[videomega] Fix extraction and update test (Fixes #5235)

youtube_dl/extractor/videomega.py

@@ -4,28 +4,21 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
+from ..compat import compat_urllib_request
-    compat_urllib_parse,
-    compat_urllib_request,
-)
-from ..utils import (
-    ExtractorError,
-    remove_start,
-)
 
 
 class VideoMegaIE(InfoExtractor):
     _VALID_URL = r'''(?x)https?://
         (?:www\.)?videomega\.tv/
-        (?:iframe\.php)?\?ref=(?P<id>[A-Za-z0-9]+)
+        (?:iframe\.php|cdn\.php)?\?ref=(?P<id>[A-Za-z0-9]+)
         '''
     _TEST = {
-        'url': 'http://videomega.tv/?ref=QR0HCUHI1661IHUCH0RQ',
+        'url': 'http://videomega.tv/?ref=4GNA688SU99US886ANG4',
         'md5': 'bf5c2f95c4c917536e80936af7bc51e1',
         'info_dict': {
-            'id': 'QR0HCUHI1661IHUCH0RQ',
+            'id': '4GNA688SU99US886ANG4',
             'ext': 'mp4',
-            'title': 'Big Buck Bunny',
+            'title': 'BigBuckBunny_320x180',
             'thumbnail': 're:^https?://.*\.jpg$',
         }
     }
@@ -33,34 +26,24 @@ class VideoMegaIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        iframe_url = 'http://videomega.tv/iframe.php?ref={0:}'.format(video_id)
+        iframe_url = 'http://videomega.tv/cdn.php?ref=%s' % video_id
         req = compat_urllib_request.Request(iframe_url)
         req.add_header('Referer', url)
         webpage = self._download_webpage(req, video_id)
 
-        try:
+        title = self._html_search_regex(
-            escaped_data = re.findall(r'unescape\("([^"]+)"\)', webpage)[-1]
+            r'<title>(.*?)</title>', webpage, 'title')
-        except IndexError:
+        title = re.sub(
-            raise ExtractorError('Unable to extract escaped data')
+            r'(?:^[Vv]ideo[Mm]ega\.tv\s-\s?|\s?-\svideomega\.tv$)', '', title)
-
-        playlist = compat_urllib_parse.unquote(escaped_data)
-
         thumbnail = self._search_regex(
-            r'image:\s*"([^"]+)"', playlist, 'thumbnail', fatal=False)
+            r'<video[^>]+?poster="([^"]+)"', webpage, 'thumbnail', fatal=False)
-        video_url = self._search_regex(r'file:\s*"([^"]+)"', playlist, 'URL')
+        video_url = self._search_regex(
-        title = remove_start(self._html_search_regex(
+            r'<source[^>]+?src="([^"]+)"', webpage, 'video URL')
-            r'<title>(.*?)</title>', webpage, 'title'), 'VideoMega.tv - ')
-
-        formats = [{
-            'format_id': 'sd',
-            'url': video_url,
-        }]
-        self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,
-            'formats': formats,
+            'url': video_url,
             'thumbnail': thumbnail,
             'http_headers': {
                 'Referer': iframe_url,