[OS-webwork] Getting rid of thread locals

2003-01-11 Thread Patrick Lightbody
In effort to ditch the servlet paradigm, any thoughts on changing
ActionContext from being a ThreadLocal to just a normal Map? I've got some
code I can commit to CVS if you'd like to see how it would work.

-Pat




---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork



Re: [OS-webwork] Getting rid of thread locals

2003-01-11 Thread Hani Suleiman
Err, how is ThreadLocal servlet related? How would any of the context 
stuff work without it?

On Saturday, January 11, 2003, at 03:56 AM, Patrick Lightbody wrote:

In effort to ditch the servlet paradigm, any thoughts on changing
ActionContext from being a ThreadLocal to just a normal Map? I've got 
some
code I can commit to CVS if you'd like to see how it would work.

-Pat




---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork




---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork



Re: [OS-webwork] Getting rid of thread locals

2003-01-11 Thread Patrick Lightbody
ThreadLocal implies that you always have a single thread throughout the
lifecycle of the action (prepare, execute, print results). This is true for
a servlet container (single thread/request), but not so in other areas.

If you look at the code, the context Map is being passed around. I don't
really see it as that bad, but I guess that's all subjective.

-Pat

- Original Message -
From: Hani Suleiman [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Saturday, January 11, 2003 4:15 AM
Subject: Re: [OS-webwork] Getting rid of thread locals


 Err, how is ThreadLocal servlet related? How would any of the context
 stuff work without it?

 On Saturday, January 11, 2003, at 03:56 AM, Patrick Lightbody wrote:

  In effort to ditch the servlet paradigm, any thoughts on changing
  ActionContext from being a ThreadLocal to just a normal Map? I've got
  some
  code I can commit to CVS if you'd like to see how it would work.
 
  -Pat
 
 
 
 
  ---
  This SF.NET email is sponsored by:
  SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
  http://www.vasoftware.com
  ___
  Opensymphony-webwork mailing list
  [EMAIL PROTECTED]
  https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork
 



 ---
 This SF.NET email is sponsored by:
 SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
 http://www.vasoftware.com
 ___
 Opensymphony-webwork mailing list
 [EMAIL PROTECTED]
 https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork



---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork



Re: [OS-webwork] Getting rid of thread locals

2003-01-11 Thread Philipp Meier
On Sat, Jan 11, 2003 at 08:11:26AM -0800, Patrick Lightbody wrote:
 ThreadLocal implies that you always have a single thread throughout the
 lifecycle of the action (prepare, execute, print results). This is true for
 a servlet container (single thread/request), but not so in other areas.

Can you please tell of any other case? I think it's not to bad to
constrain the execution environment of an action the be a single thread.
The sequence of execution is sth. like Client-Dispatcher-Action, so
the real constraint would be that the Dispatcher ensures this.

 If you look at the code, the context Map is being passed around. I don't
 really see it as that bad, but I guess that's all subjective.

The usage of ThreadLocals is two edged sword. On the one hand, Thread
locals avoid to pass a context all around, but on the other hand, it
takes some time to understand how to use them. However this would actually
only affect XWork developer, not the users of XWork. IMHO Passing a context
is no pain either because this could always done in the constructor of
the Action, View or anything because theese are used only once in
webwork. So this is a matter of taste. Personally I would vote for not
using ThreadLocals, where possible.

-billy.

-- 
Meisterbohne   Söflinger Straße 100  Tel: +49-731-399 499-0
   eLösungen   89077 Ulm Fax: +49-731-399 499-9



msg01149/pgp0.pgp
Description: PGP signature


Re: [OS-webwork] Getting rid of thread locals

2003-01-11 Thread Rickard Öberg
Patrick Lightbody wrote:

ThreadLocal implies that you always have a single thread throughout the
lifecycle of the action (prepare, execute, print results). This is true for
a servlet container (single thread/request), but not so in other areas.


No, it doesn't imply that. If the execution chain is broken, then one 
simply takes the context and reuse it whenever the chain is restarted.

/Rickard



---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork


Re: [OS-webwork] Getting rid of thread locals

2003-01-11 Thread Patrick Lightbody
I know, it is possible to do this, but it's a bit more complicated, if only
because the user's (developers, whatever) don't actually _see_ the backing
map without opening up some stack method that provides access to the
ThreadLocal. If we pass the context around, all that happens is that
execute() - execute(Map context).

Before anyone shoots this down, I really want you to look at the code that
is in CVS. It's not that bad and it makes things much simpler -- which is a
primary goal of XWork.

I really think this new code is more powerful and much simpler than the
existing WebWork core. If you check out the source (especially with
DispatcherTest as well as deploy the war and access Foo.action -- same
thing), you'll find that it's pretty appealing (IMHO). :)

-Pat

- Original Message -
From: Rickard Öberg [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Saturday, January 11, 2003 10:15 AM
Subject: Re: [OS-webwork] Getting rid of thread locals


 Patrick Lightbody wrote:
  ThreadLocal implies that you always have a single thread throughout the
  lifecycle of the action (prepare, execute, print results). This is true
for
  a servlet container (single thread/request), but not so in other areas.

 No, it doesn't imply that. If the execution chain is broken, then one
 simply takes the context and reuse it whenever the chain is restarted.

 /Rickard



 ---
 This SF.NET email is sponsored by:
 SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
 http://www.vasoftware.com
 ___
 Opensymphony-webwork mailing list
 [EMAIL PROTECTED]
 https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork



---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Opensymphony-webwork mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork