| function graphics event handling [message #90434] |
Mon, 02 March 2015 13:30  |
Helder Marchetto
Messages: 520
Registered: November 2011
|
Senior Member |
|
|
Hi,
I find my self often a bit puzzled by how the fg event handling works. Events such as mouse pressed or so, are handled with functions that with a call to "return, 1" continue the processing by the fg.
But what if I would want do insert some code after the processing of the object?
Let me make a demo example, first in words further down as code.
Say you have a widget_window with an image and an ellipse inside. The window event mouse_down_handler calls a function called MouseDownEvent. I would like to know if the user clicked on the image or on the ellipse. For that I have the getSelect() function. However, this will give you the previous object selected... the new one is selected after the "return,1" is called.
So here is an example:
function MouseDownEvent, oWin, xPos, yPos, Button, KeyMods, Clicks
;here I would like to know which object has been selected
help, oWin->getSelect()
;but this will always be the last one selected!!!
return, 1
end
pro testGetSelect
img = dist(500)
tlb = widget_base(/column)
wWindow = widget_window(tlb, xsize=500, ysize=500,mouse_down_handler ='MouseDownEvent')
widget_control, tlb, /realize
widget_control, wWindow, get_value=oWin
io = image(img, image_dimensions=[500,500], current=oWin, margin=0)
;make some objects on top
el = ellipse(0.25,0.75, major=0.125, '-r2', fill_background=0, /normal, target=io)
sq = polygon([0.2,0.3,0.3,0.2], [0.2,0.2,0.3,0.3], '-y2', fill_background=0, /normal, target=io)
rect = polygon([0.7,0.8,0.8,0.7], [0.2,0.2,0.4,0.4], '-b2', fill_background=0, /normal, target=io)
end
I will make my final question in the next few lines, but first I will say that one can use hitTest() to check which element "could" be selected.
However, hitTest() seems to be non-consistent with getSelect(). When I have three objects intersecting, it's not clear which will be returned by hit test and which will be returned by getSelect(). I had an object selected (corners highlighted) that was the second in the array of three returned from hitTest(). Figure that.
To check the effect of hit set, substitute or add in MouseDownEvent function these lines:
res = oWin->hitTest(xPos, yPos)
for i=0,n_elements(res)-1 do help, res[i]
And here is the question: Why not giving the possibility to act/interact *after* fg processing?
Is this a goofy idea?
Thanks,
Helder
|
|
|
|
| Re: function graphics event handling [message #90435 is a reply to message #90434] |
Mon, 02 March 2015 13:44  |
Helder Marchetto
Messages: 520
Registered: November 2011
|
Senior Member |
|
|
On Monday, March 2, 2015 at 10:30:58 PM UTC+1, Helder wrote:
> Hi,
> I find my self often a bit puzzled by how the fg event handling works. Events such as mouse pressed or so, are handled with functions that with a call to "return, 1" continue the processing by the fg.
> But what if I would want do insert some code after the processing of the object?
>
> Let me make a demo example, first in words further down as code.
> Say you have a widget_window with an image and an ellipse inside. The window event mouse_down_handler calls a function called MouseDownEvent. I would like to know if the user clicked on the image or on the ellipse. For that I have the getSelect() function. However, this will give you the previous object selected... the new one is selected after the "return,1" is called.
>
> So here is an example:
>
> function MouseDownEvent, oWin, xPos, yPos, Button, KeyMods, Clicks
> ;here I would like to know which object has been selected
> help, oWin->getSelect()
> ;but this will always be the last one selected!!!
> return, 1
> end
>
> pro testGetSelect
> img = dist(500)
> tlb = widget_base(/column)
> wWindow = widget_window(tlb, xsize=500, ysize=500,mouse_down_handler ='MouseDownEvent')
> widget_control, tlb, /realize
> widget_control, wWindow, get_value=oWin
> io = image(img, image_dimensions=[500,500], current=oWin, margin=0)
> ;make some objects on top
> el = ellipse(0.25,0.75, major=0.125, '-r2', fill_background=0, /normal, target=io)
> sq = polygon([0.2,0.3,0.3,0.2], [0.2,0.2,0.3,0.3], '-y2', fill_background=0, /normal, target=io)
> rect = polygon([0.7,0.8,0.8,0.7], [0.2,0.2,0.4,0.4], '-b2', fill_background=0, /normal, target=io)
> end
>
> I will make my final question in the next few lines, but first I will say that one can use hitTest() to check which element "could" be selected.
> However, hitTest() seems to be non-consistent with getSelect(). When I have three objects intersecting, it's not clear which will be returned by hit test and which will be returned by getSelect(). I had an object selected (corners highlighted) that was the second in the array of three returned from hitTest(). Figure that.
> To check the effect of hit set, substitute or add in MouseDownEvent function these lines:
> res = oWin->hitTest(xPos, yPos)
> for i=0,n_elements(res)-1 do help, res[i]
>
> And here is the question: Why not giving the possibility to act/interact *after* fg processing?
>
> Is this a goofy idea?
>
> Thanks,
> Helder
Ok, maybe for such things I should just use the mouseUp instead of mouseDown.
...mmm...
Helder
|
|
|
|
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
