[howstuffworks] Update extractor

youtube_dl/extractor/howstuffworks.py

@@ -1,19 +1,20 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import find_xpath_attr
 
 
 class HowStuffWorksIE(InfoExtractor):
     _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
     _TESTS = [
         {
-            'url': 'http://adventure.howstuffworks.com/39521-deadliest-catch-nautical-collision-video.htm',
+            'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
             'info_dict': {
-                'id': '553475',
-                'ext': 'mp4',
-                'title': 'Deadliest Catch: Nautical Collision',
-                'description': 'Check out this clip to get an exclusive look at the Deadliest Catch crew back in action.',
-                'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/39521.jpg',
+                'id': '450221',
+                'ext': 'flv',
+                'title': 'Cool Jobs - Iditarod Musher',
+                'description': 'Cold sleds, freezing temps and warm dog breath... an Iditarod musher\'s dream. Kasey-Dee Gardner jumps on a sled to find out what the big deal is.',
+                'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/5266.jpg',
             },
         },
         {
@@ -27,49 +28,73 @@ class HowStuffWorksIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://entertainment.howstuffworks.com/arts/34476-time-warp-fire-breathing-disaster-video.htm',
+            'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
             'info_dict': {
-                'id': '487036',
-                'ext': 'mp4',
-                'title': 'Time Warp: Fire-Breathing Disaster',
-                'description': 'Fire-breathers are safe from flames when they angle their heads up as they blow -- see what would happen if they looked down instead. Check out this clip from Discovery\'s "Time Warp" series to learn more.',
-                'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/34476.jpg',
+                'id': '440011',
+                'ext': 'flv',
+                'title': 'Sword Swallowing #1 by Dan Meyer',
+                'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International <www.swordswallow.org>',
+                'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/118306353233.jpg',
             },
         },
     ]
 
-    def _extract_clip_info(self, key, clip_info, name=None, **kargs):
-        if name is None:
-            name = key
-        return self._html_search_regex(
-            r"\s*%s\s*: '?([^'\n]*[^'\n,])" % key, clip_info, name, **kargs)
-
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
         clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info')
-        video_id = self._extract_clip_info('content_id', clip_info, 'video id')
+
+        def extract_clip_info(key, clip_info, name=None, **kargs):
+            if name is None:
+                name = key
+            return self._html_search_regex(
+                r"\s*%s\s*: '?([^'\n]*[^'\n,])" % key, clip_info, name, **kargs)
+
+        video_id = extract_clip_info('content_id', clip_info, 'video id')
         formats = []
-        m3u8_url = self._extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None)
+        m3u8_url = extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None)
         if m3u8_url is not None:
             formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4')
         mp4 = self._parse_json(
-            self._extract_clip_info(
+            extract_clip_info(
                 'mp4', clip_info, 'formats').replace('},]','}]'), video_id)
         for video in mp4:
             formats.append({
                 'url': video['src'],
                 'format_id': video['bitrate'],
-                'tbr': int(video['bitrate'].rstrip('k')),
+                'vbr': int(video['bitrate'].rstrip('k')),
             })
+
+        if not formats:
+            smil = self._download_xml(
+                'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
+                video_id, 'Downloading video SMIL')
+
+            http_base = find_xpath_attr(
+                smil,
+                './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
+                'name',
+                'httpBase').get('content')
+
+            URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'
+
+            for video in smil.findall(
+                    './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
+                vbr = int(video.attrib['system-bitrate']) / 1000
+                formats.append({
+                    'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
+                    'format_id': '%dk' % vbr,
+                    'vbr': vbr,
+                })
+
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'display_id': display_id,
-            'title': self._extract_clip_info('clip_title', clip_info, 'title'),
-            'description': self._extract_clip_info('caption', clip_info, 'description', fatal=False),
-            'thumbnail': self._extract_clip_info('video_still_url', clip_info, 'thumbnail'),
-            'duration': self._extract_clip_info('duration', clip_info),
+            'title': extract_clip_info('clip_title', clip_info, 'title'),
+            'description': extract_clip_info('caption', clip_info, 'description', fatal=False),
+            'thumbnail': extract_clip_info('video_still_url', clip_info, 'thumbnail'),
+            'duration': extract_clip_info('duration', clip_info),
             'formats': formats,
         }