From 51094b1b0885462c19cc2c75f545e621f8a3aa0a Mon Sep 17 00:00:00 2001 From: Celthi Date: Thu, 10 Dec 2015 21:42:12 +0800 Subject: [PATCH 1/7] add cookie and referer in headers, change the video url --- youtube_dl/extractor/youku.py | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 69ecc837a..d077e03db 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -6,6 +6,9 @@ import base64 from .common import InfoExtractor from ..compat import ( compat_urllib_parse, + compat_cookiejar, + compat_cookies, + compat_urllib_request, compat_ord, ) from ..utils import ( @@ -188,19 +191,28 @@ class YoukuIE(InfoExtractor): video_id = self._match_id(url) def retrieve_data(req_url, note): - req = sanitized_Request(req_url) + + + headers = { + 'Referer': req_url, + } + self._set_cookie('youku.com','xreferrer','http://www.youku.com') + req = sanitized_Request(req_url,headers=headers) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: req.add_header('Ytdl-request-proxy', cn_verification_proxy) raw_data = self._download_json(req, video_id, note=note) - return raw_data['data'][0] + + return raw_data['data']['security'] + video_password = self._downloader.params.get('videopassword', None) # request basic data - basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id + #basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id + basic_data_url = "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id if video_password: basic_data_url += '?password=%s' % video_password @@ -208,7 +220,8 @@ class YoukuIE(InfoExtractor): basic_data_url, 'Downloading JSON metadata 1') data2 = retrieve_data( - 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id, + #'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id, + "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id, 'Downloading JSON metadata 2') error_code = data1.get('error_code') From fdf01663d1fd362d9886dbe2e4136bca23cc1543 Mon Sep 17 00:00:00 2001 From: Celthi Date: Fri, 11 Dec 2015 17:48:40 +0800 Subject: [PATCH 2/7] able to download first part of the video, but fail in the left part --- youtube_dl/extractor/youku.py | 57 +++++++++++++++++++++-------------- 1 file changed, 35 insertions(+), 22 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index d077e03db..155286fc5 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,6 +1,7 @@ # coding: utf-8 from __future__ import unicode_literals +import json import base64 from .common import InfoExtractor @@ -84,17 +85,18 @@ class YoukuIE(InfoExtractor): return bytes(s) sid, token = yk_t( - b'becaf9be', base64.b64decode(data2['ep'].encode('ascii')) + b'becaf9be', base64.b64decode(data2['security']['encrypt_string'].encode('ascii')) ).decode('ascii').split('_') # get oip - oip = data2['ip'] + oip = data2['security']['ip'] # get fileid string_ls = list( 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890') shuffled_string_ls = [] - seed = data1['seed'] + seed = data1.get('seed') + seed = 0 N = len(string_ls) for ii in range(N): seed = (seed * 0xd3 + 0x754f) % 0x10000 @@ -103,15 +105,19 @@ class YoukuIE(InfoExtractor): del string_ls[idx] fileid_dict = {} - for format in data1['streamtypes']: - streamfileid = [ - int(i) for i in data1['streamfileids'][format].strip('*').split('*')] - fileid = ''.join( - [shuffled_string_ls[i] for i in streamfileid]) - fileid_dict[format] = fileid[:8] + '%s' + fileid[10:] + for stream in data1['stream']: + format = stream.get('stream_type') + #streamfileid = [ + # int(i) for i in data1['stream']['streamfileids']#[format].strip('*').split('*')] + #fileid = ''.join( + # [shuffled_string_ls[i] for i in streamfileid]) + fileid = stream['stream_fileid'] + fileid_dict[format] = fileid + #fileid_dict[format] = fileid[:8] + '%s' + fileid[10:] def get_fileid(format, n): - fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) + #fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) + fileid = fileid_dict[format] return fileid # get ep @@ -126,15 +132,16 @@ class YoukuIE(InfoExtractor): # generate video_urls video_urls_dict = {} - for format in data1['streamtypes']: + for stream in data1['stream']: + format = stream.get('stream_type') video_urls = [] - for dt in data1['segs'][format]: - n = str(int(dt['no'])) + for dt in stream['segs']: + n = str(int(dt['size'])) param = { - 'K': dt['k'], + 'K': dt['key'], 'hd': self.get_hd(format), 'myp': 0, - 'ts': dt['seconds'], + 'ts': dt['total_milliseconds_video'], 'ypp': 0, 'ctype': 12, 'ev': 1, @@ -161,7 +168,8 @@ class YoukuIE(InfoExtractor): 'hd2': '2', 'hd3': '3', '3gp': '0', - '3gphd': '1' + '3gphd': '1', + 'flvhd': '0' } return hd_id_dict[fm] @@ -172,7 +180,8 @@ class YoukuIE(InfoExtractor): 'hd2': 'flv', 'hd3': 'flv', '3gp': 'flv', - '3gphd': 'mp4' + '3gphd': 'mp4', + 'flvhd': 'flv' } return ext_dict[fm] @@ -180,6 +189,7 @@ class YoukuIE(InfoExtractor): _dict = { '3gp': 'h6', '3gphd': 'h5', + 'flvhd': 'h4', 'flv': 'h4', 'mp4': 'h3', 'hd2': 'h2', @@ -204,8 +214,9 @@ class YoukuIE(InfoExtractor): req.add_header('Ytdl-request-proxy', cn_verification_proxy) raw_data = self._download_json(req, video_id, note=note) + jsonDumpIn = json.dumps(raw_data,indent = 1) - return raw_data['data']['security'] + return raw_data['data'] video_password = self._downloader.params.get('videopassword', None) @@ -236,7 +247,8 @@ class YoukuIE(InfoExtractor): msg += ': ' + error raise ExtractorError(msg) - title = data1['title'] + #title = data1['title'] + title = data1['video']['title'] # generate video_urls_dict video_urls_dict = self.construct_video_urls(data1, data2) @@ -248,10 +260,11 @@ class YoukuIE(InfoExtractor): 'formats': [], # some formats are not available for all parts, we have to detect # which one has all - } for i in range(max(len(v) for v in data1['segs'].values()))] - for fm in data1['streamtypes']: + } for i in range(max(len(v) for v in data1['stream']))] + for stream in data1['stream']: + fm = stream.get('stream_type') video_urls = video_urls_dict[fm] - for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries): + for video_url, seg, entry in zip(video_urls, stream['segs'], entries): entry['formats'].append({ 'url': video_url, 'format_id': self.get_format_name(fm), From 98c3806b15adfcbd7a24fa67c8702c8a1adb0783 Mon Sep 17 00:00:00 2001 From: Celthi Date: Fri, 11 Dec 2015 19:18:14 +0800 Subject: [PATCH 3/7] fix some not important codesnips --- youtube_dl/extractor/youku.py | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 155286fc5..0e955db8a 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,15 +1,11 @@ # coding: utf-8 from __future__ import unicode_literals -import json import base64 from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_cookiejar, - compat_cookies, - compat_urllib_request, compat_ord, ) from ..utils import ( @@ -89,7 +85,7 @@ class YoukuIE(InfoExtractor): ).decode('ascii').split('_') # get oip - oip = data2['security']['ip'] + oip = data1['security']['ip'] # get fileid string_ls = list( @@ -136,7 +132,8 @@ class YoukuIE(InfoExtractor): format = stream.get('stream_type') video_urls = [] for dt in stream['segs']: - n = str(int(dt['size'])) + #n = str(int(dt['size'])) + n = str(stream['segs'].index(dt)) param = { 'K': dt['key'], 'hd': self.get_hd(format), @@ -177,6 +174,8 @@ class YoukuIE(InfoExtractor): ext_dict = { 'flv': 'flv', 'mp4': 'mp4', + 'mp4hd': 'mp4', + 'mp4hd2': 'mp4', 'hd2': 'flv', 'hd3': 'flv', '3gp': 'flv', @@ -214,7 +213,6 @@ class YoukuIE(InfoExtractor): req.add_header('Ytdl-request-proxy', cn_verification_proxy) raw_data = self._download_json(req, video_id, note=note) - jsonDumpIn = json.dumps(raw_data,indent = 1) return raw_data['data'] @@ -260,7 +258,7 @@ class YoukuIE(InfoExtractor): 'formats': [], # some formats are not available for all parts, we have to detect # which one has all - } for i in range(max(len(v) for v in data1['stream']))] + } for i in range(max(len(v.get('segs')) for v in data1['stream']))] for stream in data1['stream']: fm = stream.get('stream_type') video_urls = video_urls_dict[fm] From 5333842a1dd89320cfc0b15a483dba068e6c6af2 Mon Sep 17 00:00:00 2001 From: Celthi Date: Fri, 11 Dec 2015 20:08:14 +0800 Subject: [PATCH 4/7] According the blog and you-get fixed the issues #7627. --- youtube_dl/extractor/youku.py | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 0e955db8a..430a2bb8d 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -112,8 +112,13 @@ class