[cliphunter] Fix extraction and update test (Fixes #4362)

youtube_dl/extractor/cliphunter.py

@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 _translation_table = {
@@ -27,10 +25,10 @@ class CliphunterIE(InfoExtractor):
     '''
     _TEST = {
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'md5': 'a2ba71eebf523859fe527a61018f723e',
+        'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
             'id': '1012420',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Fun Jynx Maze solo',
             'thumbnail': 're:^https?://.*\.jpg$',
             'age_limit': 18,
@@ -44,39 +42,31 @@ class CliphunterIE(InfoExtractor):
         video_title = self._search_regex(
             r'mediaTitle = "([^"]+)"', webpage, 'title')
 
-        pl_fiji = self._search_regex(
+        fmts = {}
-            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
+        for fmt in ('mp4', 'flv'):
-        pl_c_qual = self._search_regex(
+            fmt_list = self._parse_json(self._search_regex(
-            r'pl_c_qual = "(.)"', webpage, 'video quality')
+                r'var %sjson\s*=\s*(\[.*?\]);' % fmt, webpage, '%s formats' % fmt), video_id)
-        video_url = _decode(pl_fiji)
+            for f in fmt_list:
-        formats = [{
+                fmts[f['fname']] = _decode(f['sUrl'])
-            'url': video_url,
+
-            'format_id': 'default-%s' % pl_c_qual,
+        qualities = self._parse_json(self._search_regex(
-        }]
+            r'var player_btns\s*=\s*(.*?);\n', webpage, 'quality info'), video_id)
-
+
-        qualities_json = self._search_regex(
+        formats = []
-            r'var pl_qualities\s*=\s*(.*?);\n', webpage, 'quality info')
+        for fname, url in fmts.items():
-        qualities_data = json.loads(qualities_json)
-
-        for i, t in enumerate(
-                re.findall(r"pl_fiji_([a-z0-9]+)\s*=\s*'([^']+')", webpage)):
-            quality_id, crypted_url = t
-            video_url = _decode(crypted_url)
             f = {
-                'format_id': quality_id,
+                'url': url,
-                'url': video_url,
-                'quality': i,
             }
-            if quality_id in qualities_data:
+            if fname in qualities:
-                qd = qualities_data[quality_id]
+                qual = qualities[fname]
-                m = re.match(
+                f.update({
-                    r'''(?x)<b>(?P<width>[0-9]+)x(?P<height>[0-9]+)<\\/b>
+                    'format_id': '%s_%sp' % (determine_ext(url), qual['h']),
-                        \s*\(\s*(?P<tbr>[0-9]+)\s*kb\\/s''', qd)
+                    'width': qual['w'],
-                if m:
+                    'height': qual['h'],
-                    f['width'] = int(m.group('width'))
+                    'tbr': qual['br'],
-                    f['height'] = int(m.group('height'))
+                })
-                    f['tbr'] = int(m.group('tbr'))
             formats.append(f)
+
         self._sort_formats(formats)
 
         thumbnail = self._search_regex(