Is there any way to forward a request, instead of include (as with
ns_adp_include ) ?
regards,
Sarath
[EMAIL PROTECTED]
--
AOLserver - http://www.aolserver.com/
To Remove yourself from this list, simply send an email to [EMAIL PROTECTED]
with the
body of SIGNOFF AOLSERVER in the email
Hi everybody,
As I read there are several possibilities to get AOLserver serving
virtual hosts.
There are 90 domains which shall be handeled by one AOLserver. Some
of this domains use the same document-root.
What is the best and easiest way to get those domains handeled by the
AOLserver?
Sarath PS said:
Is there any way to forward a request, instead of include (as with
ns_adp_include ) ?
Forward, as in how you can forward to another resource (JSP/Servlet) in a
J2EE server, which then is treated as a completely new page but with the
same request variables available?
Not that I
You could redirect the user, which will send the browser a 302
(redirection) HTTP code.
%
ns_returnredirect $url
ns_adp_abort
%
Note the trailing ns_adp_abort - even though the user receives the
redirection, the ADP page will continue to be parsed, which in some
cases can actually be
On 2005.09.25, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
A few years ago at one of the online meetings we had discussed content
acceleration with gzip. I know that 4.0.10 and above support on-the-fly
ADP compression but are there plans to extend it to static files?
With Tux, when a URL
I' using the same method now.
But here redirect is actually done by the client,
Can i save the round trip involved here ?
thanks for your help
On Mon, 2005-09-26 at 07:07 -0400, Nathan Folkman wrote:
You could redirect the user, which will send the browser a 302
(redirection) HTTP code.
On 2005.09.26, Bas Scheffers [EMAIL PROTECTED] wrote:
Sarath PS said:
Is there any way to forward a request, instead of include (as with
ns_adp_include ) ?
Forward, as in how you can forward to another resource (JSP/Servlet) in a
J2EE server, which then is treated as a completely new page
I use the following code to forward requests, which seems to work ok.
proc ladp_forward {adpFile} {
ns_adp_trunc
ns_adp_include $adpFile
ns_adp_break
}
Rob Seeger
Dossy Shiobara wrote on 9/26/2005, 8:08 AM:
On 2005.09.26, Bas Scheffers [EMAIL PROTECTED] wrote:
Sarath PS
Ahh, I think you want a way to do an internal redirect. Nothing
currently is exposed that would allow you to do that, but it's
definitely something we can look into adding.
- n
On Sep 26, 2005, at 7:53 AM, Sarath PS wrote:
I' using the same method now.
But here redirect is actually done
The way most people add any virtual servers is by taking the stock nsd.tcl
file for all the basic non-server specific settings.
Then when you come to the ns/server/$servername sections, you want to
automate things a little. One way is to move that to one file per server
and in the main nsd.tcl
Here's an alternate approach that doesn't use virutal servers, and
instead uses a tiny bit of logic that specifies page roots based on
the HTTP host header passed in on the request. I'd suggest
configuring a start page for this logic over using a registered
proc, since this will allow you
On Monday 26 September 2005 03:26, Christian Krause wrote:
Hi everybody,
As I read there are several possibilities to get AOLserver serving
virtual hosts.
There are 90 domains which shall be handeled by one AOLserver. Some
of this domains use the same document-root.
What is the best and
On Mon, Sep 26, 2005 at 08:08:09AM -0400, Dossy Shiobara wrote:
On 2005.09.26, Bas Scheffers [EMAIL PROTECTED] wrote:
Hmm, I wonder what the harm would be in exposing Ns_ConnRedirect() via a
Tcl command, which would give a somewhat similar internal redirect or
forward capability to Tcl/ADP
On 2005.09.26, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
Well of course, I was wondering if this would ever be part of the core?
ns_gzip works well but it is better suited for on-the-fly compression.
Is the question will ns_gzip be included as part of the core or is
there a better way of
On Mon, 2005-09-26 at 11:30 -0400, Dossy Shiobara wrote:
On 2005.09.26, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
Is the question will ns_gzip be included as part of the core or is
there a better way of integrating gzip into the core than just using
ns_gzip?
The question is neither.
Let's
On Monday 26 September 2005 08:39, Dossy Shiobara wrote:
One warning: hosting dynamic pages (ADPs) this way is very dangerous.
Software virtual servers with AOLserver is ONLY safe for static content
only.
Isn't this only true in the case of multiple users/webmasters?
tom jackson
--
Makes sense... Is there anything preventing you from adding some code
to implement this logic yourself, on top of a vanilla AOLserver
with something like the ns_gzip module?
- n
On Sep 26, 2005, at 12:41 PM, Daniel P. Stasinski wrote:
On Mon, 2005-09-26 at 11:30 -0400, Dossy Shiobara
On Monday 26 September 2005 03:37, Sarath PS wrote:
Is there any way to forward a request, instead of include (as with
ns_adp_include ) ?
To change the url, without an external redirect, you can use the nsrewrite
module.
