Start moving to ytdl-org

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

diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py
index f3825db5cc208d0fe8ae248b68198213e6ce82e1..d68f222e7b05b50a81f16b03ac4692ffe1b0d9d1 100644 (file)
--- a/youtube_dl/extractor/vlive.py
+++ b/youtube_dl/extractor/vlive.py
@@ -57,7 +57,7 @@ class VLiveIE(InfoExtractor):
         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'
@@ -108,11 +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'
             })
 
@@ -236,7 +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/ytdl-org/youtube-dl/issues/13830
+                    # 2. http://channels.vlive.tv/EDBF.
+                    'maxNumOfRows': 100,
                     '_': int(time.time()),
                     'pageNo': page_num
                 }
@@ -280,10 +285,8 @@ class VLivePlaylistIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        playlist_id = self._match_id(url)
-        video_id_match = re.match(self._VALID_URL, url)
-        assert video_id_match
-        video_id = compat_str(video_id_match.group('video_id'))
+        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'):
@@ -294,26 +297,27 @@ class VLivePlaylistIE(InfoExtractor):
                 ie=VLiveIE.ie_key(), video_id=video_id)
 
         self.to_screen(
-            'Downloading playlist %s - add --no-playlist to just download video' % playlist_id)
+            '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), video_id)
+            'http://www.vlive.tv/video/%s/playlist/%s'
+            % (video_id, playlist_id), playlist_id)
 
-        playlist_name = self._html_search_regex(
-            r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
-            webpage, 'playlist name', fatal=False)
+        item_ids = self._parse_json(
+            self._search_regex(
+                r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
+                'playlist video seqs'),
+            playlist_id)
 
-        item_ids = self._search_regex(
-            r'\bvar\s+playlistVideoSeqs\s*=\s*(\[[^]]+\])',
-            webpage, 'playlist item ids')
+        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]
 
-        entries = []
-        for item_id in self._parse_json(item_ids, playlist_id):
-            item_id = compat_str(item_id)
-            entries.append(
-                self.url_result(
-                    VIDEO_URL_TEMPLATE % item_id,
-                    ie=VLiveIE.ie_key(), video_id=item_id))
+        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)
+        return self.playlist_result(entries, playlist_id, playlist_name)