diff options
| -rw-r--r-- | test/test_utils.py | 16 | ||||
| -rw-r--r-- | youtube_dl/extractor/sohu.py | 96 | ||||
| -rw-r--r-- | youtube_dl/utils.py | 15 |
3 files changed, 114 insertions, 13 deletions
diff --git a/test/test_utils.py b/test/test_utils.py index 8f790bf0a..4f0ffd482 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -55,6 +55,7 @@ from youtube_dl.utils import ( xpath_with_ns, render_table, match_str, + url_sanitize_consecutive_slashes, ) @@ -538,6 +539,21 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4') 'like_count > 100 & dislike_count <? 50 & description', {'like_count': 190, 'dislike_count': 10})) + def test_url_sanitize_consecutive_slashes(self): + self.assertEqual(url_sanitize_consecutive_slashes( + 'http://hostname/foo//bar/filename.html'), + 'http://hostname/foo/bar/filename.html') + self.assertEqual(url_sanitize_consecutive_slashes( + 'http://hostname//foo/bar/filename.html'), + 'http://hostname/foo/bar/filename.html') + self.assertEqual(url_sanitize_consecutive_slashes( + 'http://hostname//'), 'http://hostname/') + self.assertEqual(url_sanitize_consecutive_slashes( + 'http://hostname/foo/bar/filename.html'), + 'http://hostname/foo/bar/filename.html') + self.assertEqual(url_sanitize_consecutive_slashes( + 'http://hostname/'), 'http://hostname/') + if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py index c04791997..ea5cc06b9 100644 --- a/youtube_dl/extractor/sohu.py +++ b/youtube_dl/extractor/sohu.py @@ -4,22 +4,87 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .common import compat_str +from ..compat import ( + compat_str, + compat_urllib_request +) +from ..utils import url_sanitize_consecutive_slashes class SohuIE(InfoExtractor): _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?' - _TEST = { + _TESTS = [{ + 'note': 'This video is available only in Mainland China', 'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super', - 'md5': 'bde8d9a6ffd82c63a1eefaef4eeefec7', + 'md5': '29175c8cadd8b5cc4055001e85d6b372', 'info_dict': { 'id': '382479172', 'ext': 'mp4', 'title': 'MV：Far East Movement《The Illest》', }, - 'skip': 'Only available from China', - } + 'params': { + 'cn_verification_proxy': 'proxy.uku.im:8888' + } + }, { + 'url': 'http://tv.sohu.com/20150305/n409385080.shtml', + 'md5': '699060e75cf58858dd47fb9c03c42cfb', + 'info_dict': { + 'id': '409385080', + 'ext': 'mp4', + 'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》', + } + }, { + 'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml', + 'md5': '9bf34be48f2f4dadcb226c74127e203c', + 'info_dict': { + 'id': '78693464', + 'ext': 'mp4', + 'title': '【爱范品】第31期：MWC见不到的奇葩手机', + } + }, { + 'note': 'Multipart video', + 'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml', + 'info_dict': { + 'id': '78910339', + }, + 'playlist': [{ + 'md5': 'bdbfb8f39924725e6589c146bc1883ad', + 'info_dict': { + 'id': '78910339_part1', + 'ext': 'mp4', + 'duration': 294, + 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆', + } + }, { + 'md5': '3e1f46aaeb95354fd10e7fca9fc1804e', + 'info_dict': { + 'id': '78910339_part2', + 'ext': 'mp4', + 'duration': 300, + 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆', + } + }, { + 'md5': '8407e634175fdac706766481b9443450', + 'info_dict': { + 'id': '78910339_part3', + 'ext': 'mp4', + 'duration': 150, + 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆', + } + }] + }, { + 'info': 'Video with title containing dash', + 'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml', + 'info_dict': { + 'id': '78932792', + 'ext': 'mp4', + 'title': 'youtube-dl testing video', + }, + 'params': { + 'skip_download': True + } + }] def _real_extract(self, url): @@ -29,19 +94,22 @@ class SohuIE(InfoExtractor): else: base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' - return self._download_json( - base_data_url + vid_id, video_id, - 'Downloading JSON data for %s' % vid_id) + req = compat_urllib_request.Request(base_data_url + vid_id) + + cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') + if cn_verification_proxy: + req.add_header('Ytdl-request-proxy', cn_verification_proxy) + + return self._download_json(req, video_id, + 'Downloading JSON data for %s' % vid_id) mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') mytv = mobj.group('mytv') is not None webpage = self._download_webpage(url, video_id) - raw_title = self._html_search_regex( - r'(?s)<title>(.+?)</title>', - webpage, 'video title') - title = raw_title.partition('-')[0].strip() + + title = self._og_search_title(webpage) vid = self._html_search_regex( r'var vid ?= ?["\'](\d+)["\']', @@ -77,7 +145,9 @@ class SohuIE(InfoExtractor): % (format_id, i + 1, part_count)) part_info = part_str.split('|') - video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3]) + + video_url = url_sanitize_consecutive_slashes( + '%s%s?key=%s' % (part_info[0], su[i], part_info[3])) formats.append({ 'url': video_url, diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index c3135effc..e82e3998a 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1804,3 +1804,18 @@ 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) |