+
+
+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):
+ kwargs = {}
+ if hasattr(self, '_context'): # python > 2.6
+ kwargs['context'] = self._context
+ if hasattr(self, '_check_hostname'): # python 3.x
+ kwargs['check_hostname'] = self._check_hostname
+ return self.do_open(functools.partial(
+ _create_http_connection, self, self._https_conn_class, True),
+ req, **kwargs)
+
+
+def parse_iso8601(date_str, delimiter='T', timezone=None):
+ """ Return a UNIX timestamp from the given date """
+
+ if date_str is None:
+ return None
+
+ if timezone is None:
+ m = re.search(
+ r'(\.[0-9]+)?(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
+ date_str)
+ if not m:
+ timezone = datetime.timedelta()
+ else:
+ date_str = date_str[:-len(m.group(0))]
+ if not m.group('sign'):
+ timezone = datetime.timedelta()
+ else:
+ sign = 1 if m.group('sign') == '+' else -1
+ timezone = datetime.timedelta(
+ hours=sign * int(m.group('hours')),
+ minutes=sign * int(m.group('minutes')))
+ date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
+ dt = datetime.datetime.strptime(date_str, date_format) - timezone
+ return calendar.timegm(dt.timetuple())
+
+
+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
+ date_str = date_str.replace(',', ' ')
+ # %z (UTC offset) is only supported in python>=3.2
+ if not re.match(r'^[0-9]{1,2}-[0-9]{1,2}-[0-9]{4}$', date_str):
+ date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', 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',
+ '%B %d %Y',
+ '%b %d %Y',
+ '%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',
+ '%Y/%m/%d %H:%M:%S',
+ '%Y-%m-%d %H:%M:%S',
+ '%Y-%m-%d %H:%M:%S.%f',
+ '%d.%m.%Y %H:%M',
+ '%d.%m.%Y %H.%M',
+ '%Y-%m-%dT%H:%M:%SZ',
+ '%Y-%m-%dT%H:%M:%S.%fZ',
+ '%Y-%m-%dT%H:%M:%S.%f0Z',
+ '%Y-%m-%dT%H:%M:%S',
+ '%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',
+ '%d/%m/%Y %H:%M:%S',
+ ])
+ else:
+ format_expressions.extend([
+ '%m-%d-%Y',
+ '%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')
+ except ValueError:
+ pass
+ if upload_date is None:
+ timetuple = email.utils.parsedate_tz(date_str)
+ if timetuple:
+ upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
+ return upload_date
+
+
+def determine_ext(url, default_ext='unknown_video'):
+ if url is None:
+ return default_ext
+ guess = url.partition('?')[0].rpartition('.')[2]
+ if re.match(r'^[A-Za-z0-9]+$', guess):
+ return guess
+ else:
+ return default_ext
+
+
+def subtitles_filename(filename, sub_lang, sub_format):
+ return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format
+
+
+def date_from_str(date_str):
+ """
+ Return a datetime object from a string in the format YYYYMMDD or
+ (now|today)[+-][0-9](day|week|month|year)(s)?"""
+ today = datetime.date.today()
+ if date_str in ('now', 'today'):
+ return today
+ if date_str == 'yesterday':
+ return today - datetime.timedelta(days=1)
+ match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
+ if match is not None:
+ sign = match.group('sign')
+ time = int(match.group('time'))
+ if sign == '-':
+ time = -time
+ unit = match.group('unit')
+ # A bad aproximation?
