[cctv] Improve and merge with cntv (closes #879, closes #6753, closes #8541)

youtube_dl/extractor/cctv.py

@@ -4,27 +4,84 @@
 import re
 
 from .common import InfoExtractor
-from ..utils import float_or_none
+from ..compat import compat_str
+from ..utils import (
+    float_or_none,
+    try_get,
+    unified_timestamp,
+)
 
 
 class CCTVIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://(?:.+?\.)?
+    IE_DESC = '央视网'
-        (?:
+    _VALID_URL = r'https?://(?:[^/]+)\.(?:cntv|cctv)\.(?:com|cn)/(?:[^/]+/)*?(?P<id>[^/?#&]+?)(?:/index)?\.s?html'
-            cctv\.(?:com|cn)|
-            cntv\.cn
-        )/
-        (?:
-            video/[^/]+/(?P<id>[0-9a-f]{32})|
-            \d{4}/\d{2}/\d{2}/(?P<display_id>VID[0-9A-Za-z]+)
-        )'''
     _TESTS = [{
-        'url': 'http://english.cntv.cn/2016/09/03/VIDEhnkB5y9AgHyIEVphCEz1160903.shtml',
+        'url': 'http://sports.cntv.cn/2016/02/12/ARTIaBRxv4rTT1yWf1frW2wi160212.shtml',
-        'md5': '819c7b49fc3927d529fb4cd555621823',
+        'md5': 'd61ec00a493e09da810bf406a078f691',
         'info_dict': {
-            'id': '454368eb19ad44a1925bf1eb96140a61',
+            'id': '5ecdbeab623f4973b40ff25f18b174e8',
             'ext': 'mp4',
-            'title': 'Portrait of Real Current Life 09/03/2016 Modern Inventors Part 1',
+            'title': '[NBA]二少联手砍下46分 雷霆主场击败鹈鹕（快讯）',
-        }
+            'description': 'md5:7e14a5328dc5eb3d1cd6afbbe0574e95',
+            'duration': 98,
+            'uploader': 'songjunjie',
+            'timestamp': 1455279956,
+            'upload_date': '20160212',
+        },
+    }, {
+        'url': 'http://tv.cctv.com/2016/02/05/VIDEUS7apq3lKrHG9Dncm03B160205.shtml',
+        'info_dict': {
+            'id': 'efc5d49e5b3b4ab2b34f3a502b73d3ae',
+            'ext': 'mp4',
+            'title': '[赛车]“车王”舒马赫恢复情况成谜（快讯）',
+            'description': '2月4日，蒙特泽莫罗透露了关于“车王”舒马赫恢复情况，但情况是否属实遭到了质疑。',
+            'duration': 37,
+            'uploader': 'shujun',
+            'timestamp': 1454677291,
+            'upload_date': '20160205',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://english.cntv.cn/special/four_comprehensives/index.shtml',
+        'info_dict': {
+            'id': '4bb9bb4db7a6471ba85fdeda5af0381e',
+            'ext': 'mp4',
+            'title': 'NHnews008 ANNUAL POLITICAL SEASON',
+            'description': 'Four Comprehensives',
+            'duration': 60,
+            'uploader': 'zhangyunlei',
+            'timestamp': 1425385521,
+            'upload_date': '20150303',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://cctv.cntv.cn/lm/tvseries_russian/yilugesanghua/index.shtml',
+        'info_dict': {
+            'id': 'b15f009ff45c43968b9af583fc2e04b2',
+            'ext': 'mp4',
+            'title': 'Путь，усыпанный космеями Серия 1',
+            'description': 'Путь, усыпанный космеями',
+            'duration': 2645,
+            'uploader': 'renxue',
+            'timestamp': 1477479241,
