mousemove doesn't recognize "shiftkey is down", but only under Windows

[Howard Bornstein]

I just wanted to mention that the problem detecting the shift-key being
down seems to be entirely an issue with Parallels. My test, running under
Windows 10 on a real PC had no trouble with my code that was looking for
the shift-key being down. I'm glad to see this problem wasn't within
LiveCode. There are so many layers where something can go wrong, it can be
a challenge identifying exactly where the problem is happening. It must be
a pretty specific issue with Parallels that only manifests by ignoring the
shift key.

On Tue, Sep 25, 2018 at 10:02 PM Howard Bornstein <bornstein at designeq.com>
wrote:

>
>
> On Sun, Sep 23, 2018 at 10:50 PM Curry Kenworthy via use-livecode <
> use-livecode at lists.runrev.com> wrote:
>
>>
>> Howard:
>>
>>  > All my test show that for the Shift key and the Option key, when
>>  > called within a mousemove handler, it doesn't register that the
>>  > keys are down unless the mouse button is clicked first.
>>
>> Hi Howard, I tested your code verbatim in the card script of a new stack
>> in LC 901 on native Windows 10 with a real PC laptop, Dell Intel i5.
>>
>> It works fine here.
>
>
> Thanks for testing this out.
>
>
>> I also noticed that since "shiftkey" only has two
>> states, you don't really need to test it twice; you could cut out the
>> second "if" statement. In fact, this yields similar results:
>>
>> on mousemove x,y
>>   put x,y && the shiftkey
>> end mousemove
>>
>
> Lol. Yeah, I wasn't trying for elegance. I was making sure the code and
> results were absolutely clear.
>
>
>> So, despite the very significant issues that LC 9 has on real Windows,
>> including a barely-usable script editor, it seems likely that this is
>> not one of them. That's good news! But obviously you do have a problem.
>>
>> Have you tested your code in a brand new empty stack in a fresh and
>> unmodded instance of LC, no extra plugins or scripts running? If so, if
>> it's just Parallels, I have to say that virtual Windows isn't quite the
>> same. Well, I guess an Intel Mac sporting a GUI that contradicts much of
>> Apple's original research isn't quite the same either; I'm a big
>> original Mac guy! :) But some details are different. That's why I always
>> use real Mac and real Windows.
>>
>
> I have friend with a real Windows machine that will test the code of the
> app where I applied this technique so, hopefully I'll be able to see if it
> works on a real PC and if the problem is Parallels. I'm a Mac guy too and
> only use Windows to test apps built for both platforms.
>
>
>> But if you get different results with a fresh stack and LC, if there is
>> something else going on and you would like me to test your actual stack
>> (on native Windows) or help solve your problem, feel free to contact me
>> off-list. Interesting.
>>
>
> Thanks. I'll let you know what my friend turns up.
>
>
>> (Meanwhile, hope we can get improvements for some of the real Windows
>> problems that LC has. It's difficult to type in the LC 901 script editor
>> on Windows - that is a pretty huge issue. I have a WordLib update coming
>> out in a few days, pending final tests, and it never hurts to have a
>> fully usable version of the latest LC IDE on both platforms!)
>>
>
> I did work a little bit in the script editor under Windows and didn't
> notice the awful problems you've been reporting, but again, it was just
> some tweaks and code changes. I don't do most of my scripting under
> Windows. I hope this gets fixed.
>
>
>>
>> Best wishes,
>>
>> Curry Kenworthy
>>
>> Custom Software Development
>> "Better Methods, Better Results"
>> LiveCode Training and Consulting
>> http://livecodeconsulting.com/
>>
>> _______________________________________________
>> use-livecode mailing list
>> use-livecode at lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your
>> subscription preferences:
>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>
>
>
> --
> Regards,
>
> Howard Bornstein
> -----------------------
> www.designeq.com
>


-- 
Regards,

Howard Bornstein
-----------------------
www.designeq.com