Merge remote-tracking branch 'upstream/master' into this-american-life

AUTHORS

@@ -127,3 +127,4 @@ Julian Richen
 Ping O.
 Mister Hat
 Peter Ding
+jackyzy823

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

youtube_dl/extractor/francetv.py

@@ -60,7 +60,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
                     continue
                 video_url_parsed = compat_urllib_parse_urlparse(video_url)
                 f4m_url = self._download_webpage(
-                    'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path,
+                    'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url_parsed.path,
                     video_id, 'Downloading f4m manifest token', fatal=False)
                 if f4m_url:
                     formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id))

youtube_dl/extractor/safari.py

@@ -83,7 +83,7 @@ class SafariIE(SafariBaseIE):
                                     library/view/[^/]+|
                                     api/v1/book
                                 )/
-                                (?P<course_id>\d+)/
+                                (?P<course_id>[^/]+)/
                                     (?:chapter(?:-content)?/)?
                                 (?P<part>part\d+)\.html
     '''
@@ -100,6 +100,10 @@ class SafariIE(SafariBaseIE):
     }, {
         'url': 'https://www.safaribooksonline.com/api/v1/book/9780133392838/chapter/part00.html',
         'only_matching': True,
+    }, {
+        # non-digits in course id
+        'url': 'https://www.safaribooksonline.com/library/view/create-a-nodejs/100000006A0210/part00.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -122,7 +126,7 @@ class SafariCourseIE(SafariBaseIE):
     IE_NAME = 'safari:course'
     IE_DESC = 'safaribooksonline.com online courses'
 
-    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/(?:library/view/[^/]+|api/v1/book)/(?P<id>\d+)/?(?:[#?]|$)'
+    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/(?:library/view/[^/]+|api/v1/book)/(?P<id>[^/]+)/?(?:[#?]|$)'
 
     _TESTS = [{
         'url': 'https://www.safaribooksonline.com/library/view/hadoop-fundamentals-livelessons/9780133392838/',

youtube_dl/extractor/vbox7.py

@@ -5,6 +5,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
@@ -26,11 +27,21 @@ class Vbox7IE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        redirect_page, urlh = self._download_webpage_handle(url, video_id)
+        # need to get the page 3 times for the correct jsSecretToken cookie
-        new_location = self._search_regex(r'window\.location = \'(.*)\';',
+        # which is necessary for the correct title
-                                          redirect_page, 'redirect location')
+        def get_session_id():
-        redirect_url = urlh.geturl() + new_location
+            redirect_page = self._download_webpage(url, video_id)
-        webpage = self._download_webpage(redirect_url, video_id,
+            session_id_url = self._search_regex(
+                r'var\s*url\s*=\s*\'([^\']+)\';', redirect_page,
+                'session id url')
+            self._download_webpage(
+                compat_urlparse.urljoin(url, session_id_url), video_id,
+                'Getting session id')
+
+        get_session_id()
+        get_session_id()
+
+        webpage = self._download_webpage(url, video_id,
                                          'Downloading redirect page')
 
         title = self._html_search_regex(r'<title>(.*)</title>',

youtube_dl/extractor/vk.py

@@ -13,6 +13,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     orderedSet,
+    str_to_int,
     unescapeHTML,
     unified_strdate,
 )
@@ -34,6 +35,7 @@ class VKIE(InfoExtractor):
                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
                 'duration': 195,
                 'upload_date': '20120212',
+                'view_count': int,
             },
         },
         {
@@ -45,7 +47,8 @@ class VKIE(InfoExtractor):
                 'uploader': 'Tom Cruise',
                 'title': 'No name',
                 'duration': 9,
-                'upload_date': '20130721'
+                'upload_date': '20130721',
+                'view_count': int,
             }
         },
         {
@@ -59,6 +62,7 @@ class VKIE(InfoExtractor):
                 'title': 'Lin Dan',
                 'duration': 101,
                 'upload_date': '20120730',
+                'view_count': int,
             }
         },
         {
@@ -73,7 +77,8 @@ class VKIE(InfoExtractor):
                 'uploader': 'Триллеры',
                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
                 'duration': 8352,
-                'upload_date': '20121218'
+                'upload_date': '20121218',
+                'view_count': int,
             },
             'skip': 'Requires vk account credentials',
         },
@@ -100,6 +105,7 @@ class VKIE(InfoExtractor):
                 'title': 'Книга Илая',
                 'duration': 6771,
                 'upload_date': '20140626',
+                'view_count': int,
             },
             'skip': 'Only works from Russia',
         },
@@ -175,25 +181,29 @@ class VKIE(InfoExtractor):
                 m_rutube.group(1).replace('\\', ''))
             return self.url_result(rutube_url)
 
-        m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.*?});', info_page)
+        m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
         if m_opts:
-            m_opts_url = re.search(r"url\s*:\s*'([^']+)", m_opts.group(1))
+            m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
             if m_opts_url:
                 opts_url = m_opts_url.group(1)
                 if opts_url.startswith('//'):
                     opts_url = 'http:' + opts_url
                 return self.url_result(opts_url)
 
-        data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars')
+        data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
         data = json.loads(data_json)
 
