X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdrbonanza.py;h=164e97c36b8cab223ac813abaa1f032eb7d4444a;hb=384bf91f8802805283f993714ace137e831ba45e;hp=8cbdb57b376d95cf4e0983e6d5418895d1fce991;hpb=8e7a9016d5e79679f4b948f00c794b8ff31a41e4;p=youtube-dl

diff --git a/youtube_dl/extractor/drbonanza.py b/youtube_dl/extractor/drbonanza.py
index 8cbdb57b3..164e97c36 100644
--- a/youtube_dl/extractor/drbonanza.py
+++ b/youtube_dl/extractor/drbonanza.py
@@ -1,120 +1,59 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from .common import ExtractorError
-from ..utils import parse_iso8601
-import json
 import re
 
-class DRBonanzaIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/(?:[^/]+/)+(?:[^/])+?(?:assetId=(?P<id>\d+))?(?:[#&]|$)'
+from .common import InfoExtractor
+from ..utils import (
+    js_to_json,
+    parse_duration,
+    unescapeHTML,
+)
 
-    _TESTS = [{
-        'url': 'http://www.dr.dk/bonanza/serie/portraetter/Talkshowet.htm?assetId=65517',
-        'md5': 'fe330252ddea607635cf2eb2c99a0af3',
+
+class DRBonanzaIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/[^/]+/\d+/[^/]+/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://www.dr.dk/bonanza/serie/154/matador/40312/matador---0824-komme-fremmede-',
         'info_dict': {
-            'id': '65517',
+            'id': '40312',
+            'display_id': 'matador---0824-komme-fremmede-',
             'ext': 'mp4',
-            'title': 'Talkshowet - Leonard Cohen',
-            'description': 'md5:8f34194fb30cd8c8a30ad8b27b70c0ca',
-            'timestamp': 1295537932,
-            'upload_date': '20110120',
-            'duration': 3664000,
-        },
-    },{
-        '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',
-            'timestamp': 1223274900,
-            'upload_date': '20081006',
-            'duration': 7369000,
+            'title': 'MATADOR - 08:24. "Komme fremmede".',
+            'description': 'md5:77b4c1ac4d4c1b9d610ab4395212ff84',
+            'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
+            'duration': 4613,
         },
-    }]
+    }
 
     def _real_extract(self, url):
-        url_id = self._match_id(url)
-        
-        webpage = self._download_webpage(url, url_id if url_id else "")
-        
-        if url_id:
-            info = json.loads(self._html_search_regex(r'({.*?' + 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'])
-        title = info['Title'].rstrip(' \'\"-,.:;!?')
-        duration = info['Duration']
-        timestamp = parse_iso8601(re.sub(r'\.\d+$', '', info['Created'])) # First published online. "FirstPublished" contains the date for original airing.
-        
-        def parse_filename_info(url):
-            match = re.search(r'/\d+_(?P<width>\d+)x(?P<height>\d+)x(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
-            if match:
-                return {'width': int(match.group(1)), 'height': int(match.group(2)), 'bitrate': int(match.group(3)), 'ext': match.group(4)}
-            match = re.search(r'/\d+_(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
-            if match:
-                return {'bitrate': int(match.group(1)), 'ext': match.group(2)}
-            return {'width': None, 'height': None, 'bitrate': None, 'ext': None}
-        
-        video_types = ['VideoHigh', 'VideoMid', 'VideoLow']
-        preferencemap = {
-            'VideoHigh': -1,
-            'VideoMid': -2,
-            'VideoLow': -3,
-            'Audio': -4,
-        }
-        
-        formats = []
-        for file in info['Files']:
-            if info['Type'] == "Video":
-                if file['Type'] in video_types:
-                    fileinfo = parse_filename_info(file['Location'])
-                    formats.append({
-                        'url': file['Location'],
-                        'format_id': file['Type'].replace('Video', ''),
-                        'preference': preferencemap.get(file['Type'], -10),
-                        'width': fileinfo['width'],
-                        'height': fileinfo['height'],
-                        'vbr': fileinfo['bitrate'],
-                        'ext': fileinfo['ext'],
-                    })
-                elif file['Type'] == "Thumb":
-                    thumbnail = file['Location']
-            elif info['Type'] == "Audio":
-                if file['Type'] == "Audio":
-                    fileinfo = parse_filename_info(file['Location'])
-                    formats.append({
-                        'url': file['Location'],
-                        'format_id': file['Type'],
-                        'abr': fileinfo['bitrate'],
-                        'ext': fileinfo['ext'],
-                        'vcodec': 'none',
-                    })
-                elif file['Type'] == "Thumb":
-                    thumbnail = file['Location']
-        
-        description = "{}\n{}\n{}\n".format(info['Description'], info['Actors'], info['Colophon'])
+        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = mobj.group('id', 'display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        info = self._parse_html5_media_entries(
+            url, webpage, display_id, m3u8_id='hls',
+            m3u8_entry_protocol='m3u8_native')[0]
+        self._sort_formats(info['formats'])
+
+        asset = self._parse_json(
+            self._search_regex(
+                r'(?s)currentAsset\s*=\s*({.+?})\s*</script', webpage, 'asset'),
+            display_id, transform_source=js_to_json)
+
+        title = unescapeHTML(asset['AssetTitle']).strip()
+
+        def extract(field):
+            return self._search_regex(
+                r'<div[^>]+>\s*<p>%s:<p>\s*</div>\s*<div[^>]+>\s*<p>([^<]+)</p>' % field,
+                webpage, field, default=None)
 
-        for f in formats:
-            f['url'] = f['url'].replace('rtmp://vod-bonanza.gss.dr.dk/bonanza/', 'http://vodfiles.dr.dk/')
-            f['url'] = f['url'].replace('mp4:bonanza', 'bonanza')
-        
-        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,
+        info.update({
+            'id': asset.get('AssetId') or video_id,
             'display_id': display_id,
             'title': title,
-            'formats': formats,
-            'description': description,
-            'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'duration': duration,
-        }
+            'description': extract('Programinfo'),
+            'duration': parse_duration(extract('Tid')),
+            'thumbnail': asset.get('AssetImageUrl'),
+        })
+        return info