X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2FYoutubeDL.py;h=2a3d6cd4a423227ef998d4f8b6ab4dc7c482a54b;hb=c3deacd5621d9686a32614e550eb4726e66c3899;hp=e8ce586042334c3df3a6b98cff54d762620cf7c4;hpb=e37afbe0b8a1222cb214ad0bec9a53bb7953531d;p=youtube-dl diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index e8ce58604..2a3d6cd4a 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -46,6 +46,7 @@ from .utils import ( DateRange, DEFAULT_OUTTMPL, determine_ext, + determine_protocol, DownloadError, encode_compat_str, encodeFilename, @@ -262,7 +263,7 @@ class YoutubeDL(object): the downloader (see youtube_dl/downloader/common.py): nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle, - xattr_set_filesize, external_downloader_args. + xattr_set_filesize, external_downloader_args, hls_use_mpegts. The following options are used by the post processors: prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available, @@ -898,11 +899,14 @@ class YoutubeDL(object): STR_OPERATORS = { '=': operator.eq, '!=': operator.ne, + '^=': lambda attr, value: attr.startswith(value), + '$=': lambda attr, value: attr.endswith(value), + '*=': lambda attr, value: value in attr, } str_operator_rex = re.compile(r'''(?x) \s*(?Pext|acodec|vcodec|container|protocol) \s*(?P%s)(?P\s*\?)? - \s*(?P[a-zA-Z0-9_-]+) + \s*(?P[a-zA-Z0-9._-]+) \s*$ ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) m = str_operator_rex.search(filter_spec) @@ -1244,6 +1248,12 @@ class YoutubeDL(object): except (ValueError, OverflowError, OSError): pass + # Auto generate title fields corresponding to the *_number fields when missing + # in order to always have clean titles. This is very common for TV series. + for field in ('chapter', 'season', 'episode'): + if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): + info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) + subtitles = info_dict.get('subtitles') if subtitles: for _, subtitle in subtitles.items(): @@ -1300,6 +1310,10 @@ class YoutubeDL(object): # Automatically determine file extension if missing if 'ext' not in format: format['ext'] = determine_ext(format['url']).lower() + # Automatically determine protocol if missing (useful for format + # selection purposes) + if 'protocol' not in format: + format['protocol'] = determine_protocol(format) # Add HTTP headers, so that external programs can use them from the # json output full_format_info = info_dict.copy() @@ -1986,14 +2000,19 @@ class YoutubeDL(object): https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) data_handler = compat_urllib_request_DataHandler() - unknown_handler = compat_urllib_request.UnknownHandler() - handlers = (proxy_handler, https_handler, cookie_processor, ydlh, data_handler, unknown_handler) - # we don't use build_opener because it automatically adds FileHandler, - # which can be used for malicious purposes (see + + # When passing our own FileHandler instance, build_opener won't add the + # default FileHandler and allows us to disable the file protocol, which + # can be used for malicious purposes (see # https://github.com/rg3/youtube-dl/issues/8227) - opener = compat_urllib_request.OpenerDirector() - for handler in handlers: - opener.add_handler(handler) + file_handler = compat_urllib_request.FileHandler() + + def file_open(*args, **kwargs): + raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons') + file_handler.file_open = file_open + + opener = compat_urllib_request.build_opener( + proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler) # Delete the default user-agent header, which would otherwise apply in # cases where our custom HTTP handler doesn't come into play