[Laszlo-dev] UPDATED For Review: Change 20090203-hqm-r, Summary: Fix for missing key up events in swf runtime

[André Bargull]

If that problem is limited to MacOS, you should also restrict that patch 
to MacOS, shouldn't you?
Something like:
const isMac:Boolean = flash.system.Capabilities.os.indexOf("|Mac OS") != -1;
if (isMac && kC == this.keyCodes.control) {
  ...
|

> previous change failed to actually send the keyUp event for the  
> command key itself.
>
>
>
> Change 20090203-hqm-r by hqm at badtzmaru.home on 2009-02-03 13:16:39 EST
>      in /Users/hqm/openlaszlo/trunk-clean
>      for http://svn.openlaszlo.org/openlaszlo/trunk
>
> Summary:  Fix for missing key up events in swf runtime
>
> New Features:
>
> Bugs Fixed: LPP-7012
>
> Technical Reviewer: max
> QA Reviewer: (pending)
> Doc Reviewer: (pending)
>
> Documentation:
>
> Release Notes:
>
> Details:
>
> In swf-runtime/MacOS, if you press a key sequence like 'command' 'v'
> (paste) or 'command' 'c' (copy), the kernel will not receive a key up
> event for the 'v' or 'c'. It does however get the 'command' key-up
> event, so this patch makes the keyboard service clear all down keys
> from it's downKeysHash table when it gets a 'command' key up event.
>
> This fixes the problem in the debugger where keyboard commands like  
> 'enter'
> no longer respond.
>
> It does have the effect of making the model of the down keys slightly
> wrong; if the user presses 'v', then 'command', and then releases the
> 'command' key, the 'v' key will be regarded as having gone up also.
>
> But I think that this is better than the system thinking that the key
> remains down indefinitely, which screws up the behavior LzCommand
> bindings.
>
>
> Tests:
>
> + debugger input field still responds to "enter" and up and down arrow  
> keys, even
> after using copy (command-c) and paste(command-v) keyboard commands in  
> MacOS/SWF runtime.
>
>
>
> Files:
> M      WEB-INF/lps/lfc/services/LzKeys.lzs
>
> Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20090203-hqm-r.tar
>
>