[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/bigflix.py b/youtube_dl/extractor/bigflix.py
index aeea1a002e02b1fd61e3d92bcf09e7e65a92f2b1..28e3e59f670995b747fc028f99c285e2ef7e69aa 100644 (file)
--- a/youtube_dl/extractor/bigflix.py
+++ b/youtube_dl/extractor/bigflix.py
@@ -1,24 +1,35 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from base64 import b64decode
+import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_b64decode,
+    compat_urllib_parse_unquote,
+)
 
 
 class BigflixIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?bigflix\.com/.*/(?P<id>[0-9]+)'
-    _TEST = {
-        'url': 'http://www.bigflix.com/Hindi-movies/Action-movies/Singham-Returns/16537',
-        'md5': 'ec76aa9b1129e2e5b301a474e54fab74',
+    _VALID_URL = r'https?://(?:www\.)?bigflix\.com/.+/(?P<id>[0-9]+)'
+    _TESTS = [{
+        # 2 formats
+        'url': 'http://www.bigflix.com/Tamil-movies/Drama-movies/Madarasapatinam/16070',
         'info_dict': {
-            'id': '16537',
+            'id': '16070',
             'ext': 'mp4',
-            'title': 'Singham Returns',
-            'description': 'md5:3d2ba5815f14911d5cc6a501ae0cf65d',
+            'title': 'Madarasapatinam',
+            'description': 'md5:9f0470b26a4ba8e824c823b5d95c2f6b',
+            'formats': 'mincount:2',
+        },
+        'params': {
+            'skip_download': True,
         }
-    }
+    }, {
+        # multiple formats
+        'url': 'http://www.bigflix.com/Malayalam-movies/Drama-movies/Indian-Rupee/15967',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -29,14 +40,39 @@ class BigflixIE(InfoExtractor):
             r'<div[^>]+class=["\']pagetitle["\'][^>]*>(.+?)</div>',
             webpage, 'title')
 
-        video_url = b64decode(compat_urllib_parse_unquote(self._search_regex(
-            r'file=([^&]+)', webpage, 'video url')).encode('ascii')).decode('utf-8')
+        def decode_url(quoted_b64_url):
+            return compat_b64decode(compat_urllib_parse_unquote(
+                quoted_b64_url)).decode('utf-8')
+
+        formats = []
+        for height, encoded_url in re.findall(
+                r'ContentURL_(\d{3,4})[pP][^=]+=([^&]+)', webpage):
+            video_url = decode_url(encoded_url)
+            f = {
+                'url': video_url,
+                'format_id': '%sp' % height,
+                'height': int(height),
+            }
+            if video_url.startswith('rtmp'):
+                f['ext'] = 'flv'
+            formats.append(f)
+
+        file_url = self._search_regex(
+            r'file=([^&]+)', webpage, 'video url', default=None)
+        if file_url:
+            video_url = decode_url(file_url)
+            if all(f['url'] != video_url for f in formats):
+                formats.append({
+                    'url': decode_url(file_url),
+                })
+
+        self._sort_formats(formats)
 
         description = self._html_search_meta('description', webpage)
 
         return {
             'id': video_id,
             'title': title,
-            'url': video_url,
             'description': description,
+            'formats': formats
         }