comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: border around draw widget
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: border around draw widget [message #50360 is a reply to message #50343] Thu, 28 September 2006 00:21 Go to previous messageGo to previous message
Laurens is currently offline  Laurens
Messages: 41
Registered: May 2006
Member
JD Smith wrote:
> On Wed, 27 Sep 2006 23:35:34 +0200, Laurens wrote:
>
>> Thanks very much for that explanation!
>> Could you tell me how to make such a widget-object? It sounds like
>> something I was already thinking about...
>
> It sounds fancier than it is. It's basically an object, which:
>
> 1. Sets up a widget heirarchy in the normal way (usually in its Init
> method).
> 2. Saves its "state" information not in a structure in a UVALUE but in
> the class data itself (e.g. self.*).
> 3. Calls XManager (often, but not necessarily, in its Init method) to
> generate events on that widget heirarchy.
> 4. Uses the trick I outline to inject the events flowing forth from
> the widgets created to some class method (often named "Event").
>
> The main advantages of this method:
>
> 1. You get state information "for free", quite nicely mapped to class
> data.
> 2. You automatically avoid common blocks for state info, with their
> associated collision risks if multiple identical widgets run at the
> same time.
> 3. You are never left with state information "in the air", if you use
> /NO_COPY to be efficient when retrieving your state structure from
> a UVALUE. This greatly aids debugging, since crashes to the code
> usually can be recovered from with a simple RETALL.
> 4. You quickly realize that the normal event flow embodied in "normal"
> widget prgramming is limiting, and can roll your own communication
> among objects that suits your needs. This is particularly useful
> if you have many different perhaps unrelated application components
> that need to communicate with eachother.
>
> A schematic usage would be:
>
> oDraw=obj_new('SelectableDrawPane',base)
>
> which would place a new compound widget into base. It might implement
> some methods "Select" and "DeSelect", or you could have it trap the
> selection "clicks" and automagically select/deselect itself. Once you
> have the apparatus in place, you can then have fun implementing other
> methods for your object, drawing and erasing, etc.
>
> I should note that none of this is necessary to use the "base on top
> of a base" trickery I outlined before, it just makes it easier and more
> powerful.
>
> JD
>
err...well, I'll try hehe; If I understand it correctly, this implies
writing code in the GUI.pro file, with as disadvantage that I can't use
my .prc file to regenerate GUI?
That's some strange behaviour I noticed earlier...if you change position
of a widget and save the GUI, all self-written code is simply gone :S

If I've created that object, where could I change its structure, like
the select and deselect functions?
Sorry for the quite explicit way of asking, but hey I'm not as
experienced in writing IDL as you guys huh (will one ever be haha), so
I'm just trying to learn from it...

Thanks though for what you've brought up on ideas so far...

Laurens

Cheers David :) It's quite funny, I live in the Netherlands and when we
use the word "cheers", its when we take a beer haha; so every "cheers"
underneath your msgs lets me think you're having quite a good time lol.
[Message index]
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Re: call_procedure with a dynamically created arguments list?
Next Topic: Listing of Internal IDL Routines

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Oct 10 08:35:05 PDT 2025

Total time taken to generate the page: 1.35960 seconds