From f9355dc989362f31d8e21ccb8fa765546a2360f2 Mon Sep 17 00:00:00 2001 From: PeterDing Date: Thu, 28 May 2015 17:00:09 +0800 Subject: [PATCH 01/13] [youku] update youku --- youtube_dl/extractor/youku.py | 264 ++++++++++++++++++++++------------ 1 file changed, 170 insertions(+), 94 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 97b98bbe8..8d86c3f45 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,123 +1,199 @@ # coding: utf-8 - from __future__ import unicode_literals -import math -import random import re -import time +import base64 from .common import InfoExtractor -from ..utils import ( - ExtractorError, -) - +from ..utils import ExtractorError class YoukuIE(InfoExtractor): + IE_NAME = 'youku' _VALID_URL = r'''(?x) (?: http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)| youku:) (?P[A-Za-z0-9]+)(?:\.html|/v\.swf|) ''' + _TEST = { - 'url': 'http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html', - 'md5': 'ffe3f2e435663dc2d1eea34faeff5b5b', - 'params': { - 'test': False - }, - 'info_dict': { - 'id': 'XNDgyMDQ2NTQw_part00', - 'ext': 'flv', - 'title': 'youtube-dl test video "\'/\\ä↭𝕐' - } + 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', + 'md5': '5f3af4192eabacc4501508d54a8cabd7', + 'info_dict': { + 'id': 'XMTc1ODE5Njcy', + 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', + 'ext': 'flv' + } } - def _gen_sid(self): - nowTime = int(time.time() * 1000) - random1 = random.randint(1000, 1998) - random2 = random.randint(1000, 9999) - - return "%d%d%d" % (nowTime, random1, random2) - - def _get_file_ID_mix_string(self, seed): - mixed = [] - source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890") - seed = float(seed) - for i in range(len(source)): - seed = (seed * 211 + 30031) % 65536 - index = math.floor(seed / 65536 * len(source)) - mixed.append(source[int(index)]) - source.remove(source[int(index)]) - # return ''.join(mixed) - return mixed - - def _get_file_id(self, fileId, seed): - mixed = self._get_file_ID_mix_string(seed) - ids = fileId.split('*') - realId = [] - for ch in ids: - if ch: - realId.append(mixed[int(ch)]) - return ''.join(realId) + def construct_video_urls(self, data1, data2): + # get sid, token + def yk_t(s1, s2): + ls = list(range(256)) + t = 0 + for i in range(256): + t = (t + ls[i] + ord(s1[i%len(s1)])) % 256 + ls[i], ls[t] = ls[t], ls[i] + s, x, y = '', 0, 0 + for i in range(len(s2)): + y = (y + 1) % 256 + x = (x + ls[y]) % 256 + ls[x], ls[y] = ls[y], ls[x] + s += chr((s2[i] ^ ls[(ls[x]+ls[y]) % 256])) + return s + + sid, token = yk_t( + 'becaf9be', base64.b64decode(bytes(data2['ep'], 'ascii')) + ).split('_') + + # get oip + oip = data2['ip'] + + # get fileid + string_ls = list( + 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890') + shuffled_string_ls = [] + seed = data1['seed'] + 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 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:] + + def get_fileid(format, n): + fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) + return fileid + + # get ep + def generate_ep(format, n): + fileid = get_fileid(format, n) + ep_t = yk_t( + 'bf7e5f01', + bytes('%s_%s_%s' % (sid, fileid, token), 'ascii')) + ep = base64.b64encode(bytes(ep_t, 'latin')).decode() + ep = ep.replace('+', '%2B') + ep = ep.replace('/', '%2F') + ep = ep.replace('=', '%2D') + return ep + + # generate video_urls + video_urls_dict = {} + for format in data1['streamtypes']: + video_urls = [] + for dt in data1['segs'][format]: + n = str(int(dt['no'])) + video_url = \ + 'http://k.youku.com/player/getFlvPath/' + \ + 'sid/' + sid + \ + '_' + str(int(n)+1).zfill(2) + \ + '/st/' + self.parse_ext_l(format) + \ + '/fileid/' + get_fileid(format, n) + '?' + \ + 'K=' + str(dt['k']) + \ + '&hd=' + self.get_hd(format) + \ + '&myp=0' + \ + '&ts=' + str(dt['seconds']) + \ + '&ypp=0&ctype=12&ev=1' + \ + '&token=' + str(token) + \ + '&oip=' + str(oip) + \ + '&ep=' + generate_ep(format, n) + video_urls.append(video_url) + video_urls_dict[format] = video_urls + + return video_urls_dict + + def get_hd(self, fm): + hd_id_dict = { + 'flv': '0', + 'mp4': '1', + 'hd2': '2', + 'hd3': '3', + '3gp': '0', + '3gphd': '1' + } + return hd_id_dict[fm] + + def parse_ext_l(self, fm): + ext_dict = { + 'flv': 'flv', + 'mp4': 'mp4', + 'hd2': 'flv', + 'hd3': 'flv', + '3gp': 'flv', + '3gphd': 'mp4', + } + return ext_dict[fm] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id + # request basic data + data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id + data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id - config = self._download_json(info_url, video_id) + raw_data1 = self._download_json(data1_url, video_id) + raw_data2 = self._download_json(data2_url, video_id) + data1 = raw_data1['data'][0] + data2 = raw_data2['data'][0] - error_code = config['data'][0].get('error_code') + error_code = data1.get('error_code') if error_code: # -8 means blocked outside China. - error = config['data'][0].get('error') # Chinese and English, separated by newline. - raise ExtractorError(error or 'Server reported error %i' % error_code, - expected=True) - - video_title = config['data'][0]['title'] - seed = config['data'][0]['seed'] - - format = self._downloader.params.get('format', None) - supported_format = list(config['data'][0]['streamfileids'].keys()) - - # TODO proper format selection - if format is None or format == 'best': - if 'hd2' in supported_format: - format = 'hd2' - else: - format = 'flv' - ext = 'flv' - elif format == 'worst': - format = 'mp4' - ext = 'mp4' - else: - format = 'flv' - ext = 'flv' - - fileid = config['data'][0]['streamfileids'][format] - keys = [s['k'] for s in config['data'][0]['segs'][format]] - # segs is usually a dictionary, but an empty *list* if an error occured. - - files_info = [] - sid = self._gen_sid() - fileid = self._get_file_id(fileid, seed) - - # column 8,9 of fileid represent the segment number - # fileid[7:9] should be changed - for index, key in enumerate(keys): - temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:]) - download_url = 'http://k.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key) - + # Chinese and English, separated by newline. + error = data1.get('error') + raise ExtractorError( + error or 'Server reported error %i' % + error_code, + expected=True) + + title = data1['title'] + + # generate video_urls_dict + video_urls_dict = self.construct_video_urls(data1, data2) + + # construct info + entries = [] + for fm in data1['streamtypes']: + #formats = [] + video_urls = video_urls_dict[fm] + for i in range(len(video_urls)): + if len(entries) < i+1: + entries.append({'formats': []}) + entries[i]['formats'].append( + { + 'url': video_urls[i], + 'format_id': fm, + 'ext': self.parse_ext_l(fm), + 'filesize': int(data1['segs'][fm][i]['size']) + } + ) + + for i in range(len(entries)): + entries[i].update( + { + 'id': '_part%d' % (i+1), + 'title': title, + } + ) + + if len(entries) > 1: info = { - 'id': '%s_part%02d' % (video_id, index), - 'url': download_url, - 'uploader': None, - 'upload_date': None, - 'title': video_title, - 'ext': ext, + '_type': 'multi_video', + 'id': video_id, + 'title': title, + 'entries': entries, } - files_info.append(info) + else: + info = entries[0] + info['id'] = video_id - return files_info + return info From ca45246627f5a67a7c82cd40a11e5c4ff5f68871 Mon Sep 17 00:00:00 2001 From: PeterDing Date: Thu, 28 May 2015 21:04:58 +0800 Subject: [PATCH 02/13] [youku] compatible for python > 3.3 or > 2.7 --- youtube_dl/extractor/youku.py | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 8d86c3f45..7a07c8a5f 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,6 +1,8 @@ # coding: utf-8 from __future__ import unicode_literals +import sys +pyvs = sys.version_info[0] import re import base64 @@ -34,16 +36,23 @@ class YoukuIE(InfoExtractor): for i in range(256): t = (t + ls[i] + ord(s1[i%len(s1)])) % 256 ls[i], ls[t] = ls[t], ls[i] - s, x, y = '', 0, 0 + s = '' if pyvs == 3 else b'' + x, y = 0, 0 for i in range(len(s2)): y = (y + 1) % 256 x = (x + ls[y]) % 256 ls[x], ls[y] = ls[y], ls[x] - s += chr((s2[i] ^ ls[(ls[x]+ls[y]) % 256])) + if isinstance(s2[i], int): + s += chr(s2[i] ^ ls[(ls[x]+ls[y]) % 256]) + else: + s += chr(ord(s2[i]) ^ ls[(ls[x]+ls[y]) % 256]) return s sid, token = yk_t( - 'becaf9be', base64.b64decode(bytes(data2['ep'], 'ascii')) + 'becaf9be', + base64.b64decode(bytes(data2['ep'], 'ascii')) \ + if pyvs == 3 \ + else base64.b64decode(data2['ep']) ).split('_') # get oip @@ -78,8 +87,15 @@ class YoukuIE(InfoExtractor): fileid = get_fileid(format, n) ep_t = yk_t( 'bf7e5f01', - bytes('%s_%s_%s' % (sid, fileid, token), 'ascii')) - ep = base64.b64encode(bytes(ep_t, 'latin')).decode() + bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') \ + if pyvs == 3 \ + else ('%s_%s_%s' % (sid, fileid, token)) + ) + ep = base64.b64encode( + bytes(ep_t, 'latin') \ + if pyvs == 3 \ + else ep_t + ).decode() ep = ep.replace('+', '%2B') ep = ep.replace('/', '%2F') ep = ep.replace('=', '%2D') From 1498940b10a3f43490c05045ebe7a517267a2bff Mon Sep 17 00:00:00 2001 From: PeterDing Date: Fri, 29 May 2015 10:13:09 +0800 Subject: [PATCH 03/13] [youku] compare bytes and str for compatible; use compat_urllib_parse for making video_url --- youtube_dl/extractor/youku.py | 38 +++++++++++++++++++---------------- 1 file changed, 21 insertions(+), 17 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 7a07c8a5f..063f2e10e 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,14 +1,16 @@ # coding: utf-8 from __future__ import unicode_literals -import sys -pyvs = sys.version_info[0] import re import base64 from .common import InfoExtractor from ..utils import ExtractorError +from ..compat import compat_urllib_parse + +bytes_is_str = (bytes == str) # for compatible + class YoukuIE(InfoExtractor): IE_NAME = 'youku' _VALID_URL = r'''(?x) @@ -36,7 +38,7 @@ class YoukuIE(InfoExtractor): for i in range(256): t = (t + ls[i] + ord(s1[i%len(s1)])) % 256 ls[i], ls[t] = ls[t], ls[i] - s = '' if pyvs == 3 else b'' + s = '' if not bytes_is_str else b'' x, y = 0, 0 for i in range(len(s2)): y = (y + 1) % 256 @@ -51,7 +53,7 @@ class YoukuIE(InfoExtractor): sid, token = yk_t( 'becaf9be', base64.b64decode(bytes(data2['ep'], 'ascii')) \ - if pyvs == 3 \ + if not bytes_is_str \ else base64.b64decode(data2['ep']) ).split('_') @@ -88,17 +90,14 @@ class YoukuIE(InfoExtractor): ep_t = yk_t( 'bf7e5f01', bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') \ - if pyvs == 3 \ + if not bytes_is_str \ else ('%s_%s_%s' % (sid, fileid, token)) ) ep = base64.b64encode( bytes(ep_t, 'latin') \ - if pyvs == 3 \ + if not bytes_is_str \ else ep_t ).decode() - ep = ep.replace('+', '%2B') - ep = ep.replace('/', '%2F') - ep = ep.replace('=', '%2D') return ep # generate video_urls @@ -107,20 +106,25 @@ class YoukuIE(InfoExtractor): video_urls = [] for dt in data1['segs'][format]: n = str(int(dt['no'])) + param = { + 'K': dt['k'], + 'hd': self.get_hd(format), + 'myp': 0, + 'ts': dt['seconds'], + 'ypp': 0, + 'ctype': 12, + 'ev': 1, + 'token': token, + 'oip': oip, + 'ep': generate_ep(format, n) + } video_url = \ 'http://k.youku.com/player/getFlvPath/' + \ 'sid/' + sid + \ '_' + str(int(n)+1).zfill(2) + \ '/st/' + self.parse_ext_l(format) + \ '/fileid/' + get_fileid(format, n) + '?' + \ - 'K=' + str(dt['k']) + \ - '&hd=' + self.get_hd(format) + \ - '&myp=0' + \ - '&ts=' + str(dt['seconds']) + \ - '&ypp=0&ctype=12&ev=1' + \ - '&token=' + str(token) + \ - '&oip=' + str(oip) + \ - '&ep=' + generate_ep(format, n) + compat_urllib_parse.urlencode(param) video_urls.append(video_url) video_urls_dict[format] = video_urls