+            'upload_date': '20161026',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://ent.cntv.cn/2016/01/18/ARTIjprSSJH8DryTVr5Bx8Wb160118.shtml',
+        'only_matching': True,
+    }, {
+        'url': 'http://tv.cntv.cn/video/C39296/e0210d949f113ddfb38d31f00a4e5c44',
+        'only_matching': True,
+    }, {
+        'url': 'http://english.cntv.cn/2016/09/03/VIDEhnkB5y9AgHyIEVphCEz1160903.shtml',
+        'only_matching': True,
     }, {
         'url': 'http://tv.cctv.com/2016/09/07/VIDE5C1FnlX5bUywlrjhxXOV160907.shtml',
         'only_matching': True,
@@ -34,20 +91,63 @@ class CCTVIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        video_id, display_id = re.match(self._VALID_URL, url).groups()
+        video_id = self._match_id(url)
-        if not video_id:
+        webpage = self._download_webpage(url, video_id)
-            webpage = self._download_webpage(url, display_id)
+
         video_id = self._search_regex(
-                r'(?:fo\.addVariable\("videoCenterId",\s*|guid\s*=\s*)"([0-9a-f]{32})',
+            [r'var\s+guid\s*=\s*["\']([\da-fA-F]+)',
-                webpage, 'video_id')
+             r'videoCenterId["\']\s*,\s*["\']([\da-fA-F]+)',
-        api_data = self._download_json(
+             r'"changePlayer\s*\(\s*["\']([\da-fA-F]+)',
-            'http://vdn.apps.cntv.cn/api/getHttpVideoInfo.do?pid=' + video_id, video_id)
+             r'"load[Vv]ideo\s*\(\s*["\']([\da-fA-F]+)'],
-        m3u8_url = re.sub(r'maxbr=\d+&?', '', api_data['hls_url'])
+            webpage, 'video id', default=video_id)
+
+        data = self._download_json(
+            'http://vdn.apps.cntv.cn/api/getHttpVideoInfo.do', video_id,
+            query={
+                'pid': video_id,
+                'url': url,
+                'idl': 32,
+                'idlr': 32,
+                'modifyed': 'false',
+            })
+
+        title = data['title']
+
+        formats = []
+
+        video = data.get('video')
+        if isinstance(video, dict):
+            for quality, chapters_key in enumerate(('lowChapters', 'chapters')):
+                video_url = try_get(
+                    video, lambda x: x[chapters_key][0]['url'], compat_str)
+                if video_url:
+                    formats.append({
+                        'url': video_url,
+                        'format_id': 'http',
+                        'quality': quality,
+                        'preference': -1,
+                    })
+
+        hls_url = try_get(data, lambda x: x['hls_url'], compat_str)
+        if hls_url:
+            hls_url = re.sub(r'maxbr=\d+&?', '', hls_url)
+            formats.extend(self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                m3u8_id='hls', fatal=False))
+
+        self._sort_formats(formats)
+
+        uploader = data.get('editer_name')
+        description = self._html_search_meta('description', webpage)
+        timestamp = unified_timestamp(data.get('f_pgmtime'))
+        duration = float_or_none(try_get(video, lambda x: x['totalLength']))
 
