[afreecatv] extract rtmp formats
[youtube-dl] / youtube_dl / extractor / afreecatv.py
index 4ebc61bae1f464c17dc4fe1037f8ed5839ce9247..e0a0f7c57b83c7a715e7a39c16a8a71df1cbb500 100644 (file)
@@ -11,18 +11,27 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
+    update_url_query,
+    xpath_element,
     xpath_text,
 )
 
 
 class AfreecaTVIE(InfoExtractor):
+    IE_NAME = 'afreecatv'
     IE_DESC = 'afreecatv.com'
-    _VALID_URL = r'''(?x)^
-        https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
-        (?:
-            /app/(?:index|read_ucc_bbs)\.cgi|
-            /player/[Pp]layer\.(?:swf|html))
-        \?.*?\bnTitleNo=(?P<id>\d+)'''
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
+                            (?:
+                                /app/(?:index|read_ucc_bbs)\.cgi|
+                                /player/[Pp]layer\.(?:swf|html)
+                            )\?.*?\bnTitleNo=|
+                            vod\.afreecatv\.com/PLAYER/STATION/
+                        )
+                        (?P<id>\d+)
+                    '''
     _TESTS = [{
         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
@@ -30,7 +39,7 @@ class AfreecaTVIE(InfoExtractor):
             'id': '36164052',
             'ext': 'mp4',
             'title': '데일리 에이프릴 요정들의 시상식!',
-            'thumbnail': 're:^https?://videoimg.afreecatv.com/.*$',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
             'upload_date': '20160503',
@@ -40,7 +49,7 @@ class AfreecaTVIE(InfoExtractor):
         'info_dict': {
             'id': '36153164',
             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
-            'thumbnail': 're:^https?://videoimg.afreecatv.com/.*$',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
         },
@@ -62,6 +71,12 @@ class AfreecaTVIE(InfoExtractor):
                 'upload_date': '20160502',
             },
         }],
+    }, {
+        'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
+        'only_matching': True,
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -79,11 +94,14 @@ class AfreecaTVIE(InfoExtractor):
         info_url = compat_urlparse.urlunparse(parsed_url._replace(
             netloc='afbbs.afreecatv.com:8080',
             path='/api/video/get_video_info.php'))
-        video_xml = self._download_xml(info_url, video_id)
 
-        if xpath_text(video_xml, './track/flag', default='FAIL') != 'SUCCEED':
+        video_xml = self._download_xml(
+            update_url_query(info_url, {'nTitleNo': video_id}), video_id)
+
+        if xpath_element(video_xml, './track/video/file') is None:
             raise ExtractorError('Specified AfreecaTV video does not exist',
                                  expected=True)
+
         title = xpath_text(video_xml, './track/title', 'title')
         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
@@ -93,7 +111,9 @@ class AfreecaTVIE(InfoExtractor):
 
         entries = []
         for i, video_file in enumerate(video_xml.findall('./track/video/file')):
-            video_key = self.parse_video_key(video_file.get('key'))
+            video_key = self.parse_video_key(video_file.get('key', ''))
+            if not video_key:
+                continue
             entries.append({
                 'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
                 'title': title,
@@ -116,7 +136,7 @@ class AfreecaTVIE(InfoExtractor):
             info['entries'] = entries
         elif len(entries) == 1:
             info['url'] = entries[0]['url']
-            info['upload_date'] = entries[0]['upload_date']
+            info['upload_date'] = entries[0].get('upload_date')
         else:
             raise ExtractorError(
                 'No files found for the specified AfreecaTV video, either'
@@ -124,3 +144,107 @@ class AfreecaTVIE(InfoExtractor):
                 expected=True)
 
         return info
+
+
+class AfreecaTVGlobalIE(AfreecaTVIE):
+    IE_NAME = 'afreecatv:global'
+    _VALID_URL = r'https?://(?:www\.)?afreeca\.tv/(?P<channel_id>\d+)(?:/v/(?P<video_id>\d+))?'
+    _TESTS = [{
+        'url': 'http://afreeca.tv/36853014/v/58301',
+        'info_dict': {
+            'id': '58301',
+            'title': 'tryhard top100',
+            'uploader_id': '36853014',
+            'uploader': 'makgi Hearthstone Live!',
+        },
+        'playlist_count': 3,
+    }]
+
+    def _real_extract(self, url):
+        channel_id, video_id = re.match(self._VALID_URL, url).groups()
+        video_type = 'video' if video_id else 'live'
+        query = {
+            'pt': 'view',
+            'bid': channel_id,
+        }
+        if video_id:
+            query['vno'] = video_id
+        video_data = self._download_json(
+            'http://api.afreeca.tv/%s/view_%s.php' % (video_type, video_type),
+            video_id or channel_id, query=query)['channel']
+
+        if video_data.get('result') != 1:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg']))
+
+        title = video_data['title']
+
+        info = {
+            'thumbnail': video_data.get('thumb'),
+            'view_count': int_or_none(video_data.get('vcnt')),
+            'age_limit': int_or_none(video_data.get('grade')),
+            'uploader_id': channel_id,
+            'uploader': video_data.get('cname'),
+        }
+
+        if video_id:
+            entries = []
+            for i, f in enumerate(video_data.get('flist', [])):
+                video_key = self.parse_video_key(f.get('key', ''))
+                f_url = f.get('file')
+                if not video_key or not f_url:
+                    continue
+                entries.append({
+                    'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
+                    'title': title,
+                    'upload_date': video_key.get('upload_date'),
+                    'duration': int_or_none(f.get('length')),
+                    'url': f_url,
+                    'protocol': 'm3u8_native',
+                    'ext': 'mp4',
+                })
+
+            info.update({
+                'id': video_id,
+                'title': title,
+                'duration': int_or_none(video_data.get('length')),
+            })
+            if len(entries) > 1:
+                info['_type'] = 'multi_video'
+                info['entries'] = entries
+            elif len(entries) == 1:
+                i = entries[0].copy()
+                i.update(info)
+                info = i
+        else:
+            formats = []
+            for s in video_data.get('strm', []):
+                s_url = s.get('purl')
+                if not s_url:
+                    continue
+                stype = s.get('stype')
+                if stype == 'HLS':
+                    formats.extend(self._extract_m3u8_formats(
+                        s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False))
+                elif stype == 'RTMP':
+                    format_id = [stype]
+                    label = s.get('label')
+                    if label:
+                        format_id.append(label)
+                    formats.append({
+                        'format_id': '-'.join(format_id),
+                        'url': s_url,
+                        'tbr': int_or_none(s.get('bps')),
+                        'height': int_or_none(s.get('brt')),
+                        'ext': 'flv',
+                        'rtmp_live': True,
+                    })
+            self._sort_formats(formats)
+
+            info.update({
+                'id': channel_id,
+                'title': self._live_title(title),
+                'is_live': True,
+                'formats': formats,
+            })
+
+        return info