YoukuIE(InfoExtractor): #fileid_dict[format] = fileid[:8] + '%s' + fileid[10:] def get_fileid(format, n): + number = hex(int(str(n), 10))[2:].upper() + if len(number) == 1: + number = '0' + number + streamfileids = fileid_dict[format] + fileid = streamfileids[0:8] + number + streamfileids[10:] #fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) - fileid = fileid_dict[format] + #fileid = fileid_dict[format] return fileid # get ep @@ -138,7 +143,7 @@ class YoukuIE(InfoExtractor): 'K': dt['key'], 'hd': self.get_hd(format), 'myp': 0, - 'ts': dt['total_milliseconds_video'], + #'ts': dt['total_milliseconds_video'], 'ypp': 0, 'ctype': 12, 'ev': 1, @@ -149,7 +154,7 @@ class YoukuIE(InfoExtractor): video_url = \ 'http://k.youku.com/player/getFlvPath/' + \ 'sid/' + sid + \ - '_' + str(int(n) + 1).zfill(2) + \ + '_00'+ \ '/st/' + self.parse_ext_l(format) + \ '/fileid/' + get_fileid(format, n) + '?' + \ compat_urllib_parse.urlencode(param) From 8696a7fd1396e051a3762ff241d1b4e157b7b08a Mon Sep 17 00:00:00 2001 From: Celthi Date: Sat, 12 Dec 2015 10:44:21 +0800 Subject: [PATCH 5/7] fix the keyerror(mp4hd), todo support download the video protected by password --- youtube_dl/extractor/youku.py | 35 +++++++++++++---------------------- 1 file changed, 13 insertions(+), 22 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 430a2bb8d..9fd0e1e23 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import base64 +import json from .common import InfoExtractor from ..compat import ( @@ -90,26 +91,12 @@ class YoukuIE(InfoExtractor): # get fileid string_ls = list( 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890') - shuffled_string_ls = [] - seed = data1.get('seed') - seed = 0 - N = len(string_ls) - for ii in range(N): - seed = (seed * 0xd3 + 0x754f) % 0x10000 - idx = seed * len(string_ls) // 0x10000 - shuffled_string_ls.append(string_ls[idx]) - del string_ls[idx] fileid_dict = {} for stream in data1['stream']: format = stream.get('stream_type') - #streamfileid = [ - # int(i) for i in data1['stream']['streamfileids']#[format].strip('*').split('*')] - #fileid = ''.join( - # [shuffled_string_ls[i] for i in streamfileid]) fileid = stream['stream_fileid'] fileid_dict[format] = fileid - #fileid_dict[format] = fileid[:8] + '%s' + fileid[10:] def get_fileid(format, n): number = hex(int(str(n), 10))[2:].upper() @@ -117,8 +104,6 @@ class YoukuIE(InfoExtractor): number = '0' + number streamfileids = fileid_dict[format] fileid = streamfileids[0:8] + number + streamfileids[10:] - #fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) - #fileid = fileid_dict[format] return fileid # get ep @@ -171,7 +156,9 @@ class YoukuIE(InfoExtractor): 'hd3': '3', '3gp': '0', '3gphd': '1', - 'flvhd': '0' + 'flvhd': '0', + 'mp4hd': '1', + 'mp4hd2': '1' } return hd_id_dict[fm] @@ -180,7 +167,8 @@ class YoukuIE(InfoExtractor): 'flv': 'flv', 'mp4': 'mp4', 'mp4hd': 'mp4', - 'mp4hd2': 'mp4', + 'mp4hd2': 'flv', + 'mp4hd3': 'flv', 'hd2': 'flv', 'hd3': 'flv', '3gp': 'flv', @@ -197,7 +185,10 @@ class YoukuIE(InfoExtractor): 'flv': 'h4', 'mp4': 'h3', 'hd2': 'h2', - 'hd3': 'h1' + 'hd3': 'h1', + 'mp4hd': 'h3', + 'mp4hd3': 'h4', + 'mp4hd2': 'h4' } return _dict[fm] @@ -218,6 +209,7 @@ class YoukuIE(InfoExtractor): req.add_header('Ytdl-request-proxy', cn_verification_proxy) raw_data = self._download_json(req, video_id, note=note) + js = json.dumps(raw_data) return raw_data['data'] @@ -225,7 +217,6 @@ class YoukuIE(InfoExtractor): video_password = self._downloader.params.get('videopassword', None) # request basic data - #basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id basic_data_url = "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id if video_password: basic_data_url += '?password=%s' % video_password @@ -234,7 +225,6 @@ class YoukuIE(InfoExtractor): basic_data_url, 'Downloading JSON metadata 1') data2 = retrieve_data( - #'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id, "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id, 'Downloading JSON metadata 2') @@ -250,9 +240,10 @@ class YoukuIE(InfoExtractor): msg += ': ' + error raise ExtractorError(msg) - #title = data1['title'] + #get video title title = data1['video']['title'] + # generate video_urls_dict video_urls_dict = self.construct_video_urls(data1, data2) From 14c17cafa18292112e5ea09e23f8598892ab3a0b Mon Sep 17 00:00:00 2001 From: Celthi Date: Sat, 12 Dec 2015 11:21:44 +0800 Subject: [PATCH 6/7] add support to video protected by password --- youtube_dl/extractor/youku.py | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 9fd0e1e23..f1bb482c3 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -219,25 +219,25 @@ class YoukuIE(InfoExtractor): # request basic data basic_data_url = "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id if video_password: - basic_data_url += '?password=%s' % video_password + basic_data_url += '&pwd=%s' % video_password data1 = retrieve_data( basic_data_url, 'Downloading JSON metadata 1') data2 = retrieve_data( - "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id, + basic_data_url, 'Downloading JSON metadata 2') - error_code = data1.get('error_code') - if error_code: - error = data1.get('error') - if error is not None and '因版权原因无法观看此视频' in error: + error = data1.get('error') + if error: + error_note = error.get('note') + if error_note is not None and '因版权原因无法观看此视频' in error_note: raise ExtractorError( 'Youku said: Sorry, this video is available in China only', expected=True) else: - msg = 'Youku server reported error %i' % error_code + msg = 'Youku server reported error %i' % error.get('code') if error is not None: - msg += ': ' + error + msg += ': ' + error_note raise ExtractorError(msg) #get video title From 7e37c394858ac215b02d1c4ea425d20a17bf4655 Mon Sep 17 00:00:00 2001 From: Celthi Date: Sat, 12 Dec 2015 11:26:15 +0800 Subject: [PATCH 7/7] merge data1 and data2 --- youtube_dl/extractor/youku.py | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index f1bb482c3..5110efcb4 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -64,7 +64,7 @@ class YoukuIE(InfoExtractor): }, }] - def construct_video_urls(self, data1, data2): + def construct_video_urls(self, data): # get sid, token def yk_t(s1, s2): ls = list(range(256)) @@ -82,18 +82,18 @@ class YoukuIE(InfoExtractor): return bytes(s) sid, token = yk_t( - b'becaf9be', base64.b64decode(data2['security']['encrypt_string'].encode('ascii')) + b'becaf9be', base64.b64decode(data['security']['encrypt_string'].encode('ascii')) ).decode('ascii').split('_') # get oip - oip = data1['security']['ip'] + oip = data['security']['ip'] # get fileid string_ls = list( 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890') fileid_dict = {} - for stream in data1['stream']: + for stream in data['stream']: format = stream.get('stream_type') fileid = stream['stream_fileid'] fileid_dict[format] = fileid @@ -118,7 +118,7 @@ class YoukuIE(InfoExtractor): # generate video_urls video_urls_dict = {} - for stream in data1['stream']: + for stream in data['stream']: format = stream.get('stream_type') video_urls = [] for dt in stream['segs']: @@ -221,14 +221,11 @@ class YoukuIE(InfoExtractor): if video_password: basic_data_url += '&pwd=%s' % video_password - data1 = retrieve_data( + data = retrieve_data( basic_data_url, 'Downloading JSON metadata 1') - data2 = retrieve_data( - basic_data_url, - 'Downloading JSON metadata 2') - error = data1.get('error') + error = data.get('error') if error: error_note = error.get('note') if error_note is not None and '因版权原因无法观看此视频' in error_note: @@ -241,11 +238,11 @@ class YoukuIE(InfoExtractor): raise ExtractorError(msg) #get video title - title = data1['video']['title'] + title = data['video']['title'] # generate video_urls_dict - video_urls_dict = self.construct_video_urls(data1, data2) + video_urls_dict = self.construct_video_urls(data) # construct info entries = [{ @@ -254,8 +251,8 @@ class YoukuIE(InfoExtractor): 'formats': [], # some formats are not available for all parts, we have to detect # which one has all - } for i in range(max(len(v.get('segs')) for v in data1['stream']))] - for stream in data1['stream']: + } for i in range(max(len(v.get('segs')) for v in data['stream']))] + for stream in data['stream']: fm = stream.get('stream_type') video_urls = video_urls_dict[fm] for video_url, seg, entry in zip(video_urls, stream['segs'], entries):