From d9dd3584e1051b159a748da32cd7d7a1da0bb787 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Sergey=20M=E2=80=A4?= <dstftw@gmail.com>
Date: Wed, 28 May 2014 19:38:44 +0700
Subject: [PATCH] [cinemassacre] Improve formats extraction and modernize

---
 youtube_dl/extractor/cinemassacre.py | 43 ++++++++++++++++++----------
 1 file changed, 28 insertions(+), 15 deletions(-)

diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py
index 8167a2370..496271be4 100644
--- a/youtube_dl/extractor/cinemassacre.py
+++ b/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',
             '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',
             'info_dict': {
+                'id': '521be8ef82b16',
+                'ext': 'mp4',
                 'upload_date': '20131002',
                 'title': 'The Mummy’s Hand (1940)',
             },
@@ -50,29 +54,38 @@ 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)
-        video_thumbnail = self._search_regex(r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False)
+        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')
-        
-        videolist = self._download_webpage(videolist_url, video_id)
+
+        videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
+
         formats = []
         baseurl = sd_url[:sd_url.rfind('/')+1]
-        for match in re.finditer('<video src="mp4:(?P<file>[^"]+_(?P<format_id>[^"]+)\.[^"]+)" system-bitrate="(?P<br>\d+)"(?: width="(?P<width>\d+)" height="(?P<height>\d+)")?/>', videolist):
+        for video in videolist.findall('.//video'):
+            src = video.get('src')
+            if not src:
+                continue
+            file_ = src.partition(':')[-1]
+            width = int_or_none(video.get('width'))
+            height = int_or_none(video.get('height'))
+            bitrate = int_or_none(video.get('system-bitrate'))
             format = {
-                'url': baseurl + match.group('file'),
-                'format_id': match.group('format_id')
+                'url': baseurl + file_,
+                'format_id': src.rpartition('.')[0].rpartition('_')[-1],
             }
-            if match.group('width'):
+            if width or height:
                 format.update({
-                    'tbr': int(match.group('br')) // 1000,
-                    'width': int(match.group('width')),
-                    'height': int(match.group('height'))
+                    'tbr': bitrate // 1000 if bitrate else None,
+                    'width': width,
+                    'height': height,
                 })
             else:
                 format.update({
-                    'abr': int(match.group('br')) // 1000,
-                    'vcodec': 'none'
+                    'abr': bitrate // 1000 if bitrate else None,
+                    'vcodec': 'none',
                 })
             formats.append(format)
         self._sort_formats(formats)