+ if unit == 'month':
+ unit = 'day'
+ time *= 30
+ elif unit == 'year':
+ unit = 'day'
+ time *= 365
+ unit += 's'
+ delta = datetime.timedelta(**{unit: time})
+ return today + delta
+ return datetime.datetime.strptime(date_str, "%Y%m%d").date()
+
+
+def hyphenate_date(date_str):
+ """
+ Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
+ match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str)
+ if match is not None:
+ return '-'.join(match.groups())
+ else:
+ return date_str
+
+
+class DateRange(object):
+ """Represents a time interval between two dates"""
+
+ def __init__(self, start=None, end=None):
+ """start and end must be strings in the format accepted by date"""
+ if start is not None:
+ self.start = date_from_str(start)
+ else:
+ self.start = datetime.datetime.min.date()
+ if end is not None:
+ self.end = date_from_str(end)
+ else:
+ self.end = datetime.datetime.max.date()
+ if self.start > self.end:
+ raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
+
+ @classmethod
+ def day(cls, day):
+ """Returns a range that only contains the given day"""
+ return cls(day, day)
+
+ def __contains__(self, date):
+ """Check if the date is in the range"""
+ if not isinstance(date, datetime.date):
+ date = date_from_str(date)
+ return self.start <= date <= self.end
+
+ def __str__(self):
+ return '%s - %s' % (self.start.isoformat(), self.end.isoformat())
+
+
+def platform_name():
+ """ Returns the platform name as a compat_str """
+ res = platform.platform()
+ if isinstance(res, bytes):
+ res = res.decode(preferredencoding())
+
+ assert isinstance(res, compat_str)
+ return res
+
+
+def _windows_write_string(s, out):
+ """ Returns True if the string was written using special methods,
+ False if it has yet to be written out."""
+ # Adapted from http://stackoverflow.com/a/3259271/35070
+
+ import ctypes
+ import ctypes.wintypes
+
+ WIN_OUTPUT_IDS = {
+ 1: -11,
+ 2: -12,
+ }
+
+ try:
+ fileno = out.fileno()
+ 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
+
+ GetStdHandle = ctypes.WINFUNCTYPE(
+ ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)(
+ (b"GetStdHandle", ctypes.windll.kernel32))
+ h = GetStdHandle(WIN_OUTPUT_IDS[fileno])
+
+ WriteConsoleW = ctypes.WINFUNCTYPE(
+ ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR,
+ ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD),
+ ctypes.wintypes.LPVOID)((b"WriteConsoleW", ctypes.windll.kernel32))
+ written = ctypes.wintypes.DWORD(0)
+
+ GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)((b"GetFileType", ctypes.windll.kernel32))
+ FILE_TYPE_CHAR = 0x0002
+ FILE_TYPE_REMOTE = 0x8000
+ GetConsoleMode = ctypes.WINFUNCTYPE(
+ ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE,
+ ctypes.POINTER(ctypes.wintypes.DWORD))(
+ (b"GetConsoleMode", ctypes.windll.kernel32))
+ INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value
+
+ def not_a_console(handle):
+ if handle == INVALID_HANDLE_VALUE or handle is None:
+ return True
+ return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR or
+ GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0)
+
+ if not_a_console(h):
+ return False
+
+ def next_nonbmp_pos(s):
+ try:
+ return next(i for i, c in enumerate(s) if ord(c) > 0xffff)
+ except StopIteration:
+ return len(s)
+
+ while s:
+ count = min(next_nonbmp_pos(s), 1024)
+
+ ret = WriteConsoleW(
+ h, s, count if count else 2, ctypes.byref(written), None)
+ if ret == 0:
+ raise OSError('Failed to write string')
+ if not count: # We just wrote a non-BMP character
+ assert written.value == 2
+ s = s[1:]
+ else:
+ assert written.value > 0
+ s = s[written.value:]
+ return True
+
+
+def write_string(s, out=None, encoding=None):
+ if out is None:
+ out = sys.stderr
+ assert type(s) == compat_str
+
+ if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'):
+ if _windows_write_string(s, out):
