[Lazarus] Maybe a bug: click on TImage behind another control with SetShape

[Fabio Luis Girardi]

I think that we aren't talking about the same thing.

I think that a transparent control must retain clicks to another controls
if it is uppermost and in front of these controls. I think that this is
correct.

Polyline is a different control. The area of line is the area of the
control, other area will be full transparent and must pass the events to
controls behind. If you click on line, the click must be fired on Polyline
object, otherwise, it must be fired on the first control behind of polyline.

The bug here is: I have two TImages and one polyline. The uppermost is
TImage is positioned in a removed area (transparent) of polyline and the
other TImage is the background. When I click on uppermost TImage click
isn't fired on it nor on background TImage. It's fired on TForm! The same
behavior happens if polyline is the uppermost control.

Using SetShape make the "click" ignore TImages and TSpeedButtons (in front
or behind) positioned in a removed area of a Shaped control.

Since the click messages are processed on parent, I made the message
redirection. Now I'm studing a way to check if the click is on a control
area or not, to make a TControl.CMHitTest fix.




2014-11-22 7:57 GMT-02:00 Frederic Da Vitoria <davitofrg at gmail.com>:

> I may be wrong, but this seems different from the problem you are
> describing: in the bugtracker issue, the click triggers the Form handler
> instead of being sent to a Control. In your problem, from your description,
> I thought the click triggered the polyline control while you would have
> wanted it to be handled by the TImage which is behind the polyline. In a
> way, your problem is not enough transparency (in terms of who gets to
> handle the click event) while the bugtracker bug is too much transparency.
> Or did I misunderstand you?
>
>
> 2014-11-22 10:01 GMT+01:00 Fabio Luis Girardi <fluisgirardi at gmail.com>:
>
>> Here is it: http://bugs.freepascal.org/view.php?id=20381
>>
>> 2014-11-21 22:24 GMT-02:00 Fabio Luis Girardi <fluisgirardi at gmail.com>:
>>
>> I found this bug on bug tracker. I followed your idea and send a patch
>>> that solves this. I'll try another approach tomorrow.
>>>
>>> Thank you very much!
>>> Em 21/11/2014 20:48, "Frederic Da Vitoria" <davitofrg at gmail.com>
>>> escreveu:
>>>
>>> 2014-11-21 16:01 GMT+01:00 Frederic Da Vitoria <davitofrg at gmail.com>:
>>>>
>>>>>
>>>>> 2014-11-20 19:21 GMT+01:00 Fabio Luis Girardi <fluisgirardi at gmail.com>
>>>>> :
>>>>>
>>>>>> I don't know if this is a bug or not. I created a control from a
>>>>>> TCustomControl that draws a polyline with a transparent background. The
>>>>>> transparent effect has been using SetShape procedure using a bitmap of 1
>>>>>> bit color depth.
>>>>>>
>>>>>> The "bug" (I don't know if this is a bug) is that a TImage behind of
>>>>>> a instance of my polyline control don't receive the click messages, but a
>>>>>> TButton instance receives.
>>>>>>
>>>>>> So, I don't know if this is a bug or I have to do something in my
>>>>>> control to send these messages to TImage behind.
>>>>>>
>>>>>
>>>>> I had once a similar issue with a tPanel behind another control, in
>>>>> Delphi. IIRC, I indeed had to forward the messages in some way to the
>>>>> tPanel. I can check in my archives, if you want, but I won't be able to do
>>>>> it before tomorrow.
>>>>>
>>>>
>>>> I just checked. Sorry, my memory was wrong, I did use TButtons. But it
>>>> makes sense that the uppermost control would prevent any deeper control
>>>> from receiving clicks, even if the uppermost control is visually
>>>> transparent.
>>>>
>>>
> --
> Frederic Da Vitoria
> (davitof)
>
> Membre de l'April - « promouvoir et défendre le logiciel libre » -
> http://www.april.org
>



-- 
The best regards,

Fabio Luis Girardi
PascalSCADA Project
http://sourceforge.net/projects/pascalscada
http://www.pascalscada.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20141122/c1079c87/attachment-0003.html>