[youtube] Extract chapters

[youtube-dl] / youtube_dl / extractor / sportbox.py

diff --git a/youtube_dl/extractor/sportbox.py b/youtube_dl/extractor/sportbox.py
index becdf658f6e0ce8b209dffc0ce4c96a2857099dc..e7bd5bf91921752757e1bc9beb390798b86a9c8a 100644 (file)
--- a/youtube_dl/extractor/sportbox.py
+++ b/youtube_dl/extractor/sportbox.py
@@ -4,73 +4,69 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    parse_duration,
-    parse_iso8601,
-)
+from ..utils import js_to_json
 
 
-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$',
-                'timestamp': 1411896237,
-                'upload_date': '20140928',
-                'duration': 4846,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        }, {
-            'url': 'http://news.sportbox.ru/Vidy_sporta/billiard/spbvideo_NI486287_CHempionat-mira-po-dinamichnoy-piramide-4',
-            'only_matching': True,
-        }
-    ]
+class SportBoxEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://news\.sportbox\.ru/vdl/player(?:/[^/]+/|\?.*?\bn?id=)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://news.sportbox.ru/vdl/player/ci/211355',
+        'info_dict': {
+            'id': '211355',
+            'ext': 'mp4',
+            'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+        '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,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src="(https?://news\.sportbox\.ru/vdl/player[^"]+)"',
+            webpage)
 
     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, video_id)
+
+        formats = []
 
-        webpage = self._download_webpage(url, display_id)
+        def cleanup_js(code):
+            # desktop_advert_config contains complex Javascripts and we don't need it
+            return js_to_json(re.sub(r'desktop_advert_config.*', '', code))
 
-        video_id = self._search_regex(
-            r'src="/vdl/player/media/(\d+)"', webpage, 'video id')
+        jwplayer_data = self._parse_json(self._search_regex(
+            r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
+            transform_source=cleanup_js)
 
-        player = self._download_webpage(
-            'http://news.sportbox.ru/vdl/player/media/%s' % video_id,
-            display_id, 'Downloading player webpage')
+        hls_url = jwplayer_data.get('hls_url')
+        if hls_url:
+            formats.extend(self._extract_m3u8_formats(
+                hls_url, video_id, ext='mp4', m3u8_id='hls'))
 
-        hls = self._search_regex(
-            r"var\s+original_hls_file\s*=\s*'([^']+)'", player, 'hls file')
+        rtsp_url = jwplayer_data.get('rtsp_url')
+        if rtsp_url:
+            formats.append({
+                'url': rtsp_url,
+                'format_id': 'rtsp',
+            })
 
-        formats = self._extract_m3u8_formats(hls, display_id, 'mp4')
+        self._sort_formats(formats)
 
-        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))
+        title = jwplayer_data['node_title']
+        thumbnail = jwplayer_data.get('image_url')
 
         return {
             'id': video_id,
-            'display_id': display_id,
             'title': title,
-            'description': description,
             'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'duration': duration,
             'formats': formats,
         }