+ return
+
+ if ('b' in getattr(out, 'mode', '') or
+ sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr
+ byt = s.encode(encoding or preferredencoding(), 'ignore')
+ out.write(byt)
+ elif hasattr(out, 'buffer'):
+ enc = encoding or getattr(out, 'encoding', None) or preferredencoding()
+ byt = s.encode(enc, 'ignore')
+ out.buffer.write(byt)
+ else:
+ out.write(s)
+ out.flush()
+
+
+def bytes_to_intlist(bs):
+ if not bs:
+ return []
+ if isinstance(bs[0], int): # Python 3
+ return list(bs)
+ else:
+ return [ord(c) for c in bs]
+
+
+def intlist_to_bytes(xs):
+ if not xs:
+ return b''
+ return struct_pack('%dB' % len(xs), *xs)
+
+
+# Cross-platform file locking
+if sys.platform == 'win32':
+ import ctypes.wintypes
+ import msvcrt
+
+ class OVERLAPPED(ctypes.Structure):
+ _fields_ = [
+ ('Internal', ctypes.wintypes.LPVOID),
+ ('InternalHigh', ctypes.wintypes.LPVOID),
+ ('Offset', ctypes.wintypes.DWORD),
+ ('OffsetHigh', ctypes.wintypes.DWORD),
+ ('hEvent', ctypes.wintypes.HANDLE),
+ ]
+
+ kernel32 = ctypes.windll.kernel32
+ LockFileEx = kernel32.LockFileEx
+ LockFileEx.argtypes = [
+ ctypes.wintypes.HANDLE, # hFile
+ ctypes.wintypes.DWORD, # dwFlags
+ ctypes.wintypes.DWORD, # dwReserved
+ ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
+ ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
+ ctypes.POINTER(OVERLAPPED) # Overlapped
+ ]
+ LockFileEx.restype = ctypes.wintypes.BOOL
+ UnlockFileEx = kernel32.UnlockFileEx
+ UnlockFileEx.argtypes = [
+ ctypes.wintypes.HANDLE, # hFile
+ ctypes.wintypes.DWORD, # dwReserved
+ ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
+ ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
+ ctypes.POINTER(OVERLAPPED) # Overlapped
+ ]
+ UnlockFileEx.restype = ctypes.wintypes.BOOL
+ whole_low = 0xffffffff
+ whole_high = 0x7fffffff
+
+ def _lock_file(f, exclusive):
+ overlapped = OVERLAPPED()
+ overlapped.Offset = 0
+ overlapped.OffsetHigh = 0
+ overlapped.hEvent = 0
+ f._lock_file_overlapped_p = ctypes.pointer(overlapped)
+ handle = msvcrt.get_osfhandle(f.fileno())
+ if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
+ whole_low, whole_high, f._lock_file_overlapped_p):
+ raise OSError('Locking file failed: %r' % ctypes.FormatError())
+
+ def _unlock_file(f):
+ assert f._lock_file_overlapped_p
+ handle = msvcrt.get_osfhandle(f.fileno())
+ if not UnlockFileEx(handle, 0,
+ whole_low, whole_high, f._lock_file_overlapped_p):
+ raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
+
+else:
+ import fcntl
+
+ def _lock_file(f, exclusive):
+ fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
+
+ def _unlock_file(f):
+ fcntl.flock(f, fcntl.LOCK_UN)
+
+
+class locked_file(object):
+ def __init__(self, filename, mode, encoding=None):
+ assert mode in ['r', 'a', 'w']
+ self.f = io.open(filename, mode, encoding=encoding)
+ self.mode = mode
+
+ def __enter__(self):
+ exclusive = self.mode != 'r'
+ try:
+ _lock_file(self.f, exclusive)
+ except IOError:
+ self.f.close()
+ raise
+ return self
+
+ def __exit__(self, etype, value, traceback):
+ try:
+ _unlock_file(self.f)
+ finally:
+ self.f.close()
+
+ def __iter__(self):
+ return iter(self.f)
+
+ def write(self, *args):
+ return self.f.write(*args)
+
+ def read(self, *args):
+ return self.f.read(*args)
+
+
+def get_filesystem_encoding():
+ encoding = sys.getfilesystemencoding()
+ return encoding if encoding is not None else 'utf-8'
+
+
+def shell_quote(args):
+ quoted_args = []
+ encoding = get_filesystem_encoding()
+ for a in args:
+ if isinstance(a, bytes):
+ # We may get a filename encoded with 'encodeFilename'
+ a = a.decode(encoding)
+ quoted_args.append(pipes.quote(a))
+ return ' '.join(quoted_args)
+
+
+def smuggle_url(url, data):
+ """ Pass additional data in a URL for internal use. """
+
+ sdata = compat_urllib_parse.urlencode(
+ {'__youtubedl_smuggle': json.dumps(data)})
