X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=ed9ed9ed63ec9b40d929f83cb2e56ee4d63f9e7f;hb=9354a5fad4521687eb9f08c1a42848621857400b;hp=ef14f9a367197164ad15fc0ab6d15f47ef2c9b59;hpb=55969016e96fded28b97b2ef3bbf66efa83d6afb;p=youtube-dl diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index ef14f9a36..ed9ed9ed6 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -37,6 +37,7 @@ from .compat import ( compat_chr, compat_html_entities, compat_http_client, + compat_kwargs, compat_parse_qs, compat_socket_create_connection, compat_str, @@ -75,7 +76,7 @@ def preferredencoding(): try: pref = locale.getpreferredencoding() 'TEST'.encode(pref) - except: + except Exception: pref = 'UTF-8' return pref @@ -114,7 +115,7 @@ def write_json_file(obj, fn): 'encoding': 'utf-8', }) - tf = tempfile.NamedTemporaryFile(**args) + tf = tempfile.NamedTemporaryFile(**compat_kwargs(args)) try: with tf: @@ -127,7 +128,7 @@ def write_json_file(obj, fn): except OSError: pass os.rename(tf.name, fn) - except: + except Exception: try: os.remove(tf.name) except OSError: @@ -252,15 +253,12 @@ def sanitize_open(filename, open_mode): raise # In case of error, try to remove win32 forbidden chars - alt_filename = os.path.join( - re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part) - for path_part in os.path.split(filename) - ) + alt_filename = sanitize_path(filename) if alt_filename == filename: raise else: # An exception here should be caught in the caller - stream = open(encodeFilename(filename), open_mode) + stream = open(encodeFilename(alt_filename), open_mode) return (stream, alt_filename) @@ -311,6 +309,24 @@ def sanitize_filename(s, restricted=False, is_id=False): return result +def sanitize_path(s): + """Sanitizes and normalizes path on Windows""" + if sys.platform != 'win32': + return s + drive_or_unc, _ = os.path.splitdrive(s) + if sys.version_info < (2, 7) and not drive_or_unc: + drive_or_unc, _ = os.path.splitunc(s) + norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) + if drive_or_unc: + norm_path.pop(0) + sanitized_path = [ + path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|\.$)', '#', 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) + + def orderedSet(iterable): """ Remove all duplicates from the input iterable """ res = [] @@ -326,7 +342,7 @@ def _htmlentity_transform(entity): if entity in compat_html_entities.name2codepoint: return compat_chr(compat_html_entities.name2codepoint[entity]) - mobj = re.match(r'#(x?[0-9]+)', entity) + mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) if mobj is not None: numstr = mobj.group(1) if numstr.startswith('x'): @@ -349,6 +365,18 @@ def unescapeHTML(s): r'&([^;]+);', lambda m: _htmlentity_transform(m.group(1)), s) +def get_subprocess_encoding(): + if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: + # For subprocess calls, encode with locale encoding + # Refer to http://stackoverflow.com/a/9951851/35070 + encoding = preferredencoding() + else: + encoding = sys.getfilesystemencoding() + if encoding is None: + encoding = 'utf-8' + return encoding + + def encodeFilename(s, for_subprocess=False): """ @param s The name of the file @@ -360,21 +388,24 @@ def encodeFilename(s, for_subprocess=False): if sys.version_info >= (3, 0): return s - if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: - # 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: - return s - else: - # For subprocess calls, encode with locale encoding - # Refer to http://stackoverflow.com/a/9951851/35070 - encoding = preferredencoding() - else: - encoding = sys.getfilesystemencoding() - if encoding is None: - encoding = 'utf-8' - return s.encode(encoding, 'ignore') + # 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 and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: + return s + + return s.encode(get_subprocess_encoding(), 'ignore') + + +def decodeFilename(b, for_subprocess=False): + + if sys.version_info >= (3, 0): + return b + + if not isinstance(b, bytes): + return b + + return b.decode(get_subprocess_encoding(), 'ignore') def encodeArgument(s): @@ -386,6 +417,10 @@ def encodeArgument(s): return encodeFilename(s, True) +def decodeArgument(b): + return decodeFilename(b, True) + + def decodeOption(optval): if optval is None: return optval @@ -430,6 +465,17 @@ def make_HTTPS_handler(params, **kwargs): return YoutubeDLHTTPSHandler(params, context=context, **kwargs) +def bug_reports_message(): + 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.' + return msg + + class ExtractorError(Exception): """Error during info extraction.""" @@ -445,13 +491,7 @@ class ExtractorError(Exception): if cause: msg += ' (caused by %r)' % cause if not expected: - 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.' + msg += bug_reports_message() super(ExtractorError, self).__init__(msg) self.traceback = tb @@ -708,7 +748,8 @@ def unified_strdate(date_str, day_first=True): # 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) + 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) @@ -737,6 +778,7 @@ 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', @@ -744,6 +786,7 @@ def unified_strdate(date_str, day_first=True): ]) else: format_expressions.extend([ + '%m-%d-%Y', '%m.%d.%Y', '%m/%d/%Y', '%m/%d/%y', @@ -1079,15 +1122,6 @@ def shell_quote(args): return ' '.join(quoted_args) -def takewhile_inclusive(pred, seq): - """ Like itertools.takewhile, but include the latest evaluated element - (the first element so that Not pred(e)) """ - for e in seq: - yield e - if not pred(e): - return - - def smuggle_url(url, data): """ Pass additional data in a URL for internal use. """ @@ -1308,9 +1342,19 @@ def parse_duration(s): return res -def prepend_extension(filename, ext): +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}{2}'.format(name, ext, real_ext) + 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=[]): @@ -1329,7 +1373,7 @@ def get_exe_version(exe, args=['--version'], or False if the executable is not present """ try: out, _ = subprocess.Popen( - [exe] + args, + [encodeArgument(exe)] + args, stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() except OSError: return False @@ -1435,6 +1479,14 @@ def uppercase_escape(s): 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): @@ -1555,7 +1607,7 @@ def js_to_json(code): '(?:[^'\\]*(?:\\\\|\\['"nu]))*[^'\\]*'| [a-zA-Z_][.a-zA-Z_0-9]* ''', fix_kv, code) - res = re.sub(r',(\s*\])', lambda m: m.group(1), res) + res = re.sub(r',(\s*[\]}])', lambda m: m.group(1), res) return res @@ -1770,6 +1822,59 @@ def match_filter_func(filter_str): return _match_func +def parse_dfxp_time_expr(time_expr): + if not time_expr: + return 0.0 + + mobj = re.match(r'^(?P\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 @@ -1789,18 +1894,3 @@ class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): return None # No Proxy return compat_urllib_request.ProxyHandler.proxy_open( self, req, proxy, type) - - -def url_sanitize_consecutive_slashes(url): - """Sanitize URLs with consecutive slashes - - For example, transform both - http://hostname/foo//bar/filename.html - and - http://hostname//foo/bar/filename.html - into - http://hostname/foo/bar/filename.html - """ - parsed_url = list(compat_urlparse.urlparse(url)) - parsed_url[2] = re.sub(r'/{2,}', '/', parsed_url[2]) - return compat_urlparse.urlunparse(parsed_url)