[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/expotv.py b/youtube_dl/extractor/expotv.py
index 23a38c7c1673c2929e56a03390a772547ea387e0..95a8977821d3c292470e42f0f9170674ed9a6aa2 100644 (file)
--- a/youtube_dl/extractor/expotv.py
+++ b/youtube_dl/extractor/expotv.py
@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -10,46 +8,48 @@ from ..utils import (
 
 
 class ExpoTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
+    _VALID_URL = r'https?://(?:www\.)?expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
     _TEST = {
-        'url': 'http://www.expotv.com/videos/reviews/1/24/LinneCardscom/17561',
-        'md5': '2985e6d7a392b2f7a05e0ca350fe41d0',
+        'url': 'http://www.expotv.com/videos/reviews/3/40/NYX-Butter-lipstick/667916',
+        'md5': 'fe1d728c3a813ff78f595bc8b7a707a8',
         'info_dict': {
-            'id': '17561',
+            'id': '667916',
             'ext': 'mp4',
-            'upload_date': '20060212',
-            'title': 'My Favorite Online Scrapbook Store',
+            'title': 'NYX Butter Lipstick Little Susie',
+            'description': 'Goes on like butter, but looks better!',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Stephanie S.',
+            'upload_date': '20150520',
             'view_count': int,
-            'description': 'You\'ll find most everything you need at this virtual store front.',
-            'uploader': 'Anna T.',
-            'thumbnail': 're:^https?://.*\.jpg$',
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
         player_key = self._search_regex(
             r'<param name="playerKey" value="([^"]+)"', webpage, 'player key')
         config = self._download_json(
-                'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key),
-                video_id,
-                note='Downloading video configuration')
+            'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key),
+            video_id, 'Downloading video configuration')
 
         formats = []
         for fcfg in config['sources']:
-            if fcfg['type'] == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(fcfg['file'], video_id))
+            media_url = fcfg.get('file')
+            if not media_url:
+                continue
+            if fcfg.get('type') == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    media_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls'))
             else:
                 formats.append({
-                    'url': fcfg['file'],
+                    'url': media_url,
                     'height': int_or_none(fcfg.get('height')),
                     'format_id': fcfg.get('label'),
                     'ext': self._search_regex(
-                        r'filename=.*\.([a-z0-9_A-Z]+)&', fcfg['file'],
-                        'file extension', default=None),
+                        r'filename=.*\.([a-z0-9_A-Z]+)&', media_url,
+                        'file extension', default=None) or fcfg.get('type'),
                 })
         self._sort_formats(formats)
 
@@ -63,7 +63,7 @@ class ExpoTVIE(InfoExtractor):
             fatal=False)
         upload_date = unified_strdate(self._search_regex(
             r'<h5>Reviewed on ([0-9/.]+)</h5>', webpage, 'upload date',
-            fatal=False))
+            fatal=False), day_first=False)
 
         return {
             'id': video_id,