Merge pull request #5116 from yan12125/letv_fix

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

diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py
index 62d5707fee87302b5f69544f768b1a56a5e3976b..7fb4e57dfb81029c0fc46dbab83ced5b2fbc766b 100644 (file)
--- a/youtube_dl/extractor/niconico.py
+++ b/youtube_dl/extractor/niconico.py
@@ -5,13 +5,16 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
-    unified_strdate,
-    parse_duration,
+)
+from ..utils import (
+    ExtractorError,
     int_or_none,
+    parse_duration,
+    unified_strdate,
 )
 
 
@@ -38,7 +41,7 @@ class NiconicoIE(InfoExtractor):
         },
     }
 
-    _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/((?:[a-z]{2})?[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
     _NETRC_MACHINE = 'niconico'
     # Determine whether the downloader used authentication to download video
     _AUTHENTICATED = False
@@ -73,8 +76,7 @@ class NiconicoIE(InfoExtractor):
         return True
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = self._match_id(url)
 
         # Get video webpage. We are not actually interested in it, but need
         # the cookies in order to be able to download the info webpage
@@ -108,6 +110,9 @@ class NiconicoIE(InfoExtractor):
                 flv_info_request, video_id,
                 note='Downloading flv info', errnote='Unable to download flv info')
 
+        if 'deleted=' in flv_info_webpage:
+            raise ExtractorError('The video has been deleted.',
+                                 expected=True)
         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 
         # Start extracting information
@@ -166,12 +171,13 @@ class NiconicoPlaylistIE(InfoExtractor):
         webpage = self._download_webpage(url, list_id)
 
         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
-            webpage, 'entries')
+                                          webpage, 'entries')
         entries = json.loads(entries_json)
         entries = [{
             '_type': 'url',
             'ie_key': NiconicoIE.ie_key(),
-            'url': 'http://www.nicovideo.jp/watch/%s' % entry['item_id'],
+            'url': ('http://www.nicovideo.jp/watch/%s' %
+                    entry['item_data']['video_id']),
         } for entry in entries]
 
         return {