X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=d22b0313460f4dc42453bcd7772bd59f5dcfe074;hb=3fcfb8e9faf3cf1dcadedd6fecc5158a86d07065;hp=d4951c406c73d8216803f5d9149200787883676b;hpb=ecd1936695e73ba850d0618828b4a40d7d16c091;p=youtube-dl
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index d4951c406..d22b03134 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -10,6 +10,7 @@ import ctypes
import datetime
import email.utils
import errno
+import functools
import gzip
import itertools
import io
@@ -34,7 +35,9 @@ from .compat import (
compat_chr,
compat_getenv,
compat_html_entities,
+ compat_http_client,
compat_parse_qs,
+ compat_socket_create_connection,
compat_str,
compat_urllib_error,
compat_urllib_parse,
@@ -205,6 +208,10 @@ def get_element_by_attribute(attribute, value, html):
def clean_html(html):
"""Clean an HTML snippet into a readable string"""
+
+ if html is None: # Convenience for sanitizing descriptions etc.
+ return html
+
# Newline vs
html = html.replace('\n', ' ')
html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html)
@@ -280,6 +287,8 @@ def sanitize_filename(s, restricted=False, is_id=False):
return '_'
return char
+ # Handle timestamps
+ s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
result = ''.join(map(replace_insane, s))
if not is_id:
while '__' in result:
@@ -387,46 +396,29 @@ def formatSeconds(secs):
return '%d' % secs
-def make_HTTPS_handler(opts_no_check_certificate, **kwargs):
+def make_HTTPS_handler(params, **kwargs):
+ opts_no_check_certificate = params.get('nocheckcertificate', False)
if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9
- context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH)
+ context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH)
if opts_no_check_certificate:
+ context.check_hostname = False
context.verify_mode = ssl.CERT_NONE
try:
- return compat_urllib_request.HTTPSHandler(context=context, **kwargs)
+ return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
except TypeError:
# Python 2.7.8
# (create_default_context present but HTTPSHandler has no context=)
pass
if sys.version_info < (3, 2):
- import httplib
-
- class HTTPSConnectionV3(httplib.HTTPSConnection):
- def __init__(self, *args, **kwargs):
- httplib.HTTPSConnection.__init__(self, *args, **kwargs)
-
- def connect(self):
- sock = socket.create_connection((self.host, self.port), self.timeout)
- if getattr(self, '_tunnel_host', False):
- self.sock = sock
- self._tunnel()
- try:
- self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_TLSv1)
- except ssl.SSLError:
- self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23)
-
- class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler):
- def https_open(self, req):
- return self.do_open(HTTPSConnectionV3, req)
- return HTTPSHandlerV3(**kwargs)
+ return YoutubeDLHTTPSHandler(params, **kwargs)
else: # Python < 3.4
- context = ssl.SSLContext(ssl.PROTOCOL_SSLv23)
+ context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
context.verify_mode = (ssl.CERT_NONE
if opts_no_check_certificate
else ssl.CERT_REQUIRED)
context.set_default_verify_paths()
- return compat_urllib_request.HTTPSHandler(context=context, **kwargs)
+ return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
class ExtractorError(Exception):
@@ -540,6 +532,28 @@ class ContentTooShortError(Exception):
self.expected = expected
+def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
+ hc = http_class(*args, **kwargs)
+ source_address = ydl_handler._params.get('source_address')
+ if source_address is not None:
+ 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(
+ (self.host, self.port), self.timeout, sa)
+ if is_https:
+ self.sock = ssl.wrap_socket(
+ sock, self.key_file, self.cert_file,
+ ssl_version=ssl.PROTOCOL_TLSv1)
+ else:
+ self.sock = sock
+ hc.connect = functools.partial(_hc_connect, hc)
+
+ return hc
+
+
class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
"""Handler for HTTP requests and responses.
@@ -558,6 +572,15 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
public domain.
