Hi Przemek,
Ok, I understand, and in the same line that my previous post for if
it
could be more easier without making many changes.
And if we encode pHrbPtr in PHB_SYMBOLS szModuleName f.e.
"pcode.hrb" +
'\0' + pHrbPtr in hb_hrbReturn() and in hb_vmDoExitFunctions()
identified it and decoded for set pHrbPtr->fExit = FALSE,
PHB_SYMBOLS is
secuential. Does it be possible that solvent all the things?
Yes, it's one of possible solutions. Probably the easiest one but
it resolves only problem with .hrb files. It's possible that
I'll implement sth like that to not make deeper modifications in
core code before 1.0 release. I'll have to think about it.
But there is still open question if we should allow to register
new modules when HVM is in QUIT state.
Please remember that similar situation can appear with modules
loaded/unloaded from .dll/.so/.dyn or other shared libraries.
In practice recent modification doesn't change anything here
and the problem exists from the beginning. Defining that HVM
will not accept new modules when it enter EXIT state seems to
be reasonable for me.
It seems a reasonable restriction, and most probably won't
limit real world usage of this feature. And it's definitely
better than leaving it undefined or buggy.
Brgds,
Viktor
_______________________________________________
Harbour mailing list
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour
