Issue 844577
Created on 2003-11-18 18:03 by tepperly, last changed 2003-12-05 03:27 by richard.
File name |
Uploaded |
Description |
Edit |
Remove |
roundup063_client_py.txt
|
tepperly,
2003-11-18 18:03
|
Patch for roundup-0.6.3/roundup/cgi/client.py |
|
|
msg2668 |
Author: [hidden] (tepperly) |
Date: 2003-11-18 18:03 |
|
Attached is a three line diff for
round-0.6.3/roundup/cgi/client.py. Since the content
length is easily available, we should send it. I can't
find it now, but I remember reading somewhere when I
was doing a Java servlet that setting Content-Length
allowed browsers to do things faster.
References:
http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.13
|
msg2669 |
Author: [hidden] (tepperly) |
Date: 2003-11-18 18:29 |
|
Logged In: YES
user_id=94539
For Java, setting Content-Length allows the server to use a
persistent connection. I don't know if anything similar
applies to the Python server.
http://java.sun.com/products/servlet/2.2/javadoc/javax/servlet/http/HttpServlet.html#doGet(javax.servlet.http.HttpServletRequest,%20javax.servlet.http.HttpServletResponse)
|
msg2670 |
Author: [hidden] (tepperly) |
Date: 2003-11-18 19:46 |
|
Logged In: YES
user_id=94539
I tried to do some testing. I didn't really see any
significant difference between setting the Content-Length
and not setting the Content-Length.
|
msg2671 |
Author: [hidden] (richard) |
Date: 2003-12-05 03:27 |
|
Logged In: YES
user_id=6405
Hurm. This patch assumes a single write() call with the entire response,
which isn't always the case. I've added the content-length header to the
file serving, as we do know the length beforehand there.
|
|
Date |
User |
Action |
Args |
2003-11-18 18:03:31 | tepperly | create | |
|