When should cmd-period work?
J. Landman Gay
jacque at hyperactivesw.com
Mon Feb 2 19:46:35 EST 2009
Devin Asay wrote:
> On Feb 2, 2009, at 2:44 PM, J. Landman Gay wrote:
>> I accidentally wrote a handler that called another handler that called
>> the first one, which called the second one, which called the first one,
>> which...you know, like that. It wasn't really recursion because each
>> handler did a finite set of actions and didn't depend on the other. At
>> any rate, there was no recursion warning.
>> What it looked like from the outside was that Rev had become
>> unresponsive, though the cursor moved. No spinning beach ball, no
>> colored pizza, just the regular arrow, but clicking on anything failed
>> and the OS X dock had "Force Quit" in its menu. It took me several
>> force-quits before I figured out the problem was ID10T error.
>> The allowInterrupts was true, but command-period didn't intercept it.
>> Should it have? I wonder if that's even possible, since from the
>> script's point of view there was nothing wrong. It was doing exactly
>> what I told it to.
> Hi Jacque,
> This has happened to me several times in Rev 3.0, but I've never been
> able to pin it down to a recipe. It's almost like what happens when a
> running script throws an error and kicks you into debug mode, and you
> try to do things with the interface, but nothing works. Then you notice
> the error message and slap yourself in the forehead and say to yourself
> "What a dork!", then click the stop button in the script editor to get
> out of debug mode, and everything's fine. But when this happens you
> can't find anything to click, and you want to slap your computer
> instead. Cmd-period doesn't work because the script is already
> interrupted. It's almost like you really are in debug mode but the
> interface hasn't been updated to reflect that. Like you, I seem to
> notice it happen when handlers are calling handlers are calling
> handlers, in different objects.
> It's very frustrating; maybe if we compare notes we can come up with a
> reliable recipe for reproducing it.
I'll see if I can make a simpler stack that reproduces it. The one I'm
working on isn't fit for pubic consumption. :) I never thought about it
acting like it's in debug mode, but now that you mention it, yeah,
that's what it seems like. Only I never got any error messages that I
I figured out real quick to save before testing...
Jacqueline Landman Gay | jacque at hyperactivesw.com
HyperActive Software | http://www.hyperactivesw.com
More information about the Use-livecode