I've written it up here: http://rmadilo.com/files/nsrewrite/
The
On 2005.09.26, Tom Jackson [EMAIL PROTECTED] wrote:
On Monday 26 September 2005 08:39, Dossy Shiobara wrote:
One warning: hosting dynamic pages (ADPs) this way is very dangerous.
Software virtual servers with AOLserver is ONLY safe for static content
only.
Isn't this only true in the
On 2005.09.22, Jade Rubick [EMAIL PROTECTED] wrote:
I wanted to alert everyone here to an announcement on the OpenACS forums:
http://openacs.org/forums/message-view?message_id=324663
Thanks for the pointer! This is a great initiative: I personally would
like to see this effort happen.
On Mon, 2005-09-26 at 12:56 -0400, Nathan Folkman wrote:
Makes sense... Is there anything preventing you from adding some code
to implement this logic yourself, on top of a vanilla AOLserver
with something like the ns_gzip module?
There is nothing preventing me from doing it, but it seems
On 2005.09.26, Nathan Folkman [EMAIL PROTECTED] wrote:
Take a look at the head version, specifically:
[... ChangeLog excerpt snipped ...]
Ah, except Ns_Compress() isn't available through a Tcl command, yet.
A lot of times, the expression XYZ isn't available in AOLserver is
roughly
On 2005.09.23, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
Useless code, yes, but it should ns_eval correctly but does not.
set code set somevar {\$blah}
ns_eval $code#doesn't work
eval $code #works
Interesting -- does this work in 4.0.x but not in
On 2005.09.23, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
Useless code, yes, but it should ns_eval correctly but does not.
set code set somevar {\$blah}
ns_eval $code#doesn't work
eval $code #works
To answer my own question:
server1:nscp 1
To be honest, the Tux approach mentioned below might be your best
solution. AOLserver is primarily optimized for dynamic serving, with
static assets (flat HTML, images, JS, CSS, etc.) generally served by
servers optimized for static serving.
- n
On Sep 26, 2005, at 1:24 PM, Daniel P.
Hi Dossy:
Would having it available in a different license make AOL more interested in the project?
:)
Jade Recognizing the broad need for web services, this project will be
open-sourced under the GPL license [...]Why GPL and not MPL or AOLserver Public License (derived from MPL)?The GPL puts
On Mon, 2005-09-26 at 13:27 -0400, Dossy Shiobara wrote:
On 2005.09.23, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
Useless code, yes, but it should ns_eval correctly but does not.
set code set somevar {\$blah}
ns_eval $code#doesn't work
eval $code
On 2005.09.26, Jade Rubick [EMAIL PROTECTED] wrote:
Would having it available in a different license make AOL more
interested in the project?
In theory, the GPL licensing shouldn't play a role in AOL's involvement
in the project: I find it hard to believe that AOL would want to
commercially
On 2005.09.26, Daniel P. Stasinski [EMAIL PROTECTED] wrote:
If all I do is add a single # comment as the very first line to the same
script, it updates fine.
I has got to be in NSAdmin but I haven't been able to re-create the
error in anything other than the package itself.
No, I don't
On Mon, Sep 26, 2005 at 01:32:08PM -0400, Dossy Shiobara wrote:
Yes, this (kinda) makes sense -- if you look at the ns_eval code, it
double-evaluates the input script (which, seems to cause lots of
confusion and grief).
Uh, where and why does it do that exactly? I'm looking at
On Mon, 2005-09-26 at 15:09 -0400, Dossy Shiobara wrote:
No, I don't think it's NSAdmin. Rob Seeger should have a better
explanation as to why this bug is happening, I think. I'll try to dig
through my email, unless he beats me to it.
Ok, I can duplicate the bug now. Here is the code. Try
On Mon, 2005-09-26 at 13:44 -0400, Nathan Folkman wrote:
To be honest, the Tux approach mentioned below might be your best
solution. AOLserver is primarily optimized for dynamic serving, with
static assets (flat HTML, images, JS, CSS, etc.) generally served by
servers optimized for
In the _ns_helper_eval proc (which ns_eval uses to actually execute
code), the line that actually runs the code is:
set code [catch {uplevel 1 [eval concat $args]} result]
The reason for this is that ns_eval, by design, can take it's inputs in
two different forms:
ns_eval set a 1
Yes, the new driver model in the head version is much better suited
for serving static content, although we have yet to do any
performance testing to actually back this up. ;-) That said, the
motivation wasn't necessarily to replace Tux, and as I mentioned
earlier, the server is primarily
On Mon, Sep 26, 2005 at 04:27:19PM -0400, Robert Seeger wrote:
In the _ns_helper_eval proc (which ns_eval uses to actually execute
code), the line that actually runs the code is:
set code [catch {uplevel 1 [eval concat $args]} result]
The reason for this is that ns_eval, by design,
On 2005.09.26, Robert Seeger [EMAIL PROTECTED] wrote:
In the _ns_helper_eval proc (which ns_eval uses to actually execute
code), the line that actually runs the code is:
[...]
Thanks, Rob. Sorry to put you on the spot with this, but I know you
knew this better than me ...
So, I retract my
36 matches
Mail list logo