"""
+ def __init__(self, params, *args, **kwargs):
+ compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs)
+ self._params = params
+
+ def http_open(self, req):
+ return self.do_open(functools.partial(
+ _create_http_connection, self, compat_http_client.HTTPConnection, False),
+ req)
+
@staticmethod
def deflate(data):
try:
@@ -575,17 +598,14 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
def http_request(self, req):
for h, v in std_headers.items():
- if h not in req.headers:
+ # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
+ # The dict keys are capitalized because of this bug by urllib
+ if h.capitalize() not in req.headers:
req.add_header(h, v)
if 'Youtubedl-no-compression' in req.headers:
if 'Accept-encoding' in req.headers:
del req.headers['Accept-encoding']
del req.headers['Youtubedl-no-compression']
- if 'Youtubedl-user-agent' in req.headers:
- if 'User-agent' in req.headers:
- del req.headers['User-agent']
- req.headers['User-agent'] = req.headers['Youtubedl-user-agent']
- del req.headers['Youtubedl-user-agent']
if sys.version_info < (2, 7) and '#' in req.get_full_url():
# Python 2.6 is brain-dead when it comes to fragments
@@ -627,6 +647,18 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
https_response = http_response
+class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
+ def __init__(self, params, https_conn_class=None, *args, **kwargs):
+ compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs)
+ self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection
+ self._params = params
+
+ def https_open(self, req):
+ return self.do_open(functools.partial(
+ _create_http_connection, self, self._https_conn_class, True),
+ req)
+
+
def parse_iso8601(date_str, delimiter='T'):
""" Return a UNIX timestamp from the given date """
@@ -673,11 +705,9 @@ def unified_strdate(date_str, day_first=True):
'%b %dst %Y %I:%M%p',
'%b %dnd %Y %I:%M%p',
'%b %dth %Y %I:%M%p',
+ '%Y %m %d',
'%Y-%m-%d',
'%Y/%m/%d',
- '%d.%m.%Y',
- '%d/%m/%Y',
- '%d/%m/%y',
'%Y/%m/%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
@@ -692,10 +722,16 @@ def unified_strdate(date_str, day_first=True):
]
if day_first:
format_expressions.extend([
+ '%d.%m.%Y',
+ '%d/%m/%Y',
+ '%d/%m/%y',
'%d/%m/%Y %H:%M:%S',
])
else:
format_expressions.extend([
+ '%m.%d.%Y',
+ '%m/%d/%Y',
+ '%m/%d/%y',
'%m/%d/%Y %H:%M:%S',
])
for expression in format_expressions:
@@ -822,6 +858,9 @@ def _windows_write_string(s, out):
except AttributeError:
# If the output stream doesn't have a fileno, it's virtual
return False
+ except io.UnsupportedOperation:
+ # Some strange Windows pseudo files?
+ return False
if fileno not in WIN_OUTPUT_IDS:
return False
@@ -1218,13 +1257,13 @@ def float_or_none(v, scale=1, invscale=1, default=None):
def parse_duration(s):
- if s is None:
+ if not isinstance(s, basestring if sys.version_info < (3, 0) else compat_str):
return None
s = s.strip()
m = re.match(
- r'''(?ix)T?
+ r'''(?ix)(?:P?T)?
(?:
(?P[0-9.]+)\s*(?:mins?|minutes?)\s*|
(?P[0-9.]+)\s*(?:hours?)|
@@ -1559,4 +1598,64 @@ def urlhandle_detect_ext(url_handle):
except AttributeError: # Python < 3
getheader = url_handle.info().getheader
+ cd = getheader('Content-Disposition')
+ if cd:
+ m = re.match(r'attachment;\s*filename="(?P[^"]+)"', cd)
+ if m:
+ e = determine_ext(m.group('filename'), default_ext=None)
+ if e:
+ return e
+
return getheader('Content-Type').split("/")[1]
+
+
+def age_restricted(content_limit, age_limit):
+ """ Returns True iff the content should be blocked """
+
+ if age_limit is None: # No limit set
+ return False
+ if content_limit is None:
+ return False # Content available for everyone
+ return age_limit < content_limit
+
+
+def is_html(first_bytes):
+ """ Detect whether a file contains HTML by examining its first bytes. """
+
+ BOMS = [
+ (b'\xef\xbb\xbf', 'utf-8'),
+ (b'\x00\x00\xfe\xff', 'utf-32-be'),
+ (b'\xff\xfe\x00\x00', 'utf-32-le'),
+ (b'\xff\xfe', 'utf-16-le'),
+ (b'\xfe\xff', 'utf-16-be'),
+ ]
+ for bom, enc in BOMS:
+ if first_bytes.startswith(bom):
+ s = first_bytes[len(bom):].decode(enc, 'replace')
+ break
+ else:
+ s = first_bytes.decode('utf-8', 'replace')
+
+ return re.match(r'^\s*<', s)
+
+
+def determine_protocol(info_dict):
+ protocol = info_dict.get('protocol')
+ if protocol is not None:
+ return protocol
+
+ url = info_dict['url']
+ if url.startswith('rtmp'):
+ return 'rtmp'
+ elif url.startswith('mms'):
+ return 'mms'
+ elif url.startswith('rtsp'):
+ return 'rtsp'
+
+ ext = determine_ext(url)
+ if ext == 'm3u8':
+ return 'm3u8'
+ elif ext == 'f4m':
+ return 'f4m'
+
+ return compat_urllib_parse_urlparse(url).scheme