X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=b8c52af74768053d27b4642173022af3b2c6723d;hb=d4f64cabf4ede444b390bb71b90ad4103ce572c0;hp=7c0fb1592914e961a6b1e790bf14fb5525eff08a;hpb=9908e0352872b252494747346589922358aa0375;p=youtube-dl
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 7c0fb1592..b8c52af74 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
@@ -34,14 +35,16 @@ from .compat import (
compat_chr,
compat_getenv,
compat_html_entities,
- compat_html_parser,
+ compat_http_client,
compat_parse_qs,
+ compat_socket_create_connection,
compat_str,
compat_urllib_error,
compat_urllib_parse,
compat_urllib_parse_urlparse,
compat_urllib_request,
compat_urlparse,
+ shlex_quote,
)
@@ -56,6 +59,7 @@ std_headers = {
'Accept-Language': 'en-us,en;q=0.5',
}
+
def preferredencoding():
"""Get preferred encoding.
@@ -64,7 +68,7 @@ def preferredencoding():
"""
try:
pref = locale.getpreferredencoding()
- u'TEST'.encode(pref)
+ 'TEST'.encode(pref)
except:
pref = 'UTF-8'
@@ -72,12 +76,25 @@ def preferredencoding():
def write_json_file(obj, fn):
- """ Encode obj as JSON and write it to fn, atomically """
+ """ Encode obj as JSON and write it to fn, atomically if possible """
+
+ fn = encodeFilename(fn)
+ if sys.version_info < (3, 0) and sys.platform != 'win32':
+ encoding = get_filesystem_encoding()
+ # os.path.basename returns a bytes object, but NamedTemporaryFile
+ # will fail if the filename contains non ascii characters unless we
+ # use a unicode object
+ path_basename = lambda f: os.path.basename(fn).decode(encoding)
+ # the same for os.path.dirname
+ path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
+ else:
+ path_basename = os.path.basename
+ path_dirname = os.path.dirname
args = {
'suffix': '.tmp',
- 'prefix': os.path.basename(fn) + '.',
- 'dir': os.path.dirname(fn),
+ 'prefix': path_basename(fn) + '.',
+ 'dir': path_dirname(fn),
'delete': False,
}
@@ -96,6 +113,13 @@ def write_json_file(obj, fn):
try:
with tf:
json.dump(obj, tf)
+ if sys.platform == 'win32':
+ # Need to remove existing file on Windows, else os.rename raises
+ # WindowsError or FileExistsError.
+ try:
+ os.unlink(fn)
+ except OSError:
+ pass
os.rename(tf.name, fn)
except:
try:
@@ -110,7 +134,7 @@ if sys.version_info >= (2, 7):
""" Find the xpath xpath[@key=val] """
assert re.match(r'^[a-zA-Z-]+$', key)
assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val)
- expr = xpath + u"[@%s='%s']" % (key, val)
+ expr = xpath + "[@%s='%s']" % (key, val)
return node.find(expr)
else:
def find_xpath_attr(node, xpath, key, val):
@@ -126,6 +150,8 @@ else:
# On python2.6 the xml.etree.ElementTree.Element methods don't support
# the namespace parameter
+
+
def xpath_with_ns(path, ns_map):
components = [c.split(':') for c in path.split('/')]
replaced = []
@@ -143,7 +169,7 @@ def xpath_text(node, xpath, name=None, fatal=False):
xpath = xpath.encode('ascii')
n = node.find(xpath)
- if n is None:
+ if n is None or n.text is None:
if fatal:
name = xpath if name is None else name
raise ExtractorError('Could not find XML element %s' % name)
@@ -182,6 +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)
@@ -204,7 +234,7 @@ def sanitize_open(filename, open_mode):
It returns the tuple (stream, definitive_file_name).
"""
try:
- if filename == u'-':
+ if filename == '-':
if sys.platform == 'win32':
import msvcrt
msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
@@ -217,9 +247,9 @@ def sanitize_open(filename, open_mode):
# In case of error, try to remove win32 forbidden chars
alt_filename = os.path.join(
- re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part)
- for path_part in os.path.split(filename)
- )
+ re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part)
+ for path_part in os.path.split(filename)
+ )
if alt_filename == filename:
raise
else:
@@ -236,6 +266,7 @@ def timeconvert(timestr):
timestamp = email.utils.mktime_tz(timetuple)
return timestamp
+
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.
