[naver] extract subtitles and hls formats and reuse code in VLiveIE

youtube_dl/extractor/naver.py

@@ -84,11 +84,17 @@ class NaverIE(InfoExtractor):
         },
     }]
 
-    def _extract_video_formats(self, formats_list):
+    def _extract_video_formats(self, formats_list, vid):
         formats = []
         for format_el in formats_list:
             url = format_el.get('source')
             if url:
+                if format_el.get('type') == 'HLS':
+                    key = format_el.get('key')
+                    if key:
+                        url += '?%s=%s' % (key['name'], key['value'])
+                    formats.extend(self._extract_m3u8_formats(url, vid, 'mp4', m3u8_id='hls'))
+                else:
                     encoding_option = format_el.get('encodingOption')
                     bitrate = format_el.get('bitrate')
                     formats.append({
@@ -106,27 +112,25 @@ class NaverIE(InfoExtractor):
             self._sort_formats(formats)
         return formats
 
-    def _extract_video_info(self, vid, key):
+    def _parse_video_info(self, play_data, vid):
-        play_data = self._download_json(
-            'http://global.apis.naver.com/linetv/rmcnmv/vod_play_videoInfo.json?' + compat_urllib_parse.urlencode({'videoId': vid, 'key': key}),
-            vid, 'Downloading video info')
         meta = play_data.get('meta')
-        user = meta.get('user')
+        user = meta.get('user', {})
 
         thumbnails = []
-        for thumbnail in play_data['thumbnails']['list']:
+        for thumbnail in play_data.get('thumbnails', {}).get('list', []):
             thumbnails.append({'url': thumbnail['source']})
 
-        formats = self._extract_video_formats(play_data['videos']['list'])
+        subtitles = {}
-        if not formats:
+        for caption in play_data.get('captions', {}).get('list', []):
-            video_info = self._download_json(
+            subtitles[caption['language']] = [
-                'http://serviceapi.rmcnmv.naver.com/mobile/getVideoInfo.nhn?' + compat_urllib_parse.urlencode({'videoId': vid, 'inKey': key, 'protocol': 'http'}),
+                {'ext': determine_ext(caption['source'], default_ext='vtt'),
-                vid, 'Downloading video info')
+                 'url': caption['source']}]
-            formats = self._extract_video_formats(video_info['videos']['list'])
+
+        formats = self._extract_video_formats(play_data['videos']['list'] + play_data.get('streams', []), vid)
 
         return {
             'id': vid,
-            'title': meta['subject'],
+            'title': meta.get('subject'),
             'formats': formats,
             'thumbnail': meta.get('cover', {}).get('source'),
             'thumbnails': thumbnails,
@@ -135,6 +139,18 @@ class NaverIE(InfoExtractor):
             'uploader': user.get('name'),
         }
 
+    def _extract_video_info(self, vid, key):
+        play_data = self._download_json(
+            'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?' + compat_urllib_parse.urlencode({'videoId': vid, 'key': key}),
+            vid, 'Downloading video info')
+        info = self._parse_video_info(play_data, vid)
+        if not info['formats']:
+            play_data = self._download_json(
+                'http://serviceapi.rmcnmv.naver.com/mobile/getVideoInfo.nhn?' + compat_urllib_parse.urlencode({'videoId': vid, 'inKey': key, 'protocol': 'http'}),
+                vid, 'Downloading video info')
+            info['formats'] = self._extract_video_formats(play_data['videos']['list'] + play_data.get('streams', []), vid)
+        return info
+
     def _extract_id_and_key(self, webpage):
         m_id = re.search(r'(?s)new\s+nhn.rmcnmv.RMCVideoPlayer\(\s*["\']([^"\']+)["\']\s*,\s*(?:{[^}]*?value[^:]*?:\s*?)?["\']([^"\']+)["\']', webpage)
         if not m_id:

youtube_dl/extractor/vlive.py

@@ -6,19 +6,19 @@ from hashlib import sha1
 from base64 import b64encode
 from time import time
 
-from .common import InfoExtractor
+from .naver import NaverIE
 from ..utils import (
     ExtractorError,
-    determine_ext
+    int_or_none,
 )
 from ..compat import compat_urllib_parse
 
 
-class VLiveIE(InfoExtractor):
+class VLiveIE(NaverIE):
     IE_NAME = 'vlive'
     # www.vlive.tv/video/ links redirect to m.vlive.tv/video/ for mobile devices
     _VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://m.vlive.tv/video/1326',
         'md5': 'cc7314812855ce56de70a06a27314983',
         'info_dict': {
@@ -27,21 +27,31 @@ class VLiveIE(InfoExtractor):
             'title': '[V] Girl\'s Day\'s Broadcast',
             'creator': 'Girl\'s Day',
         },
-    }
+    }]
     _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH'
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        status = self._download_json(
+            'http://www.vlive.tv/video/status?videoSeq=%s' % video_id,
+            video_id, note='Download status metadata')
+
+        vid = status.get('vodId')
+        if vid:
+            key = status.get('vodInKey')
+            if not key:
+                key = self._download_webpage('http://www.vlive.tv/video/inkey?vodId=%s' % vid, video_id)
+            if key:
+                video_info = self._extract_video_info(vid, key)
+        elif status['status'] not in ('CANCELED', 'COMING_SOON', 'NOT_FOUND'):
             webpage = self._download_webpage(
                 'http://m.vlive.tv/video/%s' % video_id,
                 video_id, note='Download video page')
-
             title = self._og_search_title(webpage)
             thumbnail = self._og_search_thumbnail(webpage)
             creator = self._html_search_regex(
                 r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator')
-
             url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
             msgpad = '%.0f' % (time() * 1000)
             md = b64encode(
@@ -50,37 +60,22 @@ class VLiveIE(InfoExtractor):
             )
             url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md})
             playinfo = self._download_json(url, video_id, 'Downloading video json')
-
             if playinfo.get('message', '') != 'success':
                 raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful'))
-
+            result = playinfo.get('result')
-        if not playinfo.get('result'):
+            if not result:
                 raise ExtractorError('No videos found.')
-
+            video_info = self._parse_video_info(result, video_id)
-        formats = []
+            video_info.update({
-        for vid in playinfo['result'].get('videos', {}).get('list', []):
-            formats.append({
-                'url': vid['source'],
-                'ext': 'mp4',
-                'abr': vid.get('bitrate', {}).get('audio'),
-                'vbr': vid.get('bitrate', {}).get('video'),
-                'format_id': vid['encodingOption']['name'],
-                'height': vid.get('height'),
-                'width': vid.get('width'),
-            })
-        self._sort_formats(formats)
-
-        subtitles = {}
-        for caption in playinfo['result'].get('captions', {}).get('list', []):
-            subtitles[caption['language']] = [
-                {'ext': determine_ext(caption['source'], default_ext='vtt'),
-                 'url': caption['source']}]
-
-        return {
-            'id': video_id,
                 'title': title,
-            'creator': creator,
                 'thumbnail': thumbnail,
-            'formats': formats,
+                'creator': creator,
-            'subtitles': subtitles,
+            })
-        }
+        if video_info:
+            video_info.update({
+                'id': video_id,
+                'view_count': int_or_none(status.get('playCount')),
+                'likes': int_or_none(status.get('likeCount')),
+            })
+            return video_info
+        raise ExtractorError(status['status'])