+ return url + '#' + sdata
+
+
+def unsmuggle_url(smug_url, default=None):
+ if '#__youtubedl_smuggle' not in smug_url:
+ return smug_url, default
+ url, _, sdata = smug_url.rpartition('#')
+ jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
+ data = json.loads(jsond)
+ return url, data
+
+
+def format_bytes(bytes):
+ if bytes is None:
+ return 'N/A'
+ if type(bytes) is str:
+ bytes = float(bytes)
+ if bytes == 0.0:
+ exponent = 0
+ else:
+ exponent = int(math.log(bytes, 1024.0))
+ suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
+ converted = float(bytes) / float(1024 ** exponent)
+ return '%.2f%s' % (converted, suffix)
+
+
+def parse_filesize(s):
+ if s is None:
+ return None
+
+ # The lower-case forms are of course incorrect and inofficial,
+ # but we support those too
+ _UNIT_TABLE = {
+ 'B': 1,
+ 'b': 1,
+ 'KiB': 1024,
+ 'KB': 1000,
+ 'kB': 1024,
+ 'Kb': 1000,
+ 'MiB': 1024 ** 2,
+ 'MB': 1000 ** 2,
+ 'mB': 1024 ** 2,
+ 'Mb': 1000 ** 2,
+ 'GiB': 1024 ** 3,
+ 'GB': 1000 ** 3,
+ 'gB': 1024 ** 3,
+ 'Gb': 1000 ** 3,
+ 'TiB': 1024 ** 4,
+ 'TB': 1000 ** 4,
+ 'tB': 1024 ** 4,
+ 'Tb': 1000 ** 4,
+ 'PiB': 1024 ** 5,
+ 'PB': 1000 ** 5,
+ 'pB': 1024 ** 5,
+ 'Pb': 1000 ** 5,
+ 'EiB': 1024 ** 6,
+ 'EB': 1000 ** 6,
+ 'eB': 1024 ** 6,
+ 'Eb': 1000 ** 6,
+ 'ZiB': 1024 ** 7,
+ 'ZB': 1000 ** 7,
+ 'zB': 1024 ** 7,
+ 'Zb': 1000 ** 7,
+ 'YiB': 1024 ** 8,
+ 'YB': 1000 ** 8,
+ 'yB': 1024 ** 8,
+ 'Yb': 1000 ** 8,
+ }
+
+ units_re = '|'.join(re.escape(u) for u in _UNIT_TABLE)
+ m = re.match(
+ r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)' % units_re, s)
+ if not m:
+ return None
+
+ num_str = m.group('num').replace(',', '.')
+ mult = _UNIT_TABLE[m.group('unit')]
+ return int(float(num_str) * mult)
+
+
+def month_by_name(name):
+ """ Return the number of a month by (locale-independently) English name """
+
+ try:
+ return ENGLISH_MONTH_NAMES.index(name) + 1
+ except ValueError:
+ return None
+
+
+def month_by_abbreviation(abbrev):
+ """ Return the number of a month by (locale-independently) English
+ abbreviations """
+
+ try:
+ return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1
+ except ValueError:
+ return None
+
+
+def fix_xml_ampersands(xml_str):
+ """Replace all the '&' by '&' in XML"""
+ return re.sub(
+ r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
+ '&',
+ xml_str)
+
+
+def setproctitle(title):
+ assert isinstance(title, compat_str)
+ try:
+ libc = ctypes.cdll.LoadLibrary("libc.so.6")
+ except OSError:
+ return
+ title_bytes = title.encode('utf-8')
+ buf = ctypes.create_string_buffer(len(title_bytes))
+ buf.value = title_bytes
+ try:
+ libc.prctl(15, buf, 0, 0, 0)
+ except AttributeError:
+ return # Strange libc, just skip this
+
+
+def remove_start(s, start):
+ if s.startswith(start):
+ return s[len(start):]
+ return s
+
+
+def remove_end(s, end):
+ if s.endswith(end):
+ return s[:-len(end)]
+ return s
+
+
+def url_basename(url):
+ path = compat_urlparse.urlparse(url).path
+ return path.strip('/').split('/')[-1]
+
+
+class HEADRequest(compat_urllib_request.Request):
+ def get_method(self):
+ return "HEAD"
+
+
+def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
+ if get_attr:
+ if v is not None:
+ v = getattr(v, get_attr, None)
+ if v == '':
+ v = None
+ return default if v is None else (int(v) * invscale // scale)
+
+
+def str_or_none(v, default=None):
+ return default if v is None else compat_str(v)
+
+
+def str_to_int(int_str):
+ """ A more relaxed version of int_or_none """
+ if int_str is None:
+ return None
+ int_str = re.sub(r'[,\.\+]', '', int_str)
+ return int(int_str)
+
+
+def float_or_none(v, scale=1, invscale=1, default=None):
+ return default if v is None else (float(v) * invscale / scale)
+
+
+def parse_duration(s):
+ if not isinstance(s, compat_basestring):
+ return None
+
+ s = s.strip()
+
+ m = re.match(
+ r'''(?ix)(?:P?T)?
