There is an issue filed in JIRA at:  http://openlaszlo.org/jira/browse/LPP-437 with the above-mentioned title.  Though this can clearly cause enormous problems in production, there is a dev-time use case for this behavior.

I run my server in a debugger, and make extensive use of breakpoints, etc., while debugging issues between Laszlo and the data coming from the server.  If we make it so a hanging data request is ignored or timed out, and Laszlo carries on, I won't be able to rely on my Laszlo app waiting until I'm done on the server before trying to, say, render all the data.

I'm wondering if a fix to this issue might be able to include a setting to turn on "hanging requests"?  Though I suppose there are other ways to achieve my goals, Laszlo has made it awfully easy so far

I'm also curious whether this issue is _strictly_ for SOLO apps.  I ask because there was an old mailing list thread at http://www.openlaszlo.org/pipermail/laszlo-dev/2005-April/000877.html regarding the development of a proper debugger for laszlo (with breakpoints).   It occurs to me that there's some vague chance this "bug" be leveraged to make implementing this behavior a little more viable...  am I crazy?

  Ben

---

co+optual (www.cooptual.com)

Ben Cox

Chief Technologist

831.372.7098

"Reality is that which, when you stop believing in it, does not go away."  - Philip K. Dick

_______________________________________________
Laszlo-dev mailing list
Laszlo-dev <at> openlaszlo.org
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev

A bigger win for me would be if the SOLO compiler was packaged and released as a separate, single jar file for easier integration with existing Ant or Maven based web projects.

I think Laszlo should be targeting one thing in the short term - the compiling of SOLO applications as easily as possible with the new Flash 8 features.  Improving the debugging features of a SOLO build would also be great. 

Personally, I think trying to move into the Java or .NET space is a waste of time... to me, it sounds like an awful lot of work for something that will likely not get market/mind share against competitors

If I were looking for some sort of project to promote and extend the life of Laszlo, I would seriously consider how to integrate Laszlo easily with Ruby on Rails. http://www.rubyonrails.org/    It probably involves adding a REST component in RoR to faciliate Laszlo use.  However, the combination of the two technologies would be really sweet for rapid development.  Keeping Flash as the runtime, or even trying to compile to DHTML / AJAX would be more worthwhile than a JVM target, IMO.

Sorry to be such a party pooper.. someone has to present a dissenting viewpoint.

Cheers,
  Don

---------------------------------------
Donald Dwoske
Software Journeyman
http://www.loraxis.com/ddwoske

_______________________________________________
Laszlo-dev mailing list
Laszlo-dev <at> openlaszlo.org
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev

I am going to try to add a fallback mode, so if the debugger evaluator gets a network error when trying to send an expression back to the server to compile, it will print a little warning and then try to run it through builtin Flash's eval(), so you can at least have simple inspection capabilities if there is no LPS server.

Don:

Note that objects that you present with Debug.write, or Debug.format
("%w", ...) are inspectable (by clicking on them) and this does not
require an LPS connection.  Warnings and Errors (created by
Debug.error and Debug.warn) are also inspectable, and capture a
backtrace if you have compiled with backtraces enabled.  This can be
handy if you are trying to figure out what is causing an undefined
reference warning, for instance.  Inspection does not require an LPS
back end.

Henry:

Note that the built-in Flash eval can't do anything more than look up
a variable reference, so perhaps we should write a little more of and
evaluator, that at least handles variable assignment and simple
function calls.

On 5 Oct 2005, at 14:50, Don Dwoske wrote:

> That is basically what I was talking about. Having something is
> better than
> nothing, and at the moment, I can't do much with the debugger
> (other than
> Debug.write) if there is no LPS server deployment - and with SOLO
> apps,
> that's pretty likely.
>
> That, along with simpler compilation for a solo app (preferably a
> single jar
> - with well known dependencies) inside of an Ant or Maven build
> (including
> samples), would be excellent features. If I have the time to put
> something
> useful together here I will, because I may need to do this anyway
> for my
> project.
>
> -d-
>
> On 10/5/05, Henry Minsky <henry.minsky <at> gmail.com> wrote:
>
>>
>>
>> I am going to try to add a fallback mode, so if the debugger
>> evaluator
>> gets a network error when trying to send an expression back to the
>> server to
>> compile, it will print a little warning and then try to run it
>> through
>> builtin Flash's eval(), so you can at least have simple inspection
>> capabilities if there is no LPS server.
>>
>>
>>
>
>
> --
>
> ---------------------------------------
> Donald Dwoske
> Software Journeyman
> http://www.loraxis.com/ddwoske
> _______________________________________________
> Laszlo-dev mailing list
> Laszlo-dev <at> openlaszlo.org
> http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
>