Hello Przemek

Przemyslaw Czerpak-2 wrote:
> 
> I so not think so. inkey() does not respect SetKey()
> settings. SetKey() needs explicit support to activate
> codeblok attached to given key value. See TBROWSE or
> TGET code. GT does not know anything about it.
> 
>> do while inkey()<>27
>> enddo
>> return nil
> 
> You want to make setkey() settings local to GT driver.
> I do not think it's good idea. It will stop to be thread
> local for threads which shares GT.
> Even if I'll change it then your next question will be
> making THREAD STATIC variables used in GET system also local
> to GT or sth like that.
> I believe that it may help you in some of your code but it
> will also break other code which now may need thread local
> setting because it uses single GT driver in different threads,
> f.e. with GTCTW.
> So it's not a modification which is acceptable for all users.
> I think that you should rather try to use extended CL53 functionality
> and local to given object SetKey() setting (see setkey() methods) or
> even your own function which will use CARGO value in GT driver to keep
> your local GT setting. F.e. you can store hash array in GT cargo item
> and use "SETKEY" index to keep your own SeyKey() array inside.
> 
> 

You are right. I already have figured out the problems if SetKey()
is made GT specific. Thanks for the explanation.

Regards
Pritpal Bedi

-- 
View this message in context: 
http://www.nabble.com/SetKey%28%29-and-hb_gtReload%28%29-tp22444718p22446799.html
Sent from the Harbour - Dev mailing list archive at Nabble.com.

_______________________________________________
Harbour mailing list
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour

Reply via email to