X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdiscovery.py;h=ce680a9f3192832a7fb9e35956382f85bc60d043;hb=2e7e561c1d9dedf1a8e5a206e1ef86cfa4599956;hp=14fca3cae00194967f44b27117ded4e7db1a3c71;hpb=96d7b8873ad47c1f52193e84fc6f8cfe12891aa7;p=youtube-dl

diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py
index 14fca3cae..ce680a9f3 100644
--- a/youtube_dl/extractor/discovery.py
+++ b/youtube_dl/extractor/discovery.py
@@ -1,40 +1,80 @@
 from __future__ import unicode_literals
 
-import re
-import json
 from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    parse_iso8601,
+)
+from ..compat import compat_str
 
 
 class DiscoveryIE(InfoExtractor):
-    _VALID_URL = r'http://dsc\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9\-]*)(.htm)?'
-    _TEST = {
-        'url': 'http://dsc.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
-        'file': 'mission-impossible-outtakes.mp4',
-        'md5': 'e12614f9ee303a6ccef415cb0793eba2',
+    _VALID_URL = r'''(?x)http://(?:www\.)?(?:
+            discovery|
+            investigationdiscovery|
+            discoverylife|
+            animalplanet|
+            ahctv|
+            destinationamerica|
+            sciencechannel|
+            tlc|
+            velocity
+        )\.com/(?:[^/]+/)*(?P<id>[^./?#]+)'''
+    _TESTS = [{
+        'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
         'info_dict': {
-            'title': 'MythBusters: Mission Impossible Outtakes'
+            'id': '20769',
+            'ext': 'mp4',
+            'title': 'Mission Impossible Outtakes',
+            'description': ('Watch Jamie Hyneman and Adam Savage practice being'
+                            ' each other -- to the point of confusing Jamie\'s dog -- and '
+                            'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
+                            ' back.'),
+            'duration': 156,
+            'timestamp': 1302032462,
+            'upload_date': '20110405',
+        },
+        'params': {
+            'skip_download': True,  # requires ffmpeg
         }
-    }
+    }, {
+        'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mythbusters-the-simpsons',
+        'info_dict': {
+            'id': 'mythbusters-the-simpsons',
+            'title': 'MythBusters: The Simpsons',
+        },
+        'playlist_mincount': 10,
+    }, {
+        'url': 'http://www.animalplanet.com/longfin-eels-maneaters/',
+        'info_dict': {
+            'id': '78326',
+            'ext': 'mp4',
+            'title': 'Longfin Eels: Maneaters?',
+            'description': 'Jeremy Wade tests whether or not New Zealand\'s longfin eels are man-eaters by covering himself in fish guts and getting in the water with them.',
+            'upload_date': '20140725',
+            'timestamp': 1406246400,
+            'duration': 116,
+        },
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(
-            r'(?<=\"name\": ")(?P<title>.*?)(?=\"\,)', webpage, r'Filename')
-        duration = int(self._search_regex(
-            r'(?<=\"duration\"\: )(?P<duration>.*?)(?=,)', webpage, r'Duration'))
-        formats_raw = self._search_regex(
-            r'(?<=\"mp4\":)(.*?)(}])', webpage, r'formats') + '}]'
-        formats_json = json.loads(formats_raw)
-        formats = []
-        for f in formats_json:
-            formats.append(
-                {'url': f['src'], r'ext': r'mp4', 'tbr': int(f['bitrate'][:-1])})
+        display_id = self._match_id(url)
+        info = self._download_json(url + '?flat=1', display_id)
 
-        return {
-            'id': video_id,
-            'duration': duration,
-            'title': title,
-            'formats': formats
-        }
+        video_title = info.get('playlist_title') or info.get('video_title')
+
+        entries = [{
+            'id': compat_str(video_info['id']),
+            'formats': self._extract_m3u8_formats(
+                video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls',
+                note='Download m3u8 information for video %d' % (idx + 1)),
+            'title': video_info['title'],
+            'description': video_info.get('description'),
+            'duration': parse_duration(video_info.get('video_length')),
+            'webpage_url': video_info.get('href') or video_info.get('url'),
+            'thumbnail': video_info.get('thumbnailURL'),
+            'alt_title': video_info.get('secondary_title'),
+            'timestamp': parse_iso8601(video_info.get('publishedDate')),
+        } for idx, video_info in enumerate(info['playlist'])]
+
+        return self.playlist_result(entries, display_id, video_title)