From 08f7db20c16743a2bd3040eb7dac11d675011eef Mon Sep 17 00:00:00 2001 From: PeterDing Date: Sat, 30 May 2015 10:03:32 +0800 Subject: [PATCH 04/13] [youku] change format_id --- youtube_dl/extractor/youku.py | 37 +++++++++++++++++++++++------------ 1 file changed, 24 insertions(+), 13 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 063f2e10e..aed6b960a 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -132,26 +132,37 @@ class YoukuIE(InfoExtractor): def get_hd(self, fm): hd_id_dict = { - 'flv': '0', - 'mp4': '1', - 'hd2': '2', - 'hd3': '3', - '3gp': '0', - '3gphd': '1' + 'flv' : '0', + 'mp4' : '1', + 'hd2' : '2', + 'hd3' : '3', + '3gp' : '0', + '3gphd' : '1' } return hd_id_dict[fm] def parse_ext_l(self, fm): ext_dict = { - 'flv': 'flv', - 'mp4': 'mp4', - 'hd2': 'flv', - 'hd3': 'flv', - '3gp': 'flv', - '3gphd': 'mp4', + 'flv' : 'flv', + 'mp4' : 'mp4', + 'hd2' : 'flv', + 'hd3' : 'flv', + '3gp' : 'flv', + '3gphd' : 'mp4' } return ext_dict[fm] + def get_format_name(self, fm): + _dict = { + '3gp' : 'h6', + '3gphd' : 'h5', + 'flv' : 'h4', + 'mp4' : 'h3', + 'hd2' : 'h2', + 'hd3' : 'h1' + } + return _dict[fm] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') @@ -191,7 +202,7 @@ class YoukuIE(InfoExtractor): entries[i]['formats'].append( { 'url': video_urls[i], - 'format_id': fm, + 'format_id': self.get_format_name(fm), 'ext': self.parse_ext_l(fm), 'filesize': int(data1['segs'][fm][i]['size']) } From aed473ccf9d9da73b1b80ee8b06d00ee66a3769d Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 22:41:24 +0800 Subject: [PATCH 05/13] [youku] PEP8 --- youtube_dl/extractor/youku.py | 80 +++++++++++++++++------------------ 1 file changed, 40 insertions(+), 40 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index aed6b960a..4e47fca8a 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -11,6 +11,7 @@ from ..compat import compat_urllib_parse bytes_is_str = (bytes == str) # for compatible + class YoukuIE(InfoExtractor): IE_NAME = 'youku' _VALID_URL = r'''(?x) @@ -21,13 +22,13 @@ class YoukuIE(InfoExtractor): ''' _TEST = { - 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', - 'md5': '5f3af4192eabacc4501508d54a8cabd7', - 'info_dict': { - 'id': 'XMTc1ODE5Njcy', - 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', - 'ext': 'flv' - } + 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', + 'md5': '5f3af4192eabacc4501508d54a8cabd7', + 'info_dict': { + 'id': 'XMTc1ODE5Njcy', + 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', + 'ext': 'flv' + } } def construct_video_urls(self, data1, data2): @@ -36,7 +37,7 @@ class YoukuIE(InfoExtractor): ls = list(range(256)) t = 0 for i in range(256): - t = (t + ls[i] + ord(s1[i%len(s1)])) % 256 + t = (t + ls[i] + ord(s1[i % len(s1)])) % 256 ls[i], ls[t] = ls[t], ls[i] s = '' if not bytes_is_str else b'' x, y = 0, 0 @@ -45,16 +46,16 @@ class YoukuIE(InfoExtractor): x = (x + ls[y]) % 256 ls[x], ls[y] = ls[y], ls[x] if isinstance(s2[i], int): - s += chr(s2[i] ^ ls[(ls[x]+ls[y]) % 256]) + s += chr(s2[i] ^ ls[(ls[x] + ls[y]) % 256]) else: - s += chr(ord(s2[i]) ^ ls[(ls[x]+ls[y]) % 256]) + s += chr(ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256]) return s sid, token = yk_t( 'becaf9be', - base64.b64decode(bytes(data2['ep'], 'ascii')) \ - if not bytes_is_str \ - else base64.b64decode(data2['ep']) + base64.b64decode(bytes(data2['ep'], 'ascii')) + if not bytes_is_str + else base64.b64decode(data2['ep']) ).split('_') # get oip @@ -89,13 +90,13 @@ class YoukuIE(InfoExtractor): fileid = get_fileid(format, n) ep_t = yk_t( 'bf7e5f01', - bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') \ - if not bytes_is_str \ + bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') + if not bytes_is_str else ('%s_%s_%s' % (sid, fileid, token)) ) ep = base64.b64encode( - bytes(ep_t, 'latin') \ - if not bytes_is_str \ + bytes(ep_t, 'latin') + if not bytes_is_str else ep_t ).decode() return ep @@ -121,9 +122,9 @@ class YoukuIE(InfoExtractor): video_url = \ 'http://k.youku.com/player/getFlvPath/' + \ 'sid/' + sid + \ - '_' + str(int(n)+1).zfill(2) + \ + '_' + str(int(n) + 1).zfill(2) + \ '/st/' + self.parse_ext_l(format) + \ - '/fileid/' + get_fileid(format, n) + '?' + \ + '/fileid/' + get_fileid(format, n) + '?' + \ compat_urllib_parse.urlencode(param) video_urls.append(video_url) video_urls_dict[format] = video_urls @@ -132,34 +133,34 @@ class YoukuIE(InfoExtractor): def get_hd(self, fm): hd_id_dict = { - 'flv' : '0', - 'mp4' : '1', - 'hd2' : '2', - 'hd3' : '3', - '3gp' : '0', - '3gphd' : '1' + 'flv': '0', + 'mp4': '1', + 'hd2': '2', + 'hd3': '3', + '3gp': '0', + '3gphd': '1' } return hd_id_dict[fm] def parse_ext_l(self, fm): ext_dict = { - 'flv' : 'flv', - 'mp4' : 'mp4', - 'hd2' : 'flv', - 'hd3' : 'flv', - '3gp' : 'flv', - '3gphd' : 'mp4' + 'flv': 'flv', + 'mp4': 'mp4', + 'hd2': 'flv', + 'hd3': 'flv', + '3gp': 'flv', + '3gphd': 'mp4' } return ext_dict[fm] def get_format_name(self, fm): _dict = { - '3gp' : 'h6', - '3gphd' : 'h5', - 'flv' : 'h4', - 'mp4' : 'h3', - 'hd2' : 'h2', - 'hd3' : 'h1' + '3gp': 'h6', + '3gphd': 'h5', + 'flv': 'h4', + 'mp4': 'h3', + 'hd2': 'h2', + 'hd3': 'h1' } return _dict[fm] @@ -194,10 +195,9 @@ class YoukuIE(InfoExtractor): # construct info entries = [] for fm in data1['streamtypes']: - #formats = [] video_urls = video_urls_dict[fm] for i in range(len(video_urls)): - if len(entries) < i+1: + if len(entries) < i + 1: entries.append({'formats': []}) entries[i]['formats'].append( { @@ -211,7 +211,7 @@ class YoukuIE(InfoExtractor): for i in range(len(entries)): entries[i].update( { - 'id': '_part%d' % (i+1), + 'id': '_part%d' % (i + 1), 'title': title, } ) From c203be3fb4f00388c81564dc0c85ff8a10ff4553 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 23:28:59 +0800 Subject: [PATCH 06/13] [youku] Better handling for Python 2/3 compatibility --- youtube_dl/extractor/youku.py | 37 ++++++++++++----------------------- 1 file changed, 13 insertions(+), 24 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 4e47fca8a..26e5baadc 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -7,9 +7,10 @@ import base64 from .common import InfoExtractor from ..utils import ExtractorError -from ..compat import compat_urllib_parse - -bytes_is_str = (bytes == str) # for compatible +from ..compat import ( + compat_urllib_parse, + compat_ord, +) class YoukuIE(InfoExtractor): @@ -37,26 +38,20 @@ class YoukuIE(InfoExtractor): ls = list(range(256)) t = 0 for i in range(256): - t = (t + ls[i] + ord(s1[i % len(s1)])) % 256 + t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256 ls[i], ls[t] = ls[t], ls[i] - s = '' if not bytes_is_str else b'' + s = bytearray() x, y = 0, 0 for i in range(len(s2)): y = (y + 1) % 256 x = (x + ls[y]) % 256 ls[x], ls[y] = ls[y], ls[x] - if isinstance(s2[i], int): - s += chr(s2[i] ^ ls[(ls[x] + ls[y]) % 256]) - else: - s += chr(ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256]) - return s + s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256]) + return bytes(s) sid, token = yk_t( - 'becaf9be', - base64.b64decode(bytes(data2['ep'], 'ascii')) - if not bytes_is_str - else base64.b64decode(data2['ep']) - ).split('_') + b'becaf9be', base64.b64decode(data2['ep'].encode('ascii')) + ).decode('ascii').split('_') # get oip oip = data2['ip'] @@ -89,16 +84,10 @@ class YoukuIE(InfoExtractor): def generate_ep(format, n): fileid = get_fileid(format, n) ep_t = yk_t( - 'bf7e5f01', - bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') - if not bytes_is_str - else ('%s_%s_%s' % (sid, fileid, token)) + b'bf7e5f01', + ('%s_%s_%s' % (sid, fileid, token)).encode('ascii') ) - ep = base64.b64encode( - bytes(ep_t, 'latin') - if not bytes_is_str - else ep_t - ).decode() + ep = base64.b64encode(ep_t).decode('ascii') return ep # generate video_urls From 9383e66f9475eca0e64c09972c1392d92d17570c Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 23:31:30 +0800 Subject: [PATCH 07/13] [youku] Use _match_id --- youtube_dl/extractor/youku.