Merge branch 'rzhxeo-cinemassacre'

youtube_dl/extractor/cinemassacre.py

@@ -1,10 +1,12 @@
 # encoding: utf-8
 from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    int_or_none,
 )
 
 
@@ -13,9 +15,10 @@ class CinemassacreIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
-            'file': '19911.mp4',
+            'md5': 'fde81fbafaee331785f58cd6c0d46190',
-            'md5': '782f8504ca95a0eba8fc9177c373eec7',
             'info_dict': {
+                'id': '19911',
+                'ext': 'mp4',
                 'upload_date': '20121110',
                 'title': '“Angry Video Game Nerd: The Movie” – Trailer',
                 'description': 'md5:fb87405fcb42a331742a0dce2708560b',
@@ -23,9 +26,10 @@ class CinemassacreIE(InfoExtractor):
         },
         {
             'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
-            'file': '521be8ef82b16.mp4',
+            'md5': 'd72f10cd39eac4215048f62ab477a511',
-            'md5': 'dec39ee5118f8d9cc067f45f9cbe3a35',
             'info_dict': {
+                'id': '521be8ef82b16',
+                'ext': 'mp4',
                 'upload_date': '20131002',
                 'title': 'The Mummy’s Hand (1940)',
             },
@@ -50,29 +54,40 @@ def _real_extract(self, url):
             r'<div class="entry-content">(?P<description>.+?)</div>',
             webpage, 'description', flags=re.DOTALL, fatal=False)
 
-        playerdata = self._download_webpage(playerdata_url, video_id)
+        playerdata = self._download_webpage(playerdata_url, video_id, 'Downloading player webpage')
+        video_thumbnail = self._search_regex(
+            r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False)
+        sd_url = self._search_regex(r'file: \'([^\']+)\', label: \'SD\'', playerdata, 'sd_file')
+        videolist_url = self._search_regex(r'file: \'([^\']+\.smil)\'}', playerdata, 'videolist_url')
 
-        sd_url = self._html_search_regex(r'file: \'([^\']+)\', label: \'SD\'', playerdata, 'sd_file')
+        videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
-        hd_url = self._html_search_regex(
-            r'file: \'([^\']+)\', label: \'HD\'', playerdata, 'hd_file',
-            default=None)
-        video_thumbnail = self._html_search_regex(r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False)
 
-        formats = [{
+        formats = []
-            'url': sd_url,
+        baseurl = sd_url[:sd_url.rfind('/')+1]
-            'ext': 'mp4',
+        for video in videolist.findall('.//video'):
-            'format': 'sd',
+            src = video.get('src')
-            'format_id': 'sd',
+            if not src:
-            'quality': 1,
+                continue
-        }]
+            file_ = src.partition(':')[-1]
-        if hd_url:
+            width = int_or_none(video.get('width'))
-            formats.append({
+            height = int_or_none(video.get('height'))
-                'url': hd_url,
+            bitrate = int_or_none(video.get('system-bitrate'))
-                'ext': 'mp4',
+            format = {
-                'format': 'hd',
+                'url': baseurl + file_,
-                'format_id': 'hd',
+                'format_id': src.rpartition('.')[0].rpartition('_')[-1],
-                'quality': 2,
+            }
-            })
+            if width or height:
+                format.update({
+                    'tbr': bitrate // 1000 if bitrate else None,
+                    'width': width,
+                    'height': height,
+                })
+            else:
+                format.update({
+                    'abr': bitrate // 1000 if bitrate else None,
+                    'vcodec': 'none',
+                })
+            formats.append(format)
         self._sort_formats(formats)
 
         return {