responding to 'inactive - what's the catch?

Here's a question for the VID/View gurus on the list. I want to hide my window when it becomes inactive. Catching the event is no problem, but there's some little trick I'm missing to make it work. The following code works great if you paste it into a View console session, and it also works correctly if you uncomment the print statements in slide-on/off but, as it is below, it doesn't quite work. Without a console open, the deactivated window doesn't hide until you move the mouse over it. What does the console cause to happen to make it work? Thanks! --Gregg REBOL [] evt-func: func [face event][ switch event/type [ inactive [slide-off] active [slide-on] ] event ] if not find system/view/screen-face/feel/event-funcs :evt-func [ insert-event-func :evt-func ] shutdown: does [ remove-event-func :evt-func quit ] slide-off: does [ ;print ["sliding off" main-lay/offset] main-lay/offset: to-pair reduce [negate subtract main-lay/size/x 4 -22] show main-lay ] slide-on: does [ ;print ["sliding on" main-lay/offset] main-lay/offset: 0x-22 show main-lay ] main-lay: layout [ button "Quit" [shutdown] ] view/offset main-lay 0x0

On 23-May-02, Gregg Irwin wrote:

> Here's a question for the VID/View gurus on the list. > I want to hide my window when it becomes inactive. Catching the > quite work. Without a console open, the deactivated window doesn't > hide until you move the mouse over it.

> What does the console cause to happen to make it work? > Thanks! > ] > view/offset main-lay 0x0

Hi Gregg, On Thursday, May 23, 2002, 4:51:46 AM, you wrote: GI> Here's a question for the VID/View gurus on the list. It looks like the window is not actually updated until the REBOL process gets some event from Windows. When you have the console opened, it is getting timer events (cursor flashing, etc), and it works. Adding a timer to the layout like: main-lay: layout [ button "Quit" 0:0:1 [shutdown] ] makes it work as expected too. I didn't test it on the new beta or IOS, so I don't know if it goes better there, but I think it doesn't because it looks like a Windows issue. Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

Thanks Gabriele! Don't know why I didn't think to try that myself. Doh! :) Yes, Carl, the idea is just to make it slide off to the edge so only a tiny little bit is exposed, like an application desktop bar in Windows. --Gregg

Hi Gregg, On Thursday, May 23, 2002, 11:20:55 PM, you wrote: GI> Don't know why I didn't think to try that myself. Doh! :) I thought of it because I had a similar problem with reading from the serial port, where the application didn't receive data until the user moved the mouse or something. I just put a timer and it worked; now I'm wondering if the timer needs to be "fast enough" or if it is sufficient to have it set... GI> Yes, Carl, the idea is just to make it slide off to the edge so only a tiny GI> little bit is exposed, like an application desktop bar in Windows. He couldn't see that, because on the AmigaOS windows cannot go out of the screen. Regards, Gabriele. -- Gabriele Santilli <[g--santilli--tiscalinet--it]> -- REBOL Programmer Amigan -- AGI L'Aquila -- REB: http://web.tiscali.it/rebol/index.r

<< now I'm wondering if the timer needs to be "fast enough" or if it is sufficient to have it set...>> I'm not sure because I think there's a bug that causes them to trigger immeidately upon creation. Once that one message comes in, you're OK, so setting it to a very long interval should be fine. --Gregg