1. Re: IDE and copyBlt
- Posted by Derek Parnell <dparnell at BIGPOND.NET.AU>
Nov 27, 2000
-
Last edited Nov 28, 2000
On Mon, 27 Nov 2000 00:25:09 -0600, Kat <gertie at PELL.NET> wrote:
>
>Gosh, i am so glad Ralf made this suggestion for me.
>
Ralf, can you kind pass on the Ms. Kat that ...
I understood Kat to be saying that the application should be able to tell
win32lib how to behave. And I quote ...
> Just a suggestion, if converting a minor version to another is that easy,
> what if the coder could simply tell the win32lib what version to act like?
> That way, you have complete backwards compatability, and programmers
> could spec "use the latest win32lib", rather than one of the hundreds of
> 0.xx versions out there now? I know you said the changes were to the IDE
> code, but still.....
>
> Kat
>
but Ralf seemed to be suggesting that the library package include a
facility to generate code, specific to a nominated version, for use in
applications. To quote ...
>Doesn't have to be impossible, I think. Perhaps using a special code
>generation library that is part of each win32lib package rather than part
>of the IDE. Each and every IDE can than use this library to generate code,
>of whatever win32lib version.
>
>Sounds like a good idea ?
>
>Ralf N.
>nieuwen at xs4all.nl
The two ideas, are similar but NOT THE SAME. My reply to Judith was dealing
with Ralf "code generation" idea, and not optional run-time behaviour based
on old versions.
I explained that backward compatibilty is not (yet) one of my goals - the
library is still too unstable.
My suggestion of "getFuncInfo()" was specifically made to help IDE writers
generate the code to interface with win32lib. I imagine that it could also
help you with "the calling program can verify it has a win32lib that it can
use" issue. However, this still would not help changed behavior inside a
global function.
I'll try to see if I can implement runtime behaviour based on nominated
versions AFTER the release of v0.55. No promises though.
-----
cheers
Derek
