New upstream version 2018.11.07

[youtubedl] / youtube_dl / extractor / vlive.py

diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py
index b9718901b8339e3d5fee85a15d20514b038b66aa..0b5165fd03d8837a2954f864350a642b47176aff 100644 (file)
--- a/youtube_dl/extractor/vlive.py
+++ b/youtube_dl/extractor/vlive.py
@@ -49,11 +49,15 @@ class VLiveIE(InfoExtractor):
         },
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
-            'http://www.vlive.tv/video/%s' % video_id, video_id)
+            'https://www.vlive.tv/video/%s' % video_id, video_id)
 
         VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
         VIDEO_PARAMS_FIELD = 'video params'
@@ -70,9 +74,9 @@ class VLiveIE(InfoExtractor):
         status, long_video_id, key = params[2], params[5], params[6]
         status = remove_start(status, 'PRODUCT_')
 
-        if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
+        if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'):
             return self._live(video_id, webpage)
-        elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
+        elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'):
             if long_video_id and key:
                 return self._replay(video_id, webpage, long_video_id, key)
             else:
@@ -104,11 +108,11 @@ class VLiveIE(InfoExtractor):
 
     def _live(self, video_id, webpage):
         init_page = self._download_webpage(
-            'http://www.vlive.tv/video/init/view',
+            'https://www.vlive.tv/video/init/view',
             video_id, note='Downloading live webpage',
             data=urlencode_postdata({'videoSeq': video_id}),
             headers={
-                'Referer': 'http://www.vlive.tv/video/%s' % video_id,
+                'Referer': 'https://www.vlive.tv/video/%s' % video_id,
                 'Content-Type': 'application/x-www-form-urlencoded'
             })
 
@@ -232,7 +236,12 @@ class VLiveChannelIE(InfoExtractor):
                 query={
                     'app_id': app_id,
                     'channelSeq': channel_seq,
-                    'maxNumOfRows': 1000,
+                    # Large values of maxNumOfRows (~300 or above) may cause
+                    # empty responses (see [1]), e.g. this happens for [2] that
+                    # has more than 300 videos.
+                    # 1. https://github.com/rg3/youtube-dl/issues/13830
+                    # 2. http://channels.vlive.tv/EDBF.
+                    'maxNumOfRows': 100,
                     '_': int(time.time()),
                     'pageNo': page_num
                 }
@@ -261,3 +270,54 @@ class VLiveChannelIE(InfoExtractor):
 
         return self.playlist_result(
             entries, channel_code, channel_name)
+
+
+class VLivePlaylistIE(InfoExtractor):
+    IE_NAME = 'vlive:playlist'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<video_id>[0-9]+)/playlist/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.vlive.tv/video/22867/playlist/22912',
+        'info_dict': {
+            'id': '22912',
+            'title': 'Valentine Day Message from TWICE'
+        },
+        'playlist_mincount': 9
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id, playlist_id = mobj.group('video_id', 'id')
+
+        VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s'
+        if self._downloader.params.get('noplaylist'):
+            self.to_screen(
+                'Downloading just video %s because of --no-playlist' % video_id)
+            return self.url_result(
+                VIDEO_URL_TEMPLATE % video_id,
+                ie=VLiveIE.ie_key(), video_id=video_id)
+
+        self.to_screen(
+            'Downloading playlist %s - add --no-playlist to just download video'
+            % playlist_id)
+
+        webpage = self._download_webpage(
+            'http://www.vlive.tv/video/%s/playlist/%s'
+            % (video_id, playlist_id), playlist_id)
+
+        item_ids = self._parse_json(
+            self._search_regex(
+                r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
+                'playlist video seqs'),
+            playlist_id)
+
+        entries = [
+            self.url_result(
+                VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
+                video_id=compat_str(item_id))
+            for item_id in item_ids]
+
+        playlist_name = self._html_search_regex(
+            r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
+            webpage, 'playlist title', fatal=False)
+
+        return self.playlist_result(entries, playlist_id, playlist_name)