[drbonanza] Fix extraction (closes #13231)

youtube_dl/extractor/drbonanza.py

@@ -1,135 +1,59 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    int_or_none,
+    js_to_json,
-    parse_iso8601,
+    parse_duration,
+    unescapeHTML,
 )
 
 
 class DRBonanzaIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/(?:[^/]+/)+(?:[^/])+?(?:assetId=(?P<id>\d+))?(?:[#&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/[^/]+/\d+/[^/]+/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
-
+    _TEST = {
-    _TESTS = [{
+        'url': 'http://www.dr.dk/bonanza/serie/154/matador/40312/matador---0824-komme-fremmede-',
-        'url': 'http://www.dr.dk/bonanza/serie/portraetter/Talkshowet.htm?assetId=65517',
         'info_dict': {
-            'id': '65517',
+            'id': '40312',
+            'display_id': 'matador---0824-komme-fremmede-',
             'ext': 'mp4',
-            'title': 'Talkshowet - Leonard Cohen',
+            'title': 'MATADOR - 08:24. "Komme fremmede".',
-            'description': 'md5:8f34194fb30cd8c8a30ad8b27b70c0ca',
+            'description': 'md5:77b4c1ac4d4c1b9d610ab4395212ff84',
             'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
-            'timestamp': 1295537932,
+            'duration': 4613,
-            'upload_date': '20110120',
-            'duration': 3664,
         },
-        'params': {
+    }
-            'skip_download': True,  # requires rtmp
-        },
-    }, {
-        'url': 'http://www.dr.dk/bonanza/radio/serie/sport/fodbold.htm?assetId=59410',
-        'md5': '6dfe039417e76795fb783c52da3de11d',
-        'info_dict': {
-            'id': '59410',
-            'ext': 'mp3',
-            'title': 'EM fodbold 1992 Danmark - Tyskland finale Transmission',
-            'description': 'md5:501e5a195749480552e214fbbed16c4e',
-            'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
-            'timestamp': 1223274900,
-            'upload_date': '20081006',
-            'duration': 7369,
-        },
-    }]
 
     def _real_extract(self, url):
-        url_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
-        webpage = self._download_webpage(url, url_id)
+        video_id, display_id = mobj.group('id', 'display_id')
 
-        if url_id:
+        webpage = self._download_webpage(url, display_id)
-            info = json.loads(self._html_search_regex(r'({.*?%s.*})' % url_id, webpage, 'json'))
-        else:
-            # Just fetch the first video on that page
-            info = json.loads(self._html_search_regex(r'bonanzaFunctions.newPlaylist\(({.*})\)', webpage, 'json'))
 
-        asset_id = str(info['AssetId'])
+        info = self._parse_html5_media_entries(
-        title = info['Title'].rstrip(' \'\"-,.:;!?')
+            url, webpage, display_id, m3u8_id='hls',
-        duration = int_or_none(info.get('Duration'), scale=1000)
+            m3u8_entry_protocol='m3u8_native')[0]
-        # First published online. "FirstPublished" contains the date for original airing.
+        self._sort_formats(info['formats'])
-        timestamp = parse_iso8601(
-            re.sub(r'\.\d+$', '', info['Created']))
 
-        def parse_filename_info(url):
+        asset = self._parse_json(
-            match = re.search(r'/\d+_(?P<width>\d+)x(?P<height>\d+)x(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
+            self._search_regex(
-            if match:
+                r'(?s)currentAsset\s*=\s*({.+?})\s*</script', webpage, 'asset'),
-                return {
+            display_id, transform_source=js_to_json)
-                    'width': int(match.group('width')),
-                    'height': int(match.group('height')),
-                    'vbr': int(match.group('bitrate')),
-                    'ext': match.group('ext')
-                }
-            match = re.search(r'/\d+_(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
-            if match:
-                return {
-                    'vbr': int(match.group('bitrate')),
-                    'ext': match.group(2)
-                }
-            return {}
 
-        video_types = ['VideoHigh', 'VideoMid', 'VideoLow']
+        title = unescapeHTML(asset['AssetTitle']).strip()
-        preferencemap = {
-            'VideoHigh': -1,
-            'VideoMid': -2,
-            'VideoLow': -3,
-            'Audio': -4,
-        }
 
-        formats = []
+        def extract(field):
-        for file in info['Files']:
+            return self._search_regex(
-            if info['Type'] == 'Video':
+                r'<div[^>]+>\s*<p>%s:<p>\s*</div>\s*<div[^>]+>\s*<p>([^<]+)</p>' % field,
-                if file['Type'] in video_types:
+                webpage, field, default=None)
-                    format = parse_filename_info(file['Location'])
-                    format.update({
-                        'url': file['Location'],
-                        'format_id': file['Type'].replace('Video', ''),
-                        'preference': preferencemap.get(file['Type'], -10),
-                    })
-                    if format['url'].startswith('rtmp'):
-                        rtmp_url = format['url']
-                        format['rtmp_live'] = True  # --resume does not work
-                        if '/bonanza/' in rtmp_url:
-                            format['play_path'] = rtmp_url.split('/bonanza/')[1]
-                    formats.append(format)
-                elif file['Type'] == 'Thumb':
-                    thumbnail = file['Location']
-            elif info['Type'] == 'Audio':
-                if file['Type'] == 'Audio':
-                    format = parse_filename_info(file['Location'])
-                    format.update({
-                        'url': file['Location'],
-                        'format_id': file['Type'],
-                        'vcodec': 'none',
-                    })
-                    formats.append(format)
-                elif file['Type'] == 'Thumb':
-                    thumbnail = file['Location']
 
-        description = '%s\n%s\n%s\n' % (
+        info.update({
-            info['Description'], info['Actors'], info['Colophon'])
+            'id': asset.get('AssetId') or video_id,
-
-        self._sort_formats(formats)
-
-        display_id = re.sub(r'[^\w\d-]', '', re.sub(r' ', '-', title.lower())) + '-' + asset_id
-        display_id = re.sub(r'-+', '-', display_id)
-
-        return {
-            'id': asset_id,
             'display_id': display_id,
             'title': title,
-            'formats': formats,
+            'description': extract('Programinfo'),
-            'description': description,
+            'duration': parse_duration(extract('Tid')),
-            'thumbnail': thumbnail,
+            'thumbnail': asset.get('AssetImageUrl'),
-            'timestamp': timestamp,
+        })
-            'duration': duration,
+        return info
-        }