Merge pull request #3 from unit193/master

[youtubedl] / youtube_dl / extractor / sportbox.py

diff --git a/youtube_dl/extractor/sportbox.py b/youtube_dl/extractor/sportbox.py
index becdf658f6e0ce8b209dffc0ce4c96a2857099dc..b9017fd2ab67df5f3e1b855eb30047f790b53765 100644 (file)
--- a/youtube_dl/extractor/sportbox.py
+++ b/youtube_dl/extractor/sportbox.py
@@ -5,72 +5,95 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    parse_duration,
-    parse_iso8601,
+    determine_ext,
+    int_or_none,
+    js_to_json,
+    merge_dicts,

 )
 
 
 class SportBoxIE(InfoExtractor):
 )
 
 
 class SportBoxIE(InfoExtractor):

-    _VALID_URL = r'https?://news\.sportbox\.ru/Vidy_sporta/(?:[^/]+/)+spbvideo_NI\d+_(?P<display_id>.+)'
-    _TESTS = [
-        {
-            'url': 'http://news.sportbox.ru/Vidy_sporta/Avtosport/Rossijskij/spbvideo_NI483529_Gonka-2-zaezd-Obyedinenniy-2000-klassi-Turing-i-S',
-            'md5': 'ff56a598c2cf411a9a38a69709e97079',
-            'info_dict': {
-                'id': '80822',
-                'ext': 'mp4',
-                'title': 'Гонка 2  заезд ««Объединенный 2000»: классы Туринг и Супер-продакшн',
-                'description': 'md5:81715fa9c4ea3d9e7915dc8180c778ed',
-                'thumbnail': 're:^https?://.*\.jpg$',
-                'timestamp': 1411896237,
-                'upload_date': '20140928',
-                'duration': 4846,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        }, {
-            'url': 'http://news.sportbox.ru/Vidy_sporta/billiard/spbvideo_NI486287_CHempionat-mira-po-dinamichnoy-piramide-4',
-            'only_matching': True,
-        }
-    ]
+    _VALID_URL = r'https?://(?:news\.sportbox|matchtv)\.ru/vdl/player(?:/[^/]+/|\?.*?\bn?id=)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://news.sportbox.ru/vdl/player/ci/211355',
+        'info_dict': {
+            'id': '109158',
+            'ext': 'mp4',
+            'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
+            'description': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 292,
+            'view_count': int,
+            'timestamp': 1426237001,
+            'upload_date': '20150313',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
+        'only_matching': True,
+    }, {
+        'url': 'https://news.sportbox.ru/vdl/player/media/193095',
+        'only_matching': True,
+    }, {
+        'url': 'https://news.sportbox.ru/vdl/player/media/109158',
+        'only_matching': True,
+    }, {
+        'url': 'https://matchtv.ru/vdl/player/media/109158',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src="(https?://(?:news\.sportbox|matchtv)\.ru/vdl/player[^"]+)"',
+            webpage)

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
+        video_id = self._match_id(url)

 
 

-        webpage = self._download_webpage(url, display_id)
+        webpage = self._download_webpage(url, video_id)

 
 

-        video_id = self._search_regex(
-            r'src="/vdl/player/media/(\d+)"', webpage, 'video id')
+        sources = self._parse_json(
+            self._search_regex(
+                r'(?s)playerOptions\.sources(?:WithRes)?\s*=\s*(\[.+?\])\s*;\s*\n',
+                webpage, 'sources'),
+            video_id, transform_source=js_to_json)

 
 

-        player = self._download_webpage(
-            'http://news.sportbox.ru/vdl/player/media/%s' % video_id,
-            display_id, 'Downloading player webpage')
+        formats = []
+        for source in sources:
+            src = source.get('src')
+            if not src:
+                continue
+            if determine_ext(src) == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'url': src,
+                })
+        self._sort_formats(formats)

 
 

-        hls = self._search_regex(
-            r"var\s+original_hls_file\s*=\s*'([^']+)'", player, 'hls file')
+        player = self._parse_json(
+            self._search_regex(
+                r'(?s)playerOptions\s*=\s*({.+?})\s*;\s*\n', webpage,
+                'player options', default='{}'),
+            video_id, transform_source=js_to_json)
+        media_id = player['mediaId']

 
 

-        formats = self._extract_m3u8_formats(hls, display_id, 'mp4')
+        info = self._search_json_ld(webpage, media_id, default={})

 
 

-        title = self._html_search_regex(
-            r'<h1 itemprop="name">([^<]+)</h1>', webpage, 'title')
-        description = self._html_search_regex(
-            r'(?s)<div itemprop="description">(.+?)</div>', webpage, 'description', fatal=False)
-        thumbnail = self._og_search_thumbnail(webpage)
-        timestamp = parse_iso8601(self._search_regex(
-            r'<span itemprop="uploadDate">([^<]+)</span>', webpage, 'timestamp', fatal=False))
-        duration = parse_duration(self._html_search_regex(
-            r'<meta itemprop="duration" content="PT([^"]+)">', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._search_regex(
+            r'Просмотров\s*:\s*(\d+)', webpage, 'view count', default=None))

 
 

-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'duration': duration,
+        return merge_dicts(info, {
+            'id': media_id,
+            'title': self._og_search_title(webpage, default=None) or media_id,
+            'thumbnail': player.get('poster'),
+            'duration': int_or_none(player.get('duration')),
+            'view_count': view_count,

             'formats': formats,
             'formats': formats,

-        }
+        })