X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Flifenews.py;h=7d5b8621b149fb0cb60555db0f7adb0c2c5e66fc;hb=057ebeaca3da40b901b2592e2302a0d4bbab48e3;hp=8d9491f233bf578bc9274a18fe022a1538effad6;hpb=b1d65c33695feec76a801b7647fe8077869c25ad;p=youtube-dl

diff --git a/youtube_dl/extractor/lifenews.py b/youtube_dl/extractor/lifenews.py
index 8d9491f23..7d5b8621b 100644
--- a/youtube_dl/extractor/lifenews.py
+++ b/youtube_dl/extractor/lifenews.py
@@ -14,9 +14,9 @@ from ..utils import (
 class LifeNewsIE(InfoExtractor):
     IE_NAME = 'lifenews'
     IE_DESC = 'LIFE | NEWS'
-    _VALID_URL = r'http://lifenews\.ru/(?:mobile/)?news/(?P<id>\d+)'
+    _VALID_URL = r'http://lifenews\.ru/(?:mobile/)?(?P<section>news|video)/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://lifenews.ru/news/126342',
         'md5': 'e1b50a5c5fb98a6a544250f2e0db570a',
         'info_dict': {
@@ -27,16 +27,47 @@ class LifeNewsIE(InfoExtractor):
             'thumbnail': 're:http://.*\.jpg',
             'upload_date': '20140130',
         }
-    }
+    }, {
+        # video in <iframe>
+        'url': 'http://lifenews.ru/news/152125',
+        'md5': '77d19a6f0886cd76bdbf44b4d971a273',
+        'info_dict': {
+            'id': '152125',
+            'ext': 'mp4',
+            'title': 'В Сети появилось видео захвата «Правым сектором» колхозных полей ',
+            'description': 'Жители двух поселков Днепропетровской области не простили радикалам угрозу лишения плодородных земель и пошли в лобовую. ',
+            'upload_date': '20150402',
+            'uploader': 'embed.life.ru',
+        }
+    }, {
+        'url': 'http://lifenews.ru/news/153461',
+        'md5': '9b6ef8bc0ffa25aebc8bdb40d89ab795',
+        'info_dict': {
+            'id': '153461',
+            'ext': 'mp4',
+            'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве',
+            'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.',
+            'upload_date': '20150505',
+            'uploader': 'embed.life.ru',
+        }
+    }, {
+        'url': 'http://lifenews.ru/video/13035',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        section = mobj.group('section')
 
-        webpage = self._download_webpage('http://lifenews.ru/news/%s' % video_id, video_id, 'Downloading page')
+        webpage = self._download_webpage(
+            'http://lifenews.ru/%s/%s' % (section, video_id),
+            video_id, 'Downloading page')
 
         videos = re.findall(r'<video.*?poster="(?P<poster>[^"]+)".*?src="(?P<video>[^"]+)".*?></video>', webpage)
-        if not videos:
+        iframe_link = self._html_search_regex(
+            '<iframe[^>]+src=["\']([^"\']+)["\']', webpage, 'iframe link', default=None)
+        if not videos and not iframe_link:
             raise ExtractorError('No media links available for %s' % video_id)
 
         title = self._og_search_title(webpage)
@@ -47,28 +78,44 @@ class LifeNewsIE(InfoExtractor):
         description = self._og_search_description(webpage)
 
         view_count = self._html_search_regex(
-            r'<div class=\'views\'>(\d+)</div>', webpage, 'view count', fatal=False)
+            r'<div class=\'views\'>\s*(\d+)\s*</div>', webpage, 'view count', fatal=False)
         comment_count = self._html_search_regex(
-            r'<div class=\'comments\'>\s*<span class=\'counter\'>(\d+)</span>', webpage, 'comment count', fatal=False)
+            r'<div class=\'comments\'>\s*<span class=\'counter\'>\s*(\d+)\s*</span>', webpage, 'comment count', fatal=False)
 
         upload_date = self._html_search_regex(
-            r'<time datetime=\'([^\']+)\'>', webpage, 'upload date',fatal=False)
+            r'<time datetime=\'([^\']+)\'>', webpage, 'upload date', fatal=False)
         if upload_date is not None:
             upload_date = unified_strdate(upload_date)
 
+        common_info = {
+            'description': description,
+            'view_count': int_or_none(view_count),
+            'comment_count': int_or_none(comment_count),
+            'upload_date': upload_date,
+        }
+
         def make_entry(video_id, media, video_number=None):
-            return {
+            cur_info = dict(common_info)
+            cur_info.update({
                 'id': video_id,
                 'url': media[1],
                 'thumbnail': media[0],
                 'title': title if video_number is None else '%s-video%s' % (title, video_number),
-                'description': description,
-                'view_count': int_or_none(view_count),
-                'comment_count': int_or_none(comment_count),
-                'upload_date': upload_date,
-            }
+            })
+            return cur_info
+
+        if iframe_link:
+            iframe_link = self._proto_relative_url(iframe_link, 'http:')
+            cur_info = dict(common_info)
+            cur_info.update({
+                '_type': 'url_transparent',
+                'id': video_id,
+                'title': title,
+                'url': iframe_link,
+            })
+            return cur_info
 
         if len(videos) == 1:
             return make_entry(video_id, videos[0])
         else:
-            return [make_entry(video_id, media, video_number+1) for video_number, media in enumerate(videos)]
\ No newline at end of file
+            return [make_entry(video_id, media, video_number + 1) for video_number, media in enumerate(videos)]