         return {
             'id': video_id,
-            'title': api_data['title'],
+            'title': title,
-            'formats': self._extract_m3u8_formats(
+            'description': description,
-                m3u8_url, video_id, 'mp4', 'm3u8_native', fatal=False),
+            'uploader': uploader,
-            'duration': float_or_none(api_data.get('video', {}).get('totalLength')),
+            'timestamp': timestamp,
+            'duration': duration,
+            'formats': formats,
         }

youtube_dl/extractor/cntv.py

@@ -1,123 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_request
-
-class CntvIE(InfoExtractor):
-    IE_NAME = 'cntv'
-    IE_DESC = '央视网'
-    _VALID_URL = r'''(?x)
-        (?:
-            http://(?:ent|tv|sports|english|chunwan)\.(?:cntv|cctv)\.(?:com|cn)/
-            (?:
-                (?:video|special)/(?:[_A-Za-z0-9]+)|(?P<year>\d+)/(?P<month>\d+)/(?P<day>\d+)
-            )
-        )
-        /(?P<id>[A-Za-z0-9]+)(index)?(?:\.shtml|)
-    '''
-
-    _TESTS = [{
-        'url': 'http://sports.cntv.cn/2016/02/12/ARTIaBRxv4rTT1yWf1frW2wi160212.shtml',
-        'md5': 'd0f8c3e043fe43534517ed47c831fd4e',
-        'info_dict': {
-            'id': '5ecdbeab623f4973b40ff25f18b174e8',
-            'ext': 'flv',
-            'title': '[NBA]二少联手砍下46分 雷霆主场击败鹈鹕（快讯）',
-            'description': 'md5:7e14a5328dc5eb3d1cd6afbbe0574e95',
-            'uploader': 'songjunjie'
-        }
-    },{
-        'url': 'http://tv.cctv.com/2016/02/05/VIDEUS7apq3lKrHG9Dncm03B160205.shtml',
-        'md5': '3bde627de65b6bad064f3e67ec9e71a1',
-        'info_dict': {
-            'id': 'efc5d49e5b3b4ab2b34f3a502b73d3ae',
-            'ext': 'flv',
-            'title': '[赛车]“车王”舒马赫恢复情况成谜（快讯）',
-            'description': '2月4日，蒙特泽莫罗透露了关于“车王”舒马赫恢复情况，但情况是否属实遭到了质疑。',
-            'uploader': 'shujun'
-        }
-    },{
-        'url': 'http://english.cntv.cn/special/four_comprehensives/index.shtml',
-        'md5': 'f37a56d5f16647eae24c3e618f8d23a2',
-        'info_dict': {
-            'id': '4bb9bb4db7a6471ba85fdeda5af0381e',
-            'ext': 'flv',
-            'title': 'NHnews008 ANNUAL POLITICAL SEASON',
-            'description': 'Four Comprehensives',
-            'uploader': 'zhangyunlei'
-        }
-    },{
-        'url': 'http://ent.cntv.cn/2016/01/18/ARTIjprSSJH8DryTVr5Bx8Wb160118.shtml',
-        'md5': '37ad2407b5f28336ddf26aacea570ee5',
-        'info_dict': {
-            'id': '45336758dd474ef39dc6887f115e0375',
-            'ext': 'flv',
-            'title': '《爱国之恋》中国梦新歌展播  综艺视频精切',
-            'description': 'md5:856836ae2660f000e458ead597b1e2bd',
-            'uploader': 'baiyuqing'
-        }
-    },{
-        'url': 'http://tv.cntv.cn/video/C39296/e0210d949f113ddfb38d31f00a4e5c44',
-        'md5': 'd7188ecc2a2f8447c70023895724c31c',
-        'info_dict': {
-            'id': 'e0210d949f113ddfb38d31f00a4e5c44',
-            'ext': 'flv',
-            'title': '《传奇故事》 20150409 金钱“魔术”（下）',
-            'description': '本期节目主要内容：     2014年10月23日凌晨，浙江余杭警方出动2000多名警力，一举摧毁1040阳光工程传销组织，抓获65名涉嫌传销的组织头目，教育遣返400多名参与传销人员。',
-            'uploader': '陆银凯'
-        }
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_id_regex = [r'var guid[ ="]+([a-zA-Z0-9]+)"',
-                    r'"videoCenterId"[ ,"]+([a-zA-Z0-9]+)"',
-                    r'"changePlayer\(\'([a-zA-Z0-9]+)\'\)\"']
-        video_id = self._search_regex(video_id_regex, webpage, 'video_id', default=None)
-
-        if video_id is None:
-            video_id = self._match_id(url)
-
-        # refer to youku.py
-        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
-
-        get_info_url = \
-            'http://vdn.apps.cntv.cn/api/getHttpVideoInfo.do' + \
-            '?pid=%s' % video_id + \
-            '&tz=-8' + \
-            '&from=000tv' + \
-            '&url=%s' % url + \
-            '&idl=32' + \
-            '&idlr=32' + \
-            '&modifyed=false'
-
-        data = retrieve_data(
-            get_info_url,
-            'Downloading JSON metadata')
-
-        title = data.get('title')
-        uploader = data.get('editer_name')
-        hls_url = data.get('hls_url')
-        formats = self._extract_m3u8_formats(hls_url, video_id, 'flv')
-        self._sort_formats(formats)
-
-        description = self._html_search_meta('description', webpage)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'uploader': uploader,
-            'formats': formats,
-            'description': description,
-        }

youtube_dl/extractor/extractors.py

@@ -178,7 +178,6 @@
     CNNBlogsIE,
     CNNArticleIE,
 )
-from .cntv import CntvIE
 from .coub import CoubIE
 from .collegerama import CollegeRamaIE
 from .comedycentral import (