@@ -256,7 +287,9 @@ def sanitize_filename(s, restricted=False, is_id=False):
return '_'
return char
- result = u''.join(map(replace_insane, s))
+ # 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:
result = result.replace('__', '_')
@@ -268,6 +301,7 @@ def sanitize_filename(s, restricted=False, is_id=False):
result = '_'
return result
+
def orderedSet(iterable):
""" Remove all duplicates from the input iterable """
res = []
@@ -286,15 +320,15 @@ def _htmlentity_transform(entity):
mobj = re.match(r'#(x?[0-9]+)', entity)
if mobj is not None:
numstr = mobj.group(1)
- if numstr.startswith(u'x'):
+ if numstr.startswith('x'):
base = 16
- numstr = u'0%s' % numstr
+ numstr = '0%s' % numstr
else:
base = 10
return compat_chr(int(numstr, base))
# Unknown entity in name, return its literal representation
- return (u'&%s;' % entity)
+ return ('&%s;' % entity)
def unescapeHTML(s):
@@ -318,7 +352,7 @@ def encodeFilename(s, for_subprocess=False):
return s
if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
- # Pass u'' directly to use Unicode APIs on Windows 2000 and up
+ # Pass '' directly to use Unicode APIs on Windows 2000 and up
# (Detecting Windows NT 4 is tricky because 'major >= 4' would
# match Windows 9x series as well. Besides, NT 4 is obsolete.)
if not for_subprocess:
@@ -338,7 +372,7 @@ def encodeArgument(s):
if not isinstance(s, compat_str):
# Legacy code that uses byte strings
# Uncomment the following line after fixing all post processors
- #assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
+ # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
s = s.decode('ascii')
return encodeFilename(s, True)
@@ -352,6 +386,7 @@ def decodeOption(optval):
assert isinstance(optval, compat_str)
return optval
+
def formatSeconds(secs):
if secs > 3600:
return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60)
@@ -361,48 +396,34 @@ def formatSeconds(secs):
return '%d' % secs
-def make_HTTPS_handler(opts_no_check_certificate, **kwargs):
- 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)
- elif hasattr(ssl, 'create_default_context'): # Python >= 3.4
- context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH)
- context.options &= ~ssl.OP_NO_SSLv3 # Allow older, not-as-secure SSLv3
+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.SERVER_AUTH)
if opts_no_check_certificate:
+ context.check_hostname = False
context.verify_mode = ssl.CERT_NONE
- return compat_urllib_request.HTTPSHandler(context=context, **kwargs)
+ try:
+ 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):
+ 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()
- try:
- context.load_default_certs()
- except AttributeError:
- pass # Python < 3.4
- return compat_urllib_request.HTTPSHandler(context=context, **kwargs)
+ return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
+
class ExtractorError(Exception):
"""Error during info extraction."""
+
def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
""" tb, if given, is the original traceback (so that it can be printed out).
If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
@@ -413,9 +434,15 @@ class ExtractorError(Exception):
if video_id is not None:
msg = video_id + ': ' + msg
if cause:
- msg += u' (caused by %r)' % cause
+ msg += ' (caused by %r)' % cause
if not expected:
- msg = msg + u'; please report this issue on https://yt-dl.org/bug . Be sure to call youtube-dl with the --verbose flag and include its complete output. Make sure you are using the latest version; type youtube-dl -U to update.'
+ if ytdl_is_updateable():
+ update_cmd = 'type youtube-dl -U to update'
+ else:
+ update_cmd = 'see https://yt-dl.org/update on how to update'
+ msg += '; please report this issue on https://yt-dl.org/bug .'