+ (?:
+ (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*|
+ (?P<only_hours>[0-9.]+)\s*(?:hours?)|
+
+ \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*|
+ (?:
+ (?:
+ (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
+ (?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*
+ )?
+ (?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*
+ )?
+ (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?
+ )$''', s)
+ if not m:
+ return None
+ res = 0
+ if m.group('only_mins'):
+ return float_or_none(m.group('only_mins'), invscale=60)
+ if m.group('only_hours'):
+ return float_or_none(m.group('only_hours'), invscale=60 * 60)
+ if m.group('secs'):
+ res += int(m.group('secs'))
+ if m.group('mins_reversed'):
+ res += int(m.group('mins_reversed')) * 60
+ if m.group('mins'):
+ res += int(m.group('mins')) * 60
+ if m.group('hours'):
+ res += int(m.group('hours')) * 60 * 60
+ if m.group('hours_reversed'):
+ res += int(m.group('hours_reversed')) * 60 * 60
+ if m.group('days'):
+ res += int(m.group('days')) * 24 * 60 * 60
+ if m.group('ms'):
+ res += float(m.group('ms'))
+ return res
+
+
+def prepend_extension(filename, ext, expected_real_ext=None):
+ name, real_ext = os.path.splitext(filename)
+ return (
+ '{0}.{1}{2}'.format(name, ext, real_ext)
+ if not expected_real_ext or real_ext[1:] == expected_real_ext
+ else '{0}.{1}'.format(filename, ext))
+
+
+def replace_extension(filename, ext, expected_real_ext=None):
+ name, real_ext = os.path.splitext(filename)
+ return '{0}.{1}'.format(
+ name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename,
+ ext)
+
+
+def check_executable(exe, args=[]):
+ """ Checks if the given binary is installed somewhere in PATH, and returns its name.
+ args can be a list of arguments for a short output (like -version) """
+ try:
+ subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate()
+ except OSError:
+ return False
+ return exe
+
+
+def get_exe_version(exe, args=['--version'],
+ version_re=None, unrecognized='present'):
+ """ Returns the version of the specified executable,
+ or False if the executable is not present """
+ try:
+ out, _ = subprocess.Popen(
+ [encodeArgument(exe)] + args,
+ stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate()
+ except OSError:
+ return False
+ if isinstance(out, bytes): # Python 2.x
+ out = out.decode('ascii', 'ignore')
+ return detect_exe_version(out, version_re, unrecognized)
+
+
+def detect_exe_version(output, version_re=None, unrecognized='present'):
+ assert isinstance(output, compat_str)
+ if version_re is None:
+ version_re = r'version\s+([-0-9._a-zA-Z]+)'
+ m = re.search(version_re, output)
+ if m:
+ return m.group(1)
+ else:
+ return unrecognized
+
+
+class PagedList(object):
+ def __len__(self):
+ # This is only useful for tests
+ return len(self.getslice())
+
+
+class OnDemandPagedList(PagedList):
+ def __init__(self, pagefunc, pagesize):
+ self._pagefunc = pagefunc
+ self._pagesize = pagesize
+
+ def getslice(self, start=0, end=None):
+ res = []
+ for pagenum in itertools.count(start // self._pagesize):
+ firstid = pagenum * self._pagesize
+ nextfirstid = pagenum * self._pagesize + self._pagesize
+ if start >= nextfirstid:
+ continue
+
+ page_results = list(self._pagefunc(pagenum))
+
+ startv = (
+ start % self._pagesize
+ if firstid <= start < nextfirstid
+ else 0)
+
+ endv = (
+ ((end - 1) % self._pagesize) + 1
+ if (end is not None and firstid <= end <= nextfirstid)
+ else None)
+
+ if startv != 0 or endv is not None:
+ page_results = page_results[startv:endv]
+ res.extend(page_results)
+
+ # A little optimization - if current page is not "full", ie. does
+ # not contain page_size videos then we can assume that this page
+ # is the last one - there are no more ids on further pages -
+ # i.e. no need to query again.
