Fine tune the changelog.

[youtubedl] / youtube_dl / extractor / vlive.py

diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py
index 86c1cb5ef5d352608416a653bea5ef9f74c4287a..9e2aa58bd94270aee79800ec2d4d4ac18bad28e8 100644 (file)
--- a/youtube_dl/extractor/vlive.py
+++ b/youtube_dl/extractor/vlive.py
@@ -1,86 +1,88 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

-import hmac
-from hashlib import sha1
-from base64 import b64encode
-from time import time
-

 from .common import InfoExtractor
 from ..utils import (
 from .common import InfoExtractor
 from ..utils import (

-    ExtractorError,
-    determine_ext
+    dict_get,
+    float_or_none,
+    int_or_none,

 )
 from ..compat import compat_urllib_parse
 
 
 class VLiveIE(InfoExtractor):
     IE_NAME = 'vlive'
 )
 from ..compat import compat_urllib_parse
 
 
 class VLiveIE(InfoExtractor):
     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]+)'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'

     _TEST = {
     _TEST = {

-        'url': 'http://m.vlive.tv/video/1326',
+        'url': 'http://www.vlive.tv/video/1326',

         'md5': 'cc7314812855ce56de70a06a27314983',
         'info_dict': {
             'id': '1326',
             'ext': 'mp4',
         'md5': 'cc7314812855ce56de70a06a27314983',
         'info_dict': {
             'id': '1326',
             'ext': 'mp4',

-            'title': '[V] Girl\'s Day\'s Broadcast',
-            'creator': 'Girl\'s Day',
+            'title': "[V] Girl's Day's Broadcast",
+            'creator': "Girl's Day",
+            'view_count': int,

         },
     }
         },
     }

-    _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH'

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(

-            'http://m.vlive.tv/video/%s' % video_id,
-            video_id, note='Download video page')
+            'http://www.vlive.tv/video/%s' % video_id, video_id)

 
 

-        title = self._og_search_title(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
-        creator = self._html_search_regex(
-            r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator')
+        long_video_id = self._search_regex(
+            r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"',
+            webpage, 'long video id')

 
 

-        url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
-        msgpad = '%.0f' % (time() * 1000)
-        md = b64encode(
-            hmac.new(self._SECRET.encode('ascii'),
-                     (url[:255] + msgpad).encode('ascii'), sha1).digest()
-        )
-        url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md})
-        playinfo = self._download_json(url, video_id, 'Downloading video json')
+        key = self._search_regex(
+            r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"',
+            webpage, 'key')

 
 

-        if playinfo.get('message', '') != 'success':
-            raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful'))
+        title = self._og_search_title(webpage)

 
 

-        if not playinfo.get('result'):
-            raise ExtractorError('No videos found.')
+        playinfo = self._download_json(
+            'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
+            % compat_urllib_parse.urlencode({
+                'videoId': long_video_id,
+                'key': key,
+                'ptc': 'http',
+                'doct': 'json',  # document type (xml or json)
+                'cpt': 'vtt',  # captions type (vtt or ttml)
+            }), video_id)

 
 

-        formats = []
-        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'),
-            })
+        formats = [{
+            'url': vid['source'],
+            'format_id': vid.get('encodingOption', {}).get('name'),
+            'abr': float_or_none(vid.get('bitrate', {}).get('audio')),
+            'vbr': float_or_none(vid.get('bitrate', {}).get('video')),
+            'width': int_or_none(vid.get('encodingOption', {}).get('width')),
+            'height': int_or_none(vid.get('encodingOption', {}).get('height')),
+            'filesize': int_or_none(vid.get('size')),
+        } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]

         self._sort_formats(formats)
 
         self._sort_formats(formats)
 

+        thumbnail = self._og_search_thumbnail(webpage)
+        creator = self._html_search_regex(
+            r'<div[^>]+class="info_area"[^>]*>\s*<strong[^>]+class="name"[^>]*>([^<]+)</strong>',
+            webpage, 'creator', fatal=False)
+
+        view_count = int_or_none(playinfo.get('meta', {}).get('count'))
+

         subtitles = {}
         subtitles = {}

-        for caption in playinfo['result'].get('captions', {}).get('list', []):
-            subtitles[caption['language']] = [
-                {'ext': determine_ext(caption['source'], default_ext='vtt'),
-                 'url': caption['source']}]
+        for caption in playinfo.get('captions', {}).get('list', []):
+            lang = dict_get(caption, ('language', 'locale', 'country', 'label'))
+            if lang and caption.get('source'):
+                subtitles[lang] = [{
+                    'ext': 'vtt',
+                    'url': caption['source']}]

 
         return {
             'id': video_id,
             'title': title,
             'creator': creator,
             'thumbnail': thumbnail,
 
         return {
             'id': video_id,
             'title': title,
             'creator': creator,
             'thumbnail': thumbnail,

+            'view_count': view_count,

             'formats': formats,
             'subtitles': subtitles,
         }
             'formats': formats,
             'subtitles': subtitles,
         }