+ msg += ' Make sure you are using the latest version; %s.' % update_cmd
+ msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.'
super(ExtractorError, self).__init__(msg)
self.traceback = tb
@@ -426,7 +453,14 @@ class ExtractorError(Exception):
def format_traceback(self):
if self.traceback is None:
return None
- return u''.join(traceback.format_tb(self.traceback))
+ return ''.join(traceback.format_tb(self.traceback))
+
+
+class UnsupportedError(ExtractorError):
+ def __init__(self, url):
+ super(UnsupportedError, self).__init__(
+ 'Unsupported URL: %s' % url, expected=True)
+ self.url = url
class RegexNotFoundError(ExtractorError):
@@ -441,6 +475,7 @@ class DownloadError(Exception):
configured to continue on errors. They will contain the appropriate
error message.
"""
+
def __init__(self, msg, exc_info=None):
""" exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
super(DownloadError, self).__init__(msg)
@@ -462,9 +497,11 @@ class PostProcessingError(Exception):
This exception may be raised by PostProcessor's .run() method to
indicate an error in the postprocessing task.
"""
+
def __init__(self, msg):
self.msg = msg
+
class MaxDownloadsReached(Exception):
""" --max-downloads limit has been reached. """
pass
@@ -494,6 +531,29 @@ class ContentTooShortError(Exception):
self.downloaded = downloaded
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.
@@ -512,6 +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:
@@ -529,17 +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
@@ -581,6 +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 """
@@ -606,17 +684,19 @@ def parse_iso8601(date_str, delimiter='T'):
return calendar.timegm(dt.timetuple())
-def unified_strdate(date_str):
+def unified_strdate(date_str, day_first=True):
"""Return a string with the date in the format YYYYMMDD"""
if date_str is None:
return None
-
upload_date = None
- #Replace commas
+ # Replace commas
date_str = date_str.replace(',', ' ')
# %z (UTC offset) is only supported in python>=3.2
date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str)
+ # Remove AM/PM + timezone
+ date_str = re.sub(r'(?i)\s*(?:AM|PM)\s+[A-Z]+', '', date_str)
+
format_expressions = [
'%d %B %Y',
'%d %b %Y',
@@ -625,13 +705,10 @@ def unified_strdate(date_str):
'%b %dst %Y %I:%M%p',
'%b %dnd %Y %I:%M%p',
'%b %dth %Y %I:%M%p',
+ '%Y %m %d',
'%Y-%m-%d',
'%Y/%m/%d',
- '%d.%m.%Y',
- '%d/%m/%Y',
- '%d/%m/%y',
'%Y/%m/%d %H:%M:%S',
- '%d/%m/%Y %H:%M:%S',
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
'%d.%m.%Y %H:%M',
@@ -643,6 +720,20 @@ def unified_strdate(date_str):
'%Y-%m-%dT%H:%M:%S.%f',
'%Y-%m-%dT%H:%M',
]
+ if day_first:
+ format_expressions.extend([
+ '%d.%m.%Y',
+ '%d/%m/%Y',
+ '%d/%m/%y',
+ '%d/%m/%Y %H:%M:%S',
+ ])
+ else:
+ format_expressions.extend([
+ '%m.%d.%Y',
+ '%m/%d/%Y',
+ '%m/%d/%y',
+ '%m/%d/%Y %H:%M:%S',
+ ])
for expression in format_expressions:
try:
upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
@@ -654,25 +745,30 @@ def unified_strdate(date_str):
upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
return upload_date
-def determine_ext(url, default_ext=u'unknown_video'):
+
+def determine_ext(url, default_ext='unknown_video'):
if url is None:
return default_ext
- guess = url.partition(u'?')[0].rpartition(u'.')[2]
+ guess = url.partition('?')[0].rpartition('.')[2]
if re.match(r'^[A-Za-z0-9]+$', guess):
return guess
else:
return default_ext
+
def subtitles_filename(filename, sub_lang, sub_format):
- return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+ return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format
+
def date_from_str(date_str):
"""
Return a datetime object from a string in the format YYYYMMDD or
(now|today)[+-][0-9](day|week|month|year)(s)?"""
today = datetime.date.today()
- if date_str == 'now'or date_str == 'today':
+ if date_str in ('now', 'today'):
return today
+ if date_str == 'yesterday':
+ return today - datetime.timedelta(days=1)
match = re.match('(now|today)(?P[+-])(?P