+ if len(page_results) + startv < self._pagesize:
+ break
+
+ # If we got the whole page, but the next page is not interesting,
+ # break out early as well
+ if end == nextfirstid:
+ break
+ return res
+
+
+class InAdvancePagedList(PagedList):
+ def __init__(self, pagefunc, pagecount, pagesize):
+ self._pagefunc = pagefunc
+ self._pagecount = pagecount
+ self._pagesize = pagesize
+
+ def getslice(self, start=0, end=None):
+ res = []
+ start_page = start // self._pagesize
+ end_page = (
+ self._pagecount if end is None else (end // self._pagesize + 1))
+ skip_elems = start - start_page * self._pagesize
+ only_more = None if end is None else end - start
+ for pagenum in range(start_page, end_page):
+ page = list(self._pagefunc(pagenum))
+ if skip_elems:
+ page = page[skip_elems:]
+ skip_elems = None
+ if only_more is not None:
+ if len(page) < only_more:
+ only_more -= len(page)
+ else:
+ page = page[:only_more]
+ res.extend(page)
+ break
+ res.extend(page)
+ return res
+
+
+def uppercase_escape(s):
+ unicode_escape = codecs.getdecoder('unicode_escape')
+ return re.sub(
+ r'\\U[0-9a-fA-F]{8}',
+ lambda m: unicode_escape(m.group(0))[0],
+ s)
+
+
+def lowercase_escape(s):
+ unicode_escape = codecs.getdecoder('unicode_escape')
+ return re.sub(
+ r'\\u[0-9a-fA-F]{4}',
+ lambda m: unicode_escape(m.group(0))[0],
+ s)
+
+
+def escape_rfc3986(s):
+ """Escape non-ASCII characters as suggested by RFC 3986"""
+ if sys.version_info < (3, 0) and isinstance(s, compat_str):
+ s = s.encode('utf-8')
+ return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
+
+
+def escape_url(url):
+ """Escape URL as suggested by RFC 3986"""
+ url_parsed = compat_urllib_parse_urlparse(url)
+ return url_parsed._replace(
+ path=escape_rfc3986(url_parsed.path),
+ params=escape_rfc3986(url_parsed.params),
+ query=escape_rfc3986(url_parsed.query),
+ fragment=escape_rfc3986(url_parsed.fragment)
+ ).geturl()
+
+try:
+ struct.pack('!I', 0)
+except TypeError:
+ # In Python 2.6 (and some 2.7 versions), struct requires a bytes argument
+ def struct_pack(spec, *args):
+ if isinstance(spec, compat_str):
+ spec = spec.encode('ascii')
+ return struct.pack(spec, *args)
+
+ def struct_unpack(spec, *args):
+ if isinstance(spec, compat_str):
+ spec = spec.encode('ascii')
+ return struct.unpack(spec, *args)
+else:
+ struct_pack = struct.pack
+ struct_unpack = struct.unpack
+
+
+def read_batch_urls(batch_fd):
+ def fixup(url):
+ if not isinstance(url, compat_str):
+ url = url.decode('utf-8', 'replace')
+ BOM_UTF8 = '\xef\xbb\xbf'
+ if url.startswith(BOM_UTF8):
+ url = url[len(BOM_UTF8):]
+ url = url.strip()
+ if url.startswith(('#', ';', ']')):
+ return False
+ return url
+
+ with contextlib.closing(batch_fd) as fd:
+ return [url for url in map(fixup, fd) if url]
+
+
+def urlencode_postdata(*args, **kargs):
+ return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')
+
+
+try:
+ etree_iter = xml.etree.ElementTree.Element.iter
+except AttributeError: # Python <=2.6
+ etree_iter = lambda n: n.findall('.//*')
+
+
+def parse_xml(s):
+ class TreeBuilder(xml.etree.ElementTree.TreeBuilder):
+ def doctype(self, name, pubid, system):
+ pass # Ignore doctypes
+
+ parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder())
+ kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {}
+ tree = xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs)
+ # Fix up XML parser in Python 2.x
+ if sys.version_info < (3, 0):
+ for n in etree_iter(tree):
+ if n.text is not None:
+ if not isinstance(n.text, compat_str):
+ n.text = n.text.decode('utf-8')
+ return tree
+
+
+US_RATINGS = {
+ 'G': 0,
+ 'PG': 10,
+ 'PG-13': 13,
+ 'R': 16,
