Fix "invalid escape sequences" error on Python 3.6

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

diff --git a/youtube_dl/extractor/sportbox.py b/youtube_dl/extractor/sportbox.py
index 8686f9d11fa3178eefe3eb71fe5a1413e729beeb..b512cd20fbe495023eb7ec9f2c342245b5f2a44d 100644 (file)
--- a/youtube_dl/extractor/sportbox.py
+++ b/youtube_dl/extractor/sportbox.py
@@ -6,8 +6,8 @@ import re
 from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
-    parse_duration,
-    parse_iso8601,
+    js_to_json,
+    unified_strdate,
 )
 
 
@@ -20,11 +20,9 @@ class SportBoxIE(InfoExtractor):
             'id': '80822',
             'ext': 'mp4',
             'title': 'Гонка 2  заезд ««Объединенный 2000»: классы Туринг и Супер-продакшн',
-            'description': 'md5:81715fa9c4ea3d9e7915dc8180c778ed',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'timestamp': 1411896237,
+            'description': 'md5:3d72dc4a006ab6805d82f037fdc637ad',
+            'thumbnail': r're:^https?://.*\.jpg$',
             'upload_date': '20140928',
-            'duration': 4846,
         },
         'params': {
             # m3u8 download
@@ -48,17 +46,13 @@ class SportBoxIE(InfoExtractor):
             r'src="/?(vdl/player/[^"]+)"', webpage, 'player')
 
         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)
+            [r'"nodetitle"\s*:\s*"([^"]+)"', r'class="node-header_{1,2}title">([^<]+)'],
+            webpage, 'title')
+        description = self._og_search_description(webpage) or self._html_search_meta(
+            'description', webpage, 'description')
         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))
+        upload_date = unified_strdate(self._html_search_meta(
+            'dateCreated', webpage, 'upload date'))
 
         return {
             '_type': 'url_transparent',
@@ -67,8 +61,7 @@ class SportBoxIE(InfoExtractor):
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'duration': duration,
+            'upload_date': upload_date,
         }
 
 
@@ -80,7 +73,7 @@ class SportBoxEmbedIE(InfoExtractor):
             'id': '211355',
             'ext': 'mp4',
             'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
         },
         'params': {
             # m3u8 download
@@ -102,18 +95,32 @@ class SportBoxEmbedIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        hls = self._search_regex(
-            r"sportboxPlayer\.jwplayer_common_params\.file\s*=\s*['\"]([^'\"]+)['\"]",
-            webpage, 'hls file')
+        formats = []
 
-        formats = self._extract_m3u8_formats(hls, video_id, 'mp4')
+        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))
 
-        title = self._search_regex(
-            r'sportboxPlayer\.node_title\s*=\s*"([^"]+)"', webpage, 'title')
+        jwplayer_data = self._parse_json(self._search_regex(
+            r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
+            transform_source=cleanup_js)
 
-        thumbnail = self._search_regex(
-            r'sportboxPlayer\.jwplayer_common_params\.image\s*=\s*"([^"]+)"',
-            webpage, 'thumbnail', default=None)
+        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'))
+
+        rtsp_url = jwplayer_data.get('rtsp_url')
+        if rtsp_url:
+            formats.append({
+                'url': rtsp_url,
+                'format_id': 'rtsp',
+            })
+
+        self._sort_formats(formats)
+
+        title = jwplayer_data['node_title']
+        thumbnail = jwplayer_data.get('image_url')
 
         return {
             'id': video_id,