Register new version in the changelog.

[youtubedl] / youtube_dl / extractor / youku.py

diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py
index eb98298019c04334276688a7d9c6a5db8bd90664..2e81d92238307e8914769d6fc48d03befd6af2bf 100644 (file)
--- a/youtube_dl/extractor/youku.py
+++ b/youtube_dl/extractor/youku.py
@@ -1,116 +1,253 @@
 # coding: utf-8
 # coding: utf-8

+from __future__ import unicode_literals

 
 

-import json
-import math
-import random
-import re
-import time
+import base64

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
-    ExtractorError,
+from ..utils import ExtractorError
+
+from ..compat import (
+    compat_urllib_parse,
+    compat_ord,
+    compat_urllib_request,

 )
 
 
 class YoukuIE(InfoExtractor):
 )
 
 
 class YoukuIE(InfoExtractor):

-    _VALID_URL =  r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html'
-    _TEST =   {
-        u"url": u"http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html",
-        u"file": u"XNDgyMDQ2NTQw_part00.flv",
-        u"md5": u"ffe3f2e435663dc2d1eea34faeff5b5b",
-        u"params": { u"test": False },
-        u"info_dict": {
-            u"title": u"youtube-dl test video \"'/\\ä↭𝕐"
+    IE_NAME = 'youku'
+    IE_DESC = '优酷'
+    _VALID_URL = r'''(?x)
+        (?:
+            http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
+            youku:)
+        (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
+    '''
+
+    _TESTS = [{
+        'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
+        'md5': '5f3af4192eabacc4501508d54a8cabd7',
+        'info_dict': {
+            '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,
+    }, {
+        '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',
+    }, {
+        'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
+        'note': 'Video protected with password',
+        'info_dict': {
+            'id': 'XNjA1NzA2Njgw',
+            'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
+        },
+        'playlist_count': 19,
+        'params': {
+            'videopassword': '100600',
+        },
+    }]
+
+    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] + compat_ord(s1[i % len(s1)])) % 256
+                ls[i], ls[t] = ls[t], ls[i]
+            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]
+                s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
+            return bytes(s)
+
+        sid, token = yk_t(
+            b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
+        ).decode('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(
+                b'bf7e5f01',
+                ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
+            )
+            ep = base64.b64encode(ep_t).decode('ascii')
+            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']))
+                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) + '?' + \
+                    compat_urllib_parse.urlencode(param)
+                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'

         }
         }

-    }
-
-
-    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)
+        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):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('ID')
-
-        info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
-
-        jsondata = self._download_webpage(info_url, video_id)
-
-        self.report_extraction(video_id)
-        try:
-            config = json.loads(jsondata)
-
-            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())
-
-            if format is None or format == 'best':
-                if 'hd2' in supported_format:
-                    format = 'hd2'
-                else:
-                    format = 'flv'
-                ext = u'flv'
-            elif format == 'worst':
-                format = 'mp4'
-                ext = u'mp4'
+        video_id = self._match_id(url)
+
+        def retrieve_data(req_url, note):
+            req = compat_urllib_request.Request(req_url)
+
+            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]
+
+        video_password = self._downloader.params.get('videopassword', None)
+
+        # request basic data
+        basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
+        if video_password:
+            basic_data_url += '?password=%s' % video_password
+
+        data1 = retrieve_data(
+            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,
+            '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:
+                raise ExtractorError(
+                    'Youku said: Sorry, this video is available in China only', expected=True)

             else:
             else:

-                format = 'flv'
-                ext = u'flv'
-
-
-            fileid = config['data'][0]['streamfileids'][format]
-            keys = [s['k'] for s in config['data'][0]['segs'][format]]
-        except (UnicodeDecodeError, ValueError, KeyError):
-            raise ExtractorError(u'Unable to extract info section')
-
-        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://f.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
-
-            info = {
-                'id': '%s_part%02d' % (video_id, index),
-                'url': download_url,
-                'uploader': None,
-                'upload_date': None,
-                'title': video_title,
-                'ext': ext,
-            }
-            files_info.append(info)
-
-        return files_info
+                msg = 'Youku server reported error %i' % error_code
+                if error is not None:
+                    msg += ': ' + error
+                raise ExtractorError(msg)
+
+        title = data1['title']
+
+        # generate video_urls_dict
+        video_urls_dict = self.construct_video_urls(data1, data2)
+
+        # construct info
+        entries = [{
+            'id': '%s_part%d' % (video_id, i + 1),
+            'title': title,
+            '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']:
+            video_urls = video_urls_dict[fm]
+            for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries):
+                entry['formats'].append({
+                    'url': video_url,
+                    'format_id': self.get_format_name(fm),
+                    'ext': self.parse_ext_l(fm),
+                    'filesize': int(seg['size']),
+                })
+
+        return {
+            '_type': 'multi_video',
+            'id': video_id,
+            'title': title,
+            'entries': entries,
+        }