+ 'NC': 18,
+}
+
+
+def parse_age_limit(s):
+ if s is None:
+ return None
+ m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
+ return int(m.group('age')) if m else US_RATINGS.get(s, None)
+
+
+def strip_jsonp(code):
+ return re.sub(
+ r'(?s)^[a-zA-Z0-9_]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
+
+
+def js_to_json(code):
+ def fix_kv(m):
+ v = m.group(0)
+ if v in ('true', 'false', 'null'):
+ return v
+ if v.startswith('"'):
+ return v
+ if v.startswith("'"):
+ v = v[1:-1]
+ v = re.sub(r"\\\\|\\'|\"", lambda m: {
+ '\\\\': '\\\\',
+ "\\'": "'",
+ '"': '\\"',
+ }[m.group(0)], v)
+ return '"%s"' % v
+
+ res = re.sub(r'''(?x)
+ "(?:[^"\\]*(?:\\\\|\\['"nu]))*[^"\\]*"|
+ '(?:[^'\\]*(?:\\\\|\\['"nu]))*[^'\\]*'|
+ [a-zA-Z_][.a-zA-Z_0-9]*
+ ''', fix_kv, code)
+ res = re.sub(r',(\s*[\]}])', lambda m: m.group(1), res)
+ return res
+
+
+def qualities(quality_ids):
+ """ Get a numeric quality value out of a list of possible values """
+ def q(qid):
+ try:
+ return quality_ids.index(qid)
+ except ValueError:
+ return -1
+ return q
+
+
+DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s'
+
+
+def limit_length(s, length):
+ """ Add ellipses to overly long strings """
+ if s is None:
+ return None
+ ELLIPSES = '...'
+ if len(s) > length:
+ return s[:length - len(ELLIPSES)] + ELLIPSES
+ return s
+
+
+def version_tuple(v):
+ return tuple(int(e) for e in re.split(r'[-.]', v))
+
+
+def is_outdated_version(version, limit, assume_new=True):
+ if not version:
+ return not assume_new
+ try:
+ return version_tuple(version) < version_tuple(limit)
+ except ValueError:
+ return not assume_new
+
+
+def ytdl_is_updateable():
+ """ Returns if youtube-dl can be updated with -U """
+ from zipimport import zipimporter
+
+ return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen')
+
+
+def args_to_str(args):
+ # Get a short string representation for a subprocess command
+ return ' '.join(shlex_quote(a) for a in args)
+
+
+def mimetype2ext(mt):
+ _, _, res = mt.rpartition('/')
+
+ return {
+ 'x-ms-wmv': 'wmv',
+ 'x-mp4-fragmented': 'mp4',
+ }.get(res, res)
+
+
+def urlhandle_detect_ext(url_handle):
+ try:
+ url_handle.headers
+ getheader = lambda h: url_handle.headers[h]
+ except AttributeError: # Python < 3
+ getheader = url_handle.info().getheader
+
+ cd = getheader('Content-Disposition')
+ if cd:
+ m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd)
+ if m:
+ e = determine_ext(m.group('filename'), default_ext=None)
+ if e:
+ return e
+
+ return mimetype2ext(getheader('Content-Type'))
+
+
+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
+
+
+def render_table(header_row, data):
+ """ Render a list of rows, each as a list of values """
+ table = [header_row] + data
+ max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)]
+ format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s'
+ return '\n'.join(format_str % tuple(row) for row in table)
+
+
+def _match_one(filter_part, dct):
+ COMPARISON_OPERATORS = {
+ '<': operator.lt,
+ '<=': operator.le,
+ '>': operator.gt,
+ '>=': operator.ge,
+ '=': operator.eq,
+ '!=': operator.ne,
+ }
+ operator_rex = re.compile(r'''(?x)\s*
+ (?P<key>[a-z_]+)
+ \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
+ (?:
+ (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
+ (?P<strval>(?![0-9.])[a-z0-9A-Z]*)
+ )
+ \s*$
+ ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
+ m = operator_rex.search(filter_part)
+ if m:
+ op = COMPARISON_OPERATORS[m.group('op')]
+ if m.group('strval') is not None:
+ if m.group('op') not in ('=', '!='):
+ raise ValueError(
+ 'Operator %s does not support string values!' % m.group('op'))
