[viki] Enhance error message handling (#3774)

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

diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py
index 944901e1482a666ae90cc5e1c0f86e325ec2aecc..0fc1ceb19c8ece086e683da59e720489c80edbba 100644 (file)
--- a/youtube_dl/extractor/viki.py
+++ b/youtube_dl/extractor/viki.py
@@ -2,20 +2,28 @@ from __future__ import unicode_literals
 
 import re
 
+from ..compat import (
+    compat_urlparse,
+    compat_urllib_request,
+)
 from ..utils import (
     ExtractorError,
     unescapeHTML,
     unified_strdate,
     US_RATINGS,
+    clean_html,
 )
-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
 
 
-class VikiIE(SubtitlesInfoExtractor):
+class VikiIE(InfoExtractor):
     IE_NAME = 'viki'
 
+    # iPad2
+    _USER_AGENT = 'Mozilla/5.0(iPad; U; CPU OS 4_3 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8F191 Safari/6533.18.5'
+
     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
         'info_dict': {
             'id': '1023585v',
@@ -27,7 +35,17 @@ class VikiIE(SubtitlesInfoExtractor):
             'age_limit': 13,
         },
         'skip': 'Blocked in the US',
-    }
+    }, {
+        'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
+        'md5': 'ca6493e6f0a6ec07da9aa8d6304b4b2c',
+        'info_dict': {
+            'id': '1067139v',
+            'ext': 'mp4',
+            'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
+            'upload_date': '20150430',
+            'title': '\'The Avengers: Age of Ultron\' Press Conference',
+        }
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -49,13 +67,20 @@ class VikiIE(SubtitlesInfoExtractor):
             'rating information', default='').strip()
         age_limit = US_RATINGS.get(rating_str)
 
-        info_url = 'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id
+        req = compat_urllib_request.Request(
+            'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id)
+        req.add_header('User-Agent', self._USER_AGENT)
         info_webpage = self._download_webpage(
-            info_url, video_id, note='Downloading info page')
-        if re.match(r'\s*<div\s+class="video-error', info_webpage):
-            raise ExtractorError(
-                'Video %s is blocked from your location.' % video_id,
-                expected=True)
+            req, video_id, note='Downloading info page')
+        err_msg = self._html_search_regex(r'<div[^>]+class="video-error[^>]+>(.+)</div>', info_webpage, 'error message', default=None)
+        if err_msg:
+            err_msg = clean_html(err_msg)
+            if 'not available in your region' in err_msg:
+                raise ExtractorError(
+                    'Video %s is blocked from your location.' % video_id,
+                    expected=True)
+            else:
+                raise ExtractorError('Viki said: ' + err_msg)
         video_url = self._html_search_regex(
             r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
 
@@ -69,9 +94,6 @@ class VikiIE(SubtitlesInfoExtractor):
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, info_webpage)
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, info_webpage)
-            return
 
         return {
             'id': video_id,
@@ -85,12 +107,15 @@ class VikiIE(SubtitlesInfoExtractor):
             'upload_date': upload_date,
         }
 
-    def _get_available_subtitles(self, video_id, info_webpage):
+    def _get_subtitles(self, video_id, info_webpage):
         res = {}
-        for sturl_html in re.findall(r'<track src="([^"]+)"/>', info_webpage):
+        for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
             sturl = unescapeHTML(sturl_html)
             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
             if not m:
                 continue
-            res[m.group('lang')] = sturl
+            res[m.group('lang')] = [{
+                'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
+                'ext': 'vtt',
+            }]
         return res