[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py
index 11168fc4dce3ca0d9d440ef286d68ad68df29e5b..e55b1a202426026763375880b549180efbd52a42 100644 (file)
--- a/youtube_dl/extractor/livestream.py
+++ b/youtube_dl/extractor/livestream.py
@@ -14,6 +14,7 @@ from ..utils import (
     xpath_with_ns,
     xpath_text,
     orderedSet,
+    update_url_query,
     int_or_none,
     float_or_none,
     parse_iso8601,
@@ -36,7 +37,7 @@ class LivestreamIE(InfoExtractor):
             'duration': 5968.0,
             'like_count': int,
             'view_count': int,
-            'thumbnail': 're:^http://.*\.jpg$'
+            'thumbnail': r're:^http://.*\.jpg$'
         }
     }, {
         'url': 'http://new.livestream.com/tedx/cityenglish',
@@ -64,7 +65,7 @@ class LivestreamIE(InfoExtractor):
     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
         base_ele = find_xpath_attr(
             smil, self._xpath_ns('.//meta', namespace), 'name', 'httpBase')
-        base = base_ele.get('content') if base_ele else 'http://livestreamvod-f.akamaihd.net/'
+        base = base_ele.get('content') if base_ele is not None else 'http://livestreamvod-f.akamaihd.net/'
 
         formats = []
         video_nodes = smil.findall(self._xpath_ns('.//video', namespace))
@@ -72,7 +73,10 @@ class LivestreamIE(InfoExtractor):
         for vn in video_nodes:
             tbr = int_or_none(vn.attrib.get('system-bitrate'), 1000)
             furl = (
-                '%s%s?v=3.0.3&fp=WIN%%2014,0,0,145' % (base, vn.attrib['src']))
+                update_url_query(compat_urlparse.urljoin(base, vn.attrib['src']), {
+                    'v': '3.0.3',
+                    'fp': 'WIN% 14,0,0,145',
+                }))
             if 'clipBegin' in vn.attrib:
                 furl += '&ssek=' + vn.attrib['clipBegin']
             formats.append({
@@ -97,6 +101,8 @@ class LivestreamIE(InfoExtractor):
             video_url = video_data.get(key)
             if video_url:
                 ext = determine_ext(video_url)
+                if ext == 'm3u8':
+                    continue
                 bitrate = int_or_none(self._search_regex(
                     r'(\d+)\.%s' % ext, video_url, 'bitrate', default=None))
                 formats.append({
@@ -108,23 +114,18 @@ class LivestreamIE(InfoExtractor):
 
         smil_url = video_data.get('smil_url')
         if smil_url:
-            smil_formats = self._extract_smil_formats(smil_url, video_id)
-            if smil_formats:
-                formats.extend(smil_formats)
+            formats.extend(self._extract_smil_formats(smil_url, video_id, fatal=False))
 
         m3u8_url = video_data.get('m3u8_url')
         if m3u8_url:
-            m3u8_formats = self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
-            if m3u8_formats:
-                formats.extend(m3u8_formats)
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         f4m_url = video_data.get('f4m_url')
         if f4m_url:
-            f4m_formats = self._extract_f4m_formats(
-                f4m_url, video_id, f4m_id='hds', fatal=False)
-            if f4m_formats:
-                formats.extend(f4m_formats)
+            formats.extend(self._extract_f4m_formats(
+                f4m_url, video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
         comments = [{
@@ -150,23 +151,19 @@ class LivestreamIE(InfoExtractor):
         }
 
     def _extract_stream_info(self, stream_info):
-        broadcast_id = stream_info['broadcast_id']
+        broadcast_id = compat_str(stream_info['broadcast_id'])
         is_live = stream_info.get('is_live')
 
         formats = []
         smil_url = stream_info.get('play_url')
         if smil_url:
-            smil_formats = self._extract_smil_formats(smil_url, broadcast_id)
-            if smil_formats:
-                formats.extend(smil_formats)
+            formats.extend(self._extract_smil_formats(smil_url, broadcast_id))
 
-        entry_protocol = 'm3u8' if is_live else 'm3u8_native'
         m3u8_url = stream_info.get('m3u8_url')
         if m3u8_url:
-            m3u8_formats = self._extract_m3u8_formats(
-                m3u8_url, broadcast_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False)
-            if m3u8_formats:
-                formats.extend(m3u8_formats)
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, broadcast_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         rtsp_url = stream_info.get('rtsp_url')
         if rtsp_url:
@@ -207,9 +204,10 @@ class LivestreamIE(InfoExtractor):
             if not videos_info:
                 break
             for v in videos_info:
+                v_id = compat_str(v['id'])
                 entries.append(self.url_result(
-                    'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v['id']),
-                    'Livestream', v['id'], v['caption']))
+                    'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v_id),
+                    'Livestream', v_id, v.get('caption')))
             last_video = videos_info[-1]['id']
         return self.playlist_result(entries, event_id, event_data['full_name'])
 
@@ -252,7 +250,7 @@ class LivestreamOriginalIE(InfoExtractor):
         'playlist_mincount': 4,
     }, {
         # live stream
-        'url': 'http://www.livestream.com/znsbahamas',
+        'url': 'http://original.livestream.com/znsbahamas',
         'only_matching': True,
     }]
 
@@ -279,7 +277,7 @@ class LivestreamOriginalIE(InfoExtractor):
             'view_count': view_count,
         }
 
-    def _extract_video_formats(self, video_data, video_id, entry_protocol):
+    def _extract_video_formats(self, video_data, video_id):
         formats = []
 
         progressive_url = video_data.get('progressiveUrl')
@@ -291,10 +289,9 @@ class LivestreamOriginalIE(InfoExtractor):
 
         m3u8_url = video_data.get('httpUrl')
         if m3u8_url:
-            m3u8_formats = self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False)
-            if m3u8_formats:
-                formats.extend(m3u8_formats)
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         rtsp_url = video_data.get('rtspUrl')
         if rtsp_url:
@@ -341,15 +338,14 @@ class LivestreamOriginalIE(InfoExtractor):
                 info = {
                     'title': self._og_search_title(webpage),
                     'description': self._og_search_description(webpage),
-                    'thumbnail': self._search_regex(r'channelLogo.src\s*=\s*"([^"]+)"', webpage, 'thumbnail', None),
+                    'thumbnail': self._search_regex(r'channelLogo\.src\s*=\s*"([^"]+)"', webpage, 'thumbnail', None),
                 }
             video_data = self._download_json(stream_url, content_id)
             is_live = video_data.get('isLive')
-            entry_protocol = 'm3u8' if is_live else 'm3u8_native'
             info.update({
                 'id': content_id,
                 'title': self._live_title(info['title']) if is_live else info['title'],
-                'formats': self._extract_video_formats(video_data, content_id, entry_protocol),
+                'formats': self._extract_video_formats(video_data, content_id),
                 'is_live': is_live,
             })
             return info
@@ -367,7 +363,4 @@ class LivestreamShortenerIE(InfoExtractor):
         id = mobj.group('id')
         webpage = self._download_webpage(url, id)
 
-        return {
-            '_type': 'url',
-            'url': self._og_search_url(webpage),
-        }
+        return self.url_result(self._og_search_url(webpage))