+ comparison_value = m.group('strval')
+ else:
+ try:
+ comparison_value = int(m.group('intval'))
+ except ValueError:
+ comparison_value = parse_filesize(m.group('intval'))
+ if comparison_value is None:
+ comparison_value = parse_filesize(m.group('intval') + 'B')
+ if comparison_value is None:
+ raise ValueError(
+ 'Invalid integer value %r in filter part %r' % (
+ m.group('intval'), filter_part))
+ actual_value = dct.get(m.group('key'))
+ if actual_value is None:
+ return m.group('none_inclusive')
+ return op(actual_value, comparison_value)
+
+ UNARY_OPERATORS = {
+ '': lambda v: v is not None,
+ '!': lambda v: v is None,
+ }
+ operator_rex = re.compile(r'''(?x)\s*
+ (?P<op>%s)\s*(?P<key>[a-z_]+)
+ \s*$
+ ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
+ m = operator_rex.search(filter_part)
+ if m:
+ op = UNARY_OPERATORS[m.group('op')]
+ actual_value = dct.get(m.group('key'))
+ return op(actual_value)
+
+ raise ValueError('Invalid filter part %r' % filter_part)
+
+
+def match_str(filter_str, dct):
+ """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
+
+ return all(
+ _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
+
+
+def match_filter_func(filter_str):
+ def _match_func(info_dict):
+ if match_str(filter_str, info_dict):
+ return None
+ else:
+ video_title = info_dict.get('title', info_dict.get('id', 'video'))
+ return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
+ return _match_func
+
+
+def parse_dfxp_time_expr(time_expr):
+ if not time_expr:
+ return 0.0
+
+ mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
+ if mobj:
+ return float(mobj.group('time_offset'))
+
+ mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:\.\d+)?)$', time_expr)
+ if mobj:
+ return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3))
+
+
+def srt_subtitles_timecode(seconds):
+ return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
+
+
+def dfxp2srt(dfxp_data):
+ _x = functools.partial(xpath_with_ns, ns_map={'ttml': 'http://www.w3.org/ns/ttml'})
+
+ def parse_node(node):
+ str_or_empty = functools.partial(str_or_none, default='')
+
+ out = str_or_empty(node.text)
+
+ for child in node:
+ if child.tag == _x('ttml:br'):
+ out += '\n' + str_or_empty(child.tail)
+ elif child.tag == _x('ttml:span'):
+ out += str_or_empty(parse_node(child))
+ else:
+ out += str_or_empty(xml.etree.ElementTree.tostring(child))
+
+ return out
+
+ dfxp = xml.etree.ElementTree.fromstring(dfxp_data.encode('utf-8'))
+ out = []
+ paras = dfxp.findall(_x('.//ttml:p'))
+
+ for para, index in zip(paras, itertools.count(1)):
+ begin_time = parse_dfxp_time_expr(para.attrib['begin'])
+ end_time = parse_dfxp_time_expr(para.attrib.get('end'))
+ if not end_time:
+ end_time = begin_time + parse_dfxp_time_expr(para.attrib['dur'])
+ out.append('%d\n%s --> %s\n%s\n\n' % (
+ index,
+ srt_subtitles_timecode(begin_time),
+ srt_subtitles_timecode(end_time),
+ parse_node(para)))
+
+ return ''.join(out)
+
+
+class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
+ def __init__(self, proxies=None):
+ # Set default handlers
+ for type in ('http', 'https'):
+ setattr(self, '%s_open' % type,
+ lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
+ meth(r, proxy, type))
+ return compat_urllib_request.ProxyHandler.__init__(self, proxies)
+
+ def proxy_open(self, req, proxy, type):
+ req_proxy = req.headers.get('Ytdl-request-proxy')
+ if req_proxy is not None:
+ proxy = req_proxy
+ del req.headers['Ytdl-request-proxy']
+
+ if proxy == '__noproxy__':
+ return None # No Proxy
+ return compat_urllib_request.ProxyHandler.proxy_open(
+ self, req, proxy, type)
Rapsys Git / youtubedl / blobdiff