Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / cliphunter.py

diff --git a/youtube_dl/extractor/cliphunter.py b/youtube_dl/extractor/cliphunter.py
index 2edab90a33d553225b8c790b8d391f0e40b55cf8..f2ca7a337df88aab5ba7f7b070537b7b9fec3172 100644 (file)
--- a/youtube_dl/extractor/cliphunter.py
+++ b/youtube_dl/extractor/cliphunter.py
@@ -1,41 +1,41 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
-
-
-_translation_table = {
-    'a': 'h', 'd': 'e', 'e': 'v', 'f': 'o', 'g': 'f', 'i': 'd', 'l': 'n',
-    'm': 'a', 'n': 'm', 'p': 'u', 'q': 't', 'r': 's', 'v': 'p', 'x': 'r',
-    'y': 'l', 'z': 'i',
-    '$': ':', '&': '.', '(': '=', '^': '&', '=': '/',
-}
-
-
-def _decode(s):
-    return ''.join(_translation_table.get(c, c) for c in s)
+from ..utils import (
+    int_or_none,
+    url_or_none,
+)
 
 
 class CliphunterIE(InfoExtractor):
     IE_NAME = 'cliphunter'
 
-    _VALID_URL = r'''(?x)http://(?:www\.)?cliphunter\.com/w/
+    _VALID_URL = r'''(?x)https?://(?:www\.)?cliphunter\.com/w/
         (?P<id>[0-9]+)/
         (?P<seo>.+?)(?:$|[#\?])
     '''
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'md5': 'a2ba71eebf523859fe527a61018f723e',
+        'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
             'id': '1012420',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Fun Jynx Maze solo',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
             'age_limit': 18,
-        }
-    }
+        },
+        'skip': 'Video gone',
+    }, {
+        'url': 'http://www.cliphunter.com/w/2019449/ShesNew__My_booty_girlfriend_Victoria_Paradices_pussy_filled_with_jizz',
+        'md5': '55a723c67bfc6da6b0cfa00d55da8a27',
+        'info_dict': {
+            'id': '2019449',
+            'ext': 'mp4',
+            'title': 'ShesNew - My booty girlfriend, Victoria Paradice\'s pussy filled with jizz',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'age_limit': 18,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -44,39 +44,26 @@ class CliphunterIE(InfoExtractor):
         video_title = self._search_regex(
             r'mediaTitle = "([^"]+)"', webpage, 'title')
 
-        pl_fiji = self._search_regex(
-            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
-        pl_c_qual = self._search_regex(
-            r'pl_c_qual = "(.)"', webpage, 'video quality')
-        video_url = _decode(pl_fiji)
-        formats = [{
-            'url': video_url,
-            'format_id': 'default-%s' % pl_c_qual,
-        }]
-
-        qualities_json = self._search_regex(
-            r'var pl_qualities\s*=\s*(.*?);\n', webpage, 'quality info')
-        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,
+        gexo_files = self._parse_json(
+            self._search_regex(
+                r'var\s+gexoFiles\s*=\s*({.+?});', webpage, 'gexo files'),
+            video_id)
+
+        formats = []
+        for format_id, f in gexo_files.items():
+            video_url = url_or_none(f.get('url'))
+            if not video_url:
+                continue
+            fmt = f.get('fmt')
+            height = f.get('h')
+            format_id = '%s_%sp' % (fmt, height) if fmt and height else format_id
+            formats.append({
                 'url': video_url,
-                'quality': i,
-            }
-            if quality_id in qualities_data:
-                qd = qualities_data[quality_id]
-                m = re.match(
-                    r'''(?x)<b>(?P<width>[0-9]+)x(?P<height>[0-9]+)<\\/b>
-                        \s*\(\s*(?P<tbr>[0-9]+)\s*kb\\/s''', qd)
-                if m:
-                    f['width'] = int(m.group('width'))
-                    f['height'] = int(m.group('height'))
-                    f['tbr'] = int(m.group('tbr'))
-            formats.append(f)
+                'format_id': format_id,
+                'width': int_or_none(f.get('w')),
+                'height': int_or_none(height),
+                'tbr': int_or_none(f.get('br')),
+            })
         self._sort_formats(formats)
 
         thumbnail = self._search_regex(