[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / ondemandkorea.py
index 125c310c8c33c80ba4e4130257c46ca35a019616..df1ce3c1db1eaa22d03609ddb55748e404b1f4a9 100644 (file)
@@ -1,22 +1,23 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    js_to_json,
+)
 
 
 class OnDemandKoreaIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?ondemandkorea\.com/(?P<id>[^/]+)\.html'
+    _GEO_COUNTRIES = ['US', 'CA']
     _TEST = {
         'url': 'http://www.ondemandkorea.com/ask-us-anything-e43.html',
         'info_dict': {
             'id': 'ask-us-anything-e43',
             'ext': 'mp4',
             'title': 'Ask Us Anything : E43',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
         },
         'params': {
             'skip_download': 'm3u8 download'
@@ -29,30 +30,33 @@ class OnDemandKoreaIE(InfoExtractor):
 
         if not webpage:
             # Page sometimes returns captcha page with HTTP 403
-            raise ExtractorError('Unable to access page. You may have been blocked.', expected=True)
+            raise ExtractorError(
+                'Unable to access page. You may have been blocked.',
+                expected=True)
 
         if 'msg_block_01.png' in webpage:
-            raise ExtractorError('This content is not available in your region.', expected=True)
-        
+            self.raise_geo_restricted(
+                msg='This content is not available in your region',
+                countries=self._GEO_COUNTRIES)
+
         if 'This video is only available to ODK PLUS members.' in webpage:
-            raise ExtractorError('This video is only available to ODK PLUS members.', expected=True)
+            raise ExtractorError(
+                'This video is only available to ODK PLUS members.',
+                expected=True)
 
         title = self._og_search_title(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
 
-        manifest_url = self._search_regex(r'file:\s"(https?://[\S].+?/manifest\.m3u8)', webpage, 'manifest')
-        formats = self._extract_m3u8_formats(manifest_url, video_id, 'mp4', m3u8_id='hls')
-        self._sort_formats(formats)
+        jw_config = self._parse_json(
+            self._search_regex(
+                r'(?s)jwplayer\(([\'"])(?:(?!\1).)+\1\)\.setup\s*\((?P<options>.+?)\);',
+                webpage, 'jw config', group='options'),
+            video_id, transform_source=js_to_json)
+        info = self._parse_jwplayer_data(
+            jw_config, video_id, require_title=False, m3u8_id='hls',
+            base_url=url)
 
-        subs = re.findall(r'file:\s\'(?P<file>[^\']+\.vtt)\',\s+label:\s+\'(?P<lang>[^\']+)\'', webpage)
-        subtitles = {}
-        for sub in subs:
-            subtitles[sub[1]] = [{'url': 'http://www.ondemandkorea.com' + sub[0], 'ext': sub[0][-3:]}]
-
-        return {
-            'id': video_id,
+        info.update({
             'title': title,
-            'thumbnail': thumbnail,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+            'thumbnail': self._og_search_thumbnail(webpage),
+        })
+        return info