From 7465e0aee2301c3e86fe38d6e0ef5ad01c16ec79 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 3 Mar 2019 06:25:45 +0700 Subject: [PATCH] [spankbang] Fix extraction (closes #20023) --- youtube_dl/extractor/spankbang.py | 45 +++++++++++++++++++++++++++---- 1 file changed, 40 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/spankbang.py b/youtube_dl/extractor/spankbang.py index fbe6ef31a..f11d728ca 100644 --- a/youtube_dl/extractor/spankbang.py +++ b/youtube_dl/extractor/spankbang.py @@ -9,6 +9,8 @@ from ..utils import ( parse_duration, parse_resolution, str_to_int, + url_or_none, + urlencode_postdata, ) @@ -64,16 +66,49 @@ class SpankBangIE(InfoExtractor): 'Video %s is not available' % video_id, expected=True) formats = [] - for mobj in re.finditer( - r'stream_url_(?P[^\s=]+)\s*=\s*(["\'])(?P(?:(?!\2).)+)\2', - webpage): - format_id, format_url = mobj.group('id', 'url') + + def extract_format(format_id, format_url): + f_url = url_or_none(format_url) + if not f_url: + return f = parse_resolution(format_id) f.update({ - 'url': format_url, + 'url': f_url, 'format_id': format_id, }) formats.append(f) + + STREAM_URL_PREFIX = 'stream_url_' + + for mobj in re.finditer( + r'%s(?P[^\s=]+)\s*=\s*(["\'])(?P(?:(?!\2).)+)\2' + % STREAM_URL_PREFIX, webpage): + extract_format(mobj.group('id', 'url')) + + if not formats: + stream_key = self._search_regex( + r'data-streamkey\s*=\s*(["\'])(?P(?:(?!\1).)+)\1', + webpage, 'stream key', group='value') + + sb_csrf_session = self._get_cookies( + 'https://spankbang.com')['sb_csrf_session'].value + + stream = self._download_json( + 'https://spankbang.com/api/videos/stream', video_id, + 'Downloading stream JSON', data=urlencode_postdata({ + 'id': stream_key, + 'data': 0, + 'sb_csrf_session': sb_csrf_session, + }), headers={ + 'Referer': url, + 'X-CSRFToken': sb_csrf_session, + }) + + for format_id, format_url in stream.items(): + if format_id.startswith(STREAM_URL_PREFIX): + extract_format( + format_id[len(STREAM_URL_PREFIX):], format_url) + self._sort_formats(formats) title = self._html_search_regex(