Merge tag 'upstream/2014.10.30'

[youtubedl] / youtube_dl / extractor / cliphunter.py

diff --git a/youtube_dl/extractor/cliphunter.py b/youtube_dl/extractor/cliphunter.py
index 58846e8e7cfacc631015ce32ace23cbd9b8a2576..2edab90a33d553225b8c790b8d391f0e40b55cf8 100644 (file)
--- a/youtube_dl/extractor/cliphunter.py
+++ b/youtube_dl/extractor/cliphunter.py
@@ -1,11 +1,12 @@
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
 
 
-translation_table = {
+_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',
@@ -13,6 +14,10 @@ translation_table = {
 }
 
 
+def _decode(s):
+    return ''.join(_translation_table.get(c, c) for c in s)
+
+
 class CliphunterIE(InfoExtractor):
     IE_NAME = 'cliphunter'
 
@@ -22,35 +27,66 @@ class CliphunterIE(InfoExtractor):
     '''
     _TEST = {
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'file': '1012420.flv',
-        'md5': '15e7740f30428abf70f4223478dc1225',
+        'md5': 'a2ba71eebf523859fe527a61018f723e',
         'info_dict': {
+            'id': '1012420',
+            'ext': 'mp4',
             'title': 'Fun Jynx Maze solo',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
+        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_title = self._search_regex(
-            r'mediaTitle = "([^"]+)"', webpage, 'title')
-
-        video_url = ''.join(translation_table.get(c, c) for c in pl_fiji)
-
+        video_url = _decode(pl_fiji)
         formats = [{
             'url': video_url,
-            'format_id': pl_c_qual,
+            '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,
+                '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)
+        self._sort_formats(formats)
+
+        thumbnail = self._search_regex(
+            r"var\s+mov_thumb\s*=\s*'([^']+)';",
+            webpage, 'thumbnail', fatal=False)
+
         return {
             'id': video_id,
             'title': video_title,
             'formats': formats,
+            'age_limit': self._rta_search(webpage),
+            'thumbnail': thumbnail,
         }