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 26e5baadc..e41b48369 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -1,7 +1,6 @@ # coding: utf-8 from __future__ import unicode_literals -import re import base64 from .common import InfoExtractor @@ -154,8 +153,7 @@ class YoukuIE(InfoExtractor): return _dict[fm] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) # request basic data data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id From ee69799262e8344742b9d8b492fe792b4d586f6a Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 23:36:28 +0800 Subject: [PATCH 08/13] [youku] Add a v.swf test case --- youtube_dl/extractor/youku.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index e41b48369..d8162a0c5 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -21,7 +21,7 @@ class YoukuIE(InfoExtractor): (?P[A-Za-z0-9]+)(?:\.html|/v\.swf|) ''' - _TEST = { + _TESTS = [{ 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', 'md5': '5f3af4192eabacc4501508d54a8cabd7', 'info_dict': { @@ -29,7 +29,10 @@ class YoukuIE(InfoExtractor): 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', 'ext': 'flv' } - } + }, { + 'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf', + 'only_matching': True, + }] def construct_video_urls(self, data1, data2): # get sid, token From f1e66cb2eb40b48c6508acbe57207a2d99792bf0 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 23:46:07 +0800 Subject: [PATCH 09/13] [youku] Change video_id and add a multipart test case --- youtube_dl/extractor/youku.py | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index d8162a0c5..d5b73ebce 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -25,13 +25,20 @@ class YoukuIE(InfoExtractor): 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', 'md5': '5f3af4192eabacc4501508d54a8cabd7', 'info_dict': { - 'id': 'XMTc1ODE5Njcy', + 'id': 'XMTc1ODE5Njcy_part1', 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', 'ext': 'flv' } }, { 'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf', 'only_matching': True, + }, { + 'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html', + 'info_dict': { + 'id': 'XODgxNjg1Mzk2', + 'title': '武媚娘传奇 85', + }, + 'playlist_count': 11, }] def construct_video_urls(self, data1, data2): @@ -201,20 +208,14 @@ class YoukuIE(InfoExtractor): for i in range(len(entries)): entries[i].update( { - 'id': '_part%d' % (i + 1), + 'id': '%s_part%d' % (video_id, i + 1), 'title': title, } ) - if len(entries) > 1: - info = { - '_type': 'multi_video', - 'id': video_id, - 'title': title, - 'entries': entries, - } - else: - info = entries[0] - info['id'] = video_id - - return info + return { + '_type': 'multi_video', + 'id': video_id, + 'title': title, + 'entries': entries, + } From 04e7596680bce28beae2436bac0f6d1f01a45210 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Mon, 15 Jun 2015 23:54:55 +0800 Subject: [PATCH 10/13] [youku] Better error handling --- youtube_dl/extractor/youku.py | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index d5b73ebce..91f9f6bff 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -176,13 +176,15 @@ class YoukuIE(InfoExtractor): error_code = data1.get('error_code') if error_code: - # -8 means blocked outside China. - # Chinese and English, separated by newline. error = data1.get('error') - raise ExtractorError( - error or 'Server reported error %i' % - error_code, - expected=True) + if error is not None and '因版权原因无法观看此视频' in error: + raise ExtractorError( + 'Youku said: Sorry, this video is available in China only', expected=True) + else: + msg = 'Youku server reported error %i' % error_code + if error is not None: + msg += ': ' + error + raise ExtractorError(msg) title = data1['title'] From 5228b756af2c2bfc2962a5b1bb6db1e6a41c9e05 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 16 Jun 2015 00:06:23 +0800 Subject: [PATCH 11/13] [youku] Add cn_verification_proxy support and add a georestricted test case --- youtube_dl/extractor/youku.py | 32 +++++++++++++++++++++++++------- 1 file changed, 25 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 91f9f6bff..ea37dc8b2 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -9,6 +9,7 @@ from ..utils import ExtractorError from ..compat import ( compat_urllib_parse, compat_ord, + compat_urllib_request, ) @@ -39,6 +40,14 @@ class YoukuIE(InfoExtractor): 'title': '武媚娘传奇 85', }, 'playlist_count': 11, + }, { + 'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html', + 'info_dict': { + 'id': 'XMTI1OTczNDM5Mg', + 'title': '花千骨 04', + }, + 'playlist_count': 13, + 'skip': 'Available in China only', }] def construct_video_urls(self, data1, data2): @@ -165,14 +174,23 @@ class YoukuIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - # request basic data - data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id - data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id + def retrieve_data(req_url, note): + req = compat_urllib_request.Request(req_url) - raw_data1 = self._download_json(data1_url, video_id) - raw_data2 = self._download_json(data2_url, video_id) - data1 = raw_data1['data'][0] - data2 = raw_data2['data'][0] + 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] + + # request basic data + data1 = retrieve_data( + 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id, + 'Downloading JSON metadata 1') + data2 = retrieve_data( + 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id, + 'Downloading JSON metadata 2') error_code = data1.get('error_code') if error_code: From a155b7e76c5a71c650f62c4716d23a24943fc373 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 16 Jun 2015 00:15:09 +0800 Subject: [PATCH 12/13] [youku] Coding style --- youtube_dl/extractor/youku.py | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index ea37dc8b2..cab5be3a4 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -216,22 +216,18 @@ class YoukuIE(InfoExtractor): for i in range(len(video_urls)): if len(entries) < i + 1: entries.append({'formats': []}) - entries[i]['formats'].append( - { - 'url': video_urls[i], - 'format_id': self.get_format_name(fm), - 'ext': self.parse_ext_l(fm), - 'filesize': int(data1['segs'][fm][i]['size']) - } - ) + entries[i]['formats'].append({ + 'url': video_urls[i], + 'format_id': self.get_format_name(fm), + 'ext': self.parse_ext_l(fm), + 'filesize': int(data1['segs'][fm][i]['size']) + }) for i in range(len(entries)): - entries[i].update( - { - 'id': '%s_part%d' % (video_id, i + 1), - 'title': title, - } - ) + entries[i].update({ + 'id': '%s_part%d' % (video_id, i + 1), + 'title': title, + }) return { '_type': 'multi_video', From 0501bfa159db5b5e8ed7fd1ed966b9989becb3e9 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 16 Jun 2015 00:15:30 +0800 Subject: [PATCH 13/13] [YoutubeDL] Youku extractor now uses the standard format selection --- youtube_dl/YoutubeDL.py | 6 ------ 1 file changed, 6 deletions(-) diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index aacec2958..6e4b6f566 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -1033,12 +1033,6 @@ class YoutubeDL(object): info_dict['id'], info_dict.get('subtitles'), info_dict.get('automatic_captions')) - # This extractors handle format selection themselves - if info_dict['extractor'] in ['Youku']: - if download: - self.process_info(info_dict) - return info_dict - # We now pick which formats have to be downloaded if info_dict.get('formats') is None: # There's only one format available