On Thu, 09 Oct 2008, Pritpal Bedi wrote:

Hi Pritpal,

> What comes to my mind is: 
> _beginthreadex() accepts a parameter as state of the thread:
> initflag : Initial state of a new thread (0 for running or CREATE_SUSPENDED
> for suspended); use ResumeThread to execute the thread
> Only this may be important. But not needed AFAIT.

This is not portable extension.
PTHREAD intentionally does not have such functionality because
it forces additional overhead in thread startup code.
If you want to run thread in suspended mode to resume it later
then simply use mutex at thread startup.

> I will look deep into Xbase++ documentaion to understand what they 
> are doing. BTW, it seems OK to have such functionality. May be this
> behavior is simulated via access/assign methods.

It's not a problem of access/assign methods and using them
does not change anything.
The problem is in generating code which will be different
for o:var and o:var(), f.e. You have such simple code:

   function f(o)
   return o:var

and in second .prg file:

   function f(o)
   return o:var()

Compiler does not know anything about class declaration so ACCESS/
ASSIGN methods does not exist for it here.
It has to generate code which will inform HVM the in 1-st case we want
to access instance variable and in second one execute method so the code
has to be different.

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

Reply via email to