From 8389164081c357ce97b50459924328785debc805 Mon Sep 17 00:00:00 2001
From: Mort Yao <mort.yao@gmail.com>
Date: Mon, 17 Sep 2012 13:11:46 +0200
Subject: [PATCH] fix #2, #4: parse YouTube signature

---
 you_get/common.py             |  5 +++++
 you_get/downloader/youtube.py | 13 ++++++++++++-
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/you_get/common.py b/you_get/common.py
index ef1f7318..cc50c68f 100644
--- a/you_get/common.py
+++ b/you_get/common.py
@@ -43,6 +43,9 @@ def r1_of(patterns, text):
         if x:
             return x
 
+def unicodize(text):
+    return re.sub(r'\\u([0-9A-Fa-f][0-9A-Fa-f][0-9A-Fa-f][0-9A-Fa-f])', lambda x: chr(int(x.group(0)[2:], 16)), text)
+
 def escape_file_path(path):
     path = path.replace('/', '-')
     path = path.replace('\\', '-')
@@ -572,6 +575,8 @@ def script_main(script_name, download, download_playlist = None):
     set_http_proxy(proxy)
     
     for url in args:
+        if url.startswith('https://'):
+            url = url[8:]
         if not url.startswith('http://'):
             url = 'http://' + url
         
diff --git a/you_get/downloader/youtube.py b/you_get/downloader/youtube.py
index 4b12f165..918508cc 100644
--- a/you_get/downloader/youtube.py
+++ b/you_get/downloader/youtube.py
@@ -8,7 +8,17 @@ def youtube_download_by_id(id, title = None, output_dir = '.', merge = True, inf
     try:
         url = parse.parse_qs(parse.unquote(request.urlopen('http://www.youtube.com/get_video_info?&video_id=' + id).read().decode('utf-8')))['url_encoded_fmt_stream_map'][0][4:]
     except:
-        url = parse.parse_qs(parse.unquote(request.urlopen('http://www.youtube.com/watch?v=' + id).read().decode('utf-8')))['url_encoded_fmt_stream_map'][0][4:]
+        html = request.urlopen('http://www.youtube.com/watch?v=' + id).read().decode('utf-8')
+        url = parse.parse_qs(parse.unquote(html))['url_encoded_fmt_stream_map'][0][4:]
+        if not url.startswith('http://'):
+            codec = r1(r'yt.preload.start\("http:\\/\\/([^\\]+)\\/crossdomain.xml"\)', html)
+            signature = r1(r'signature=([^\\]+)\\', html)
+            urls = parse.parse_qs(parse.unquote(html))['url']
+            for u in urls:
+                if u.startswith('http://' + codec):
+                    url = "%s&signature=%s" % (u, signature)
+                    break
+    
     type, ext, size = url_info(url)
     
     print_info(site_info, title, type, size)
@@ -23,6 +33,7 @@ def youtube_download(url, output_dir = '.', merge = True, info_only = False):
     except:
         html = get_html(url, 'utf-8')
         title = r1(r'"title": "([^"]+)"', html)
+        title = unicodize(title)
     assert title
     title = parse.unquote(title)
     title = escape_file_path(title)