diff options
Diffstat (limited to 'youtube_dl/utils.py')
| -rw-r--r-- | youtube_dl/utils.py | 130 |
1 files changed, 97 insertions, 33 deletions
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 4d3cbac74..079e8d2c3 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -166,7 +166,7 @@ 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) @@ -205,6 +205,10 @@ 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 <br /> html = html.replace('\n', ' ') html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) @@ -363,7 +367,7 @@ 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) @@ -388,6 +392,17 @@ def formatSeconds(secs): def make_HTTPS_handler(opts_no_check_certificate, **kwargs): + if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 + context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH) + if opts_no_check_certificate: + context.verify_mode = ssl.CERT_NONE + try: + return compat_urllib_request.HTTPSHandler(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 @@ -409,22 +424,12 @@ def make_HTTPS_handler(opts_no_check_certificate, **kwargs): 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 - if opts_no_check_certificate: - context.verify_mode = ssl.CERT_NONE - return compat_urllib_request.HTTPSHandler(context=context, **kwargs) else: # Python < 3.4 context = ssl.SSLContext(ssl.PROTOCOL_SSLv23) 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) @@ -463,6 +468,13 @@ class ExtractorError(Exception): 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): """Error when a regex didn't match""" pass @@ -644,17 +656,19 @@ 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 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', @@ -669,7 +683,6 @@ def unified_strdate(date_str): '%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', @@ -681,6 +694,14 @@ 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 %H:%M:%S', + ]) + else: + format_expressions.extend([ + '%m/%d/%Y %H:%M:%S', + ]) for expression in format_expressions: try: upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') @@ -712,8 +733,10 @@ 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<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) if match is not None: sign = match.group('sign') @@ -808,22 +831,22 @@ def _windows_write_string(s, out): GetStdHandle = ctypes.WINFUNCTYPE( ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)( - ("GetStdHandle", ctypes.windll.kernel32)) + (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)(("WriteConsoleW", ctypes.windll.kernel32)) + ctypes.wintypes.LPVOID)((b"WriteConsoleW", ctypes.windll.kernel32)) written = ctypes.wintypes.DWORD(0) - GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(("GetFileType", ctypes.windll.kernel32)) + 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))( - ("GetConsoleMode", ctypes.windll.kernel32)) + (b"GetConsoleMode", ctypes.windll.kernel32)) INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value def not_a_console(handle): @@ -1024,7 +1047,7 @@ def smuggle_url(url, data): def unsmuggle_url(smug_url, default=None): - if not '#__youtubedl_smuggle' in smug_url: + if '#__youtubedl_smuggle' not in smug_url: return smug_url, default url, _, sdata = smug_url.rpartition('#') jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0] @@ -1090,11 +1113,14 @@ def parse_filesize(s): } 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) + m = re.match( + r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)' % units_re, s) if not m: return None - return int(float(m.group('num')) * _UNIT_TABLE[m.group('unit')]) + num_str = m.group('num').replace(',', '.') + mult = _UNIT_TABLE[m.group('unit')] + return int(float(num_str) * mult) def get_term_width(): @@ -1203,18 +1229,29 @@ def parse_duration(s): m = re.match( r'''(?ix)T? + (?: + (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*| + (?P<only_hours>[0-9.]+)\s*(?:hours?)| + (?: (?:(?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) + (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)? + )$''', s) if not m: return None - res = int(m.group('secs')) + 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'): res += int(m.group('mins')) * 60 - if m.group('hours'): - res += int(m.group('hours')) * 60 * 60 + if m.group('hours'): + res += int(m.group('hours')) * 60 * 60 if m.group('ms'): res += float(m.group('ms')) return res @@ -1236,18 +1273,25 @@ def check_executable(exe, args=[]): def get_exe_version(exe, args=['--version'], - version_re=r'version\s+([0-9._-a-zA-Z]+)', - unrecognized='present'): + version_re=None, unrecognized='present'): """ Returns the version of the specified executable, or False if the executable is not present """ try: - out, err = subprocess.Popen( + out, _ = subprocess.Popen( [exe] + args, stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() except OSError: return False - firstline = out.partition(b'\n')[0].decode('ascii', 'ignore') - m = re.search(version_re, firstline) + 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: @@ -1488,7 +1532,7 @@ def limit_length(s, length): def version_tuple(v): - return [int(e) for e in v.split('.')] + return tuple(int(e) for e in re.split(r'[-.]', v)) def is_outdated_version(version, limit, assume_new=True): @@ -1510,3 +1554,23 @@ def ytdl_is_updateable(): def args_to_str(args): # Get a short string representation for a subprocess command return ' '.join(shlex_quote(a) for a in args) + + +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 + + 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 |