_str[:-len(m.group('tz'))]
+
+ # Python only supports microseconds, so remove nanoseconds
+ m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str)
+ if m:
+ date_str = m.group(1)
+
for expression in date_formats(day_first):
try:
dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
@@ -1153,7 +1310,7 @@ def unified_timestamp(date_str, day_first=True):
def determine_ext(url, default_ext='unknown_video'):
- if url is None:
+ if url is None or '.' not in url:
return default_ext
guess = url.partition('?')[0].rpartition('.')[2]
if re.match(r'^[A-Za-z0-9]+$', guess):
@@ -1178,7 +1335,7 @@ def date_from_str(date_str):
return today
if date_str == 'yesterday':
return today - datetime.timedelta(days=1)
- match = re.match('(now|today)(?P[+-])(?P