         # Extract upload date
         upload_date = None
-        mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
+        mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
         if mobj is not None:
             mobj.group(1) + ' ' + mobj.group(2)
             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 
+        view_count = str_to_int(self._search_regex(
+            r'"mv_views_count_number"[^>]*>([\d,.]+) views<',
+            info_page, 'view count', fatal=False))
+
         formats = [{
             'format_id': k,
             'url': v,
@@ -210,6 +220,7 @@ class VKIE(InfoExtractor):
             'uploader': data.get('md_author'),
             'duration': data.get('duration'),
             'upload_date': upload_date,
+            'view_count': view_count,
         }
 
 

youtube_dl/extractor/youku.py

@@ -1,123 +1,237 @@
 # coding: utf-8
-
 from __future__ import unicode_literals
 
-import math
+import base64
-import random
-import re
-import time
 
 from .common import InfoExtractor
-from ..utils import (
+from ..utils import ExtractorError
-    ExtractorError,
+
+from ..compat import (
+    compat_urllib_parse,
+    compat_ord,
+    compat_urllib_request,
 )
 
 
 class YoukuIE(InfoExtractor):
+    IE_NAME = 'youku'
     _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|)
     '''
-    _TEST = {
+
-        'url': 'http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html',
+    _TESTS = [{
-        'md5': 'ffe3f2e435663dc2d1eea34faeff5b5b',
+        'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
-        'params': {
+        'md5': '5f3af4192eabacc4501508d54a8cabd7',
-            'test': False
-        },
         'info_dict': {
-            'id': 'XNDgyMDQ2NTQw_part00',
+            'id': 'XMTc1ODE5Njcy_part1',
-            'ext': 'flv',
+            'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
-            'title': 'youtube-dl test video "\'/\\ä↭𝕐'
+            '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',
+    }]
 
-    def _gen_sid(self):
+    def construct_video_urls(self, data1, data2):
-        nowTime = int(time.time() * 1000)
+        # get sid, token
-        random1 = random.randint(1000, 1998)
+        def yk_t(s1, s2):
-        random2 = random.randint(1000, 9999)
+            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)
 
-        return "%d%d%d" % (nowTime, random1, random2)
+        sid, token = yk_t(
+            b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
+        ).decode('ascii').split('_')
 
-    def _get_file_ID_mix_string(self, seed):
+        # get oip
-        mixed = []
+        oip = data2['ip']
-        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):
+        # get fileid
-        mixed = self._get_file_ID_mix_string(seed)
+        string_ls = list(
-        ids = fileId.split('*')
+            'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
-        realId = []
+        shuffled_string_ls = []
-        for ch in ids:
+        seed = data1['seed']
-            if ch:
+        N = len(string_ls)
-                realId.append(mixed[int(ch)])
+        for ii in range(N):
-        return ''.join(realId)
+            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'
+        }
+        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 = self._match_id(url)
-        video_id = mobj.group('id')
 
-        info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
+        def retrieve_data(req_url, note):
+            req = compat_urllib_request.Request(req_url)
 
-        config = self._download_json(info_url, video_id)
+            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
+            if cn_verification_proxy:
+                req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 
-        error_code = config['data'][0].get('error_code')
+            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:
-            # -8 means blocked outside China.
+            error = data1.get('error')
-            error = config['data'][0].get('error')  # Chinese and English, separated by newline.
+            if error is not None and '因版权原因无法观看此视频' in error:
-            raise ExtractorError(error or 'Server reported error %i' % error_code,
+                raise ExtractorError(
-                                 expected=True)
+                    'Youku said: Sorry, this video is available in China only', 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'
+                msg = 'Youku server reported error %i' % error_code
-            ext = 'flv'
+                if error is not None:
-        elif format == 'worst':
+                    msg += ': ' + error
-            format = 'mp4'
+                raise ExtractorError(msg)
-            ext = 'mp4'
-        else:
-            format = 'flv'
-            ext = 'flv'
 
-        fileid = config['data'][0]['streamfileids'][format]
+        title = data1['title']
-        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 = []
+        # generate video_urls_dict
-        sid = self._gen_sid()
+        video_urls_dict = self.construct_video_urls(data1, data2)
-        fileid = self._get_file_id(fileid, seed)
 
-        # column 8,9 of fileid represent the segment number
+        # construct info
-        # fileid[7:9] should be changed
+        entries = []
-        for index, key in enumerate(keys):
+        for fm in data1['streamtypes']:
-            temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
+            video_urls = video_urls_dict[fm]
-            download_url = 'http://k.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
+            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'])
+                })
 
-            info = {
+        for i in range(len(entries)):
-                'id': '%s_part%02d' % (video_id, index),
+            entries[i].update({
-                'url': download_url,
+                'id': '%s_part%d' % (video_id, i + 1),
-                'uploader': None,
+                'title': title,
-                'upload_date': None,
+            })
-                'title': video_title,
-                'ext': ext,
-            }
-            files_info.append(info)
 
-        return files_info
+        return {
+            '_type': 'multi_video',
+            'id': video_id,
+            'title': title,
+            'entries': entries,
+        }

youtube_dl/extractor/youtube.py

@@ -1504,7 +1504,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
 
         for pagenum in itertools.count(1):
             url_query = {
-                'search_query': query,
+                'search_query': query.encode('utf-8'),
                 'page': pagenum,
                 'spf': 'navigate',
             }