Merge branch 'vlive' of https://github.com/ping/youtube-dl into ping-vlive
[youtube-dl] / youtube_dl / extractor / crunchyroll.py
index 33a033a7f3175fec3b85725841581d2ac252033d..ce123482e8036bc50c39a65e6036ee8eb534b83a 100644 (file)
@@ -20,9 +20,11 @@ from ..utils import (
     ExtractorError,
     bytes_to_intlist,
     intlist_to_bytes,
+    int_or_none,
     remove_end,
     unified_strdate,
     urlencode_postdata,
+    xpath_text,
 )
 from ..aes import (
     aes_cbc_decrypt,
@@ -237,7 +239,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             webpage_url = 'http://www.' + mobj.group('url')
 
         webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
-        note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
+        note_m = self._html_search_regex(
+            r'<div class="showmedia-trailer-notice">(.+?)</div>',
+            webpage, 'trailer-notice', default='')
         if note_m:
             raise ExtractorError(note_m)
 
@@ -247,6 +251,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if msg.get('type') == 'error':
                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
 
+        if 'To view this, please log in to verify you are 18 or older.' in webpage:
+            self.raise_login_required()
+
         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
         video_title = re.sub(r' {2,}', ' ', video_title)
         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
@@ -281,6 +288,13 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             stream_info = streamdata.find('./{default}preload/stream_info')
             video_url = stream_info.find('./host').text
             video_play_path = stream_info.find('./file').text
+            metadata = stream_info.find('./metadata')
+            format_info = {
+                'format': video_format,
+                'format_id': video_format,
+                'height': int_or_none(xpath_text(metadata, './height')),
+                'width': int_or_none(xpath_text(metadata, './width')),
+            }
 
             if '.fplive.net/' in video_url:
                 video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
@@ -289,19 +303,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                     netloc='v.lvlt.crcdn.net',
                     path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
                 if self._is_valid_url(direct_video_url, video_id, video_format):
-                    formats.append({
+                    format_info.update({
                         'url': direct_video_url,
-                        'format_id': video_format,
                     })
+                    formats.append(format_info)
                     continue
 
-            formats.append({
+            format_info.update({
                 'url': video_url,
                 'play_path': video_play_path,
                 'ext': 'flv',
-                'format': video_format,
-                'format_id': video_format,
             })
+            formats.append(format_info)
 
         subtitles = self.extract_subtitles(video_id, webpage)