Three hours ago, Jay McCarthy wrote:
> Alright. I looked into this more.
>
> The 200 and 512 from the log are not about the request, they are about the
> response:
>
> "200 (%>s) [...]"
But the original problem was about the response size...
> But the Racket Web server only has request logging by default, not
> response logging, so it makes something up in these positions. I
> will update the docs to mention this.
...also, in what Jordan posted there was a
Content-Length: 2629
header, so the information is already there.
On Mon, Dec 12, 2011 at 11:30:59AM -0700, Jay McCarthy wrote:
>
> Is there a performance reason for not logging information about the
> response? It would seem that (especially) the response code would be
> valuable to include by default?
The common reason to not include it is when the response comes from
code, and you can't include the size before you've run the code which
means that you have to hold the complete response somewhere before you
send it. But since there was a content-length header, it looks like
the example you gave was for sending a file.
--
((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay:
http://barzilay.org/ Maze is Life!
_________________________________________________
For list-related administrative tasks:
http://lists.racket-lang.org/listinfo/users
