[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/ora.py b/youtube_dl/extractor/ora.py
index 02de1502a42f40153e912536ac721221e0dbc00d..1d42be39b3303c95952a8ec54a34abbb9d09f0b1 100644 (file)
--- a/youtube_dl/extractor/ora.py
+++ b/youtube_dl/extractor/ora.py
@@ -12,8 +12,8 @@ from ..utils import (
 
 
 class OraTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?ora\.tv/([^/]+/)*(?P<id>[^/\?#]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?(?:ora\.tv|unsafespeech\.com)/([^/]+/)*(?P<id>[^/\?#]+)'
+    _TESTS = [{
         'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq',
         'md5': 'fa33717591c631ec93b04b0e330df786',
         'info_dict': {
@@ -21,18 +21,20 @@ class OraTVIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Vine & YouTube Stars Zach King & King Bach On Their Viral Videos!',
             'description': 'md5:ebbc5b1424dd5dba7be7538148287ac1',
-            'duration': 1477,
         }
-    }
+    }, {
+        'url': 'http://www.unsafespeech.com/video/2016/5/10/student-self-censorship-and-the-thought-police-on-university-campuses-0_6622bnkppw4d',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
 
         video_data = self._search_regex(
-            r'"current"\s*:\s*({[^}]+?})', webpage, 'current video')
+            r'"(?:video|current)"\s*:\s*({[^}]+?})', webpage, 'current video')
         m3u8_url = self._search_regex(
-            r'"hls_stream"\s*:\s*"([^"]+)', video_data, 'm3u8 url', None)
+            r'hls_stream"?\s*:\s*"([^"]+)', video_data, 'm3u8 url', None)
         if m3u8_url:
             formats = self._extract_m3u8_formats(
                 m3u8_url, display_id, 'mp4', 'm3u8_native',
@@ -62,14 +64,12 @@ class OraTVIE(InfoExtractor):
 
         return {
             'id': self._search_regex(
-                r'"video_id"\s*:\s*(\d+)', video_data, 'video id'),
+                r'"id"\s*:\s*(\d+)', video_data, 'video id', default=display_id),
             'display_id': display_id,
             'title': unescapeHTML(self._og_search_title(webpage)),
             'description': get_element_by_attribute(
                 'class', 'video_txt_decription', webpage),
             'thumbnail': self._proto_relative_url(self._search_regex(
                 r'"thumb"\s*:\s*"([^"]+)', video_data, 'thumbnail', None)),
-            'duration': int(self._search_regex(
-                r'"duration"\s*:\s*(\d+)', video_data, 'duration')),
             'formats': formats,
         }