Merge pull request #8611 from remitamine/ffmpegfd

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

diff --git a/youtube_dl/extractor/vgtv.py b/youtube_dl/extractor/vgtv.py
index 347410a78c2c3f49da2812070e9ac7d3b5943ed2..e148b1ef513321376efe1795056503ea2a8bcad8 100644 (file)
--- a/youtube_dl/extractor/vgtv.py
+++ b/youtube_dl/extractor/vgtv.py
@@ -20,6 +20,7 @@ class VGTVIE(XstreamIE):
         'aftenbladet.no/tv': 'satv',
         'fvn.no/fvntv': 'fvntv',
         'aftenposten.no/webtv': 'aptv',
+        'ap.vgtv.no/webtv': 'aptv',
     }
 
     _APP_NAME_TO_VENDOR = {
@@ -35,7 +36,7 @@ class VGTVIE(XstreamIE):
                     (?P<host>
                         %s
                     )
-                    /
+                    /?
                     (?:
                         \#!/(?:video|live)/|
                         embed?.*id=
@@ -104,7 +105,7 @@ class VGTVIE(XstreamIE):
         },
         {
             'url': 'http://www.aftenposten.no/webtv/#!/video/21039/trailer-sweatshop-i-can-t-take-any-more',
-            'md5': '7fbc265a3ca4933a423c7a66aa879a67',
+            'md5': 'fd828cd29774a729bf4d4425fe192972',
             'info_dict': {
                 'id': '21039',
                 'ext': 'mp4',
@@ -114,12 +115,20 @@ class VGTVIE(XstreamIE):
                 'timestamp': 1417002452,
                 'upload_date': '20141126',
                 'view_count': int,
-            }
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
         },
         {
             'url': 'http://www.bt.no/tv/#!/video/100250/norling-dette-er-forskjellen-paa-1-divisjon-og-eliteserien',
             'only_matching': True,
         },
+        {
+            'url': 'http://ap.vgtv.no/webtv#!/video/111084/de-nye-bysyklene-lettere-bedre-gir-stoerre-hjul-og-feste-til-mobil',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):
@@ -144,8 +153,6 @@ class VGTVIE(XstreamIE):
         if len(video_id) == 5:
             if appname == 'bttv':
                 info = self._extract_video_info('btno', video_id)
-            elif appname == 'aptv':
-                info = self._extract_video_info('ap', video_id)
 
         streams = data['streamUrls']
         stream_type = data.get('streamType')
@@ -154,18 +161,19 @@ class VGTVIE(XstreamIE):
 
         hls_url = streams.get('hls')
         if hls_url:
-            m3u8_formats = self._extract_m3u8_formats(
-                hls_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
-            if m3u8_formats:
-                formats.extend(m3u8_formats)
+            formats.extend(self._extract_m3u8_formats(
+                hls_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 
         hds_url = streams.get('hds')
-        # wasLive hds are always 404
-        if hds_url and stream_type != 'wasLive':
+        if hds_url:
+            hdcore_sign = 'hdcore=3.7.0'
             f4m_formats = self._extract_f4m_formats(
-                hds_url + '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18', video_id, f4m_id='hds', fatal=False)
+                hds_url + '?%s' % hdcore_sign, video_id, f4m_id='hds', fatal=False)
             if f4m_formats:
-                formats.extend(f4m_formats)
+                for entry in f4m_formats:
+                    # URLs without the extra param induce an 404 error
+                    entry.update({'extra_param_to_segment_url': hdcore_sign})
+                    formats.append(entry)
 
         mp4_urls = streams.get('pseudostreaming') or []
         mp4_url = streams.get('mp4')
@@ -174,16 +182,15 @@ class VGTVIE(XstreamIE):
         for mp4_url in mp4_urls:
             format_info = {
                 'url': mp4_url,
-                'preference': 1,
             }
             mobj = re.search('(\d+)_(\d+)_(\d+)', mp4_url)
             if mobj:
-                vbr = int(mobj.group(3))
+                tbr = int(mobj.group(3))
                 format_info.update({
                     'width': int(mobj.group(1)),
                     'height': int(mobj.group(2)),
-                    'vbr': vbr,
-                    'format_id': 'mp4-%s' % vbr,
+                    'tbr': tbr,
+                    'format_id': 'mp4-%s' % tbr,
                 })
             formats.append(format_info)
 
@@ -210,7 +217,7 @@ class BTArticleIE(InfoExtractor):
     _VALID_URL = 'http://(?:www\.)?bt\.no/(?:[^/]+/)+(?P<id>[^/]+)-\d+\.html'
     _TEST = {
         'url': 'http://www.bt.no/nyheter/lokalt/Kjemper-for-internatet-1788214.html',
-        'md5': 'd055e8ee918ef2844745fcfd1a4175fb',
+        'md5': '2acbe8ad129b3469d5ae51b1158878df',
         'info_dict': {
             'id': '23199',
             'ext': 'mp4',
@@ -227,7 +234,7 @@ class BTArticleIE(InfoExtractor):
     def _real_extract(self, url):
         webpage = self._download_webpage(url, self._match_id(url))
         video_id = self._search_regex(
-            r'SVP\.Player\.load\(\s*(\d+)', webpage, 'video id')
+            r'<video[^>]+data-id="(\d+)"', webpage, 'video id')
         return self.url_result('bttv:%s' % video_id, 'VGTV')
 
 
@@ -235,7 +242,7 @@ class BTVestlendingenIE(InfoExtractor):
     IE_NAME = 'bt:vestlendingen'
     IE_DESC = 'Bergens Tidende - Vestlendingen'
     _VALID_URL = 'http://(?:www\.)?bt\.no/spesial/vestlendingen/#!/(?P<id>\d+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.bt.no/spesial/vestlendingen/#!/86588',
         'md5': 'd7d17e3337dc80de6d3a540aefbe441b',
         'info_dict': {
@@ -246,7 +253,19 @@ class BTVestlendingenIE(InfoExtractor):
             'timestamp': 1430473209,
             'upload_date': '20150501',
         },
-    }
+        'skip': '404 Error',
+    }, {
+        'url': 'http://www.bt.no/spesial/vestlendingen/#!/86255',
+        'md5': 'a2893f8632e96389f4bdf36aa9463ceb',
+        'info_dict': {
+            'id': '86255',
+            'ext': 'mov',
+            'title': 'Du må tåle å fryse og være sulten',
+            'description': 'md5:b8046f4d022d5830ddab04865791d063',
+            'upload_date': '20150321',
+            'timestamp': 1426942023,
+        },
+    }]
 
     def _real_extract(self, url):
         return self.url_result('bttv:%s' % self._match_id(url), 'VGTV')