Re: [Plib-devel] Creating shared libs or plib under Linux

[Hans de G. <j.w...@hh...>]

steve wrote:
> Hans de Goede wrote:
> 
>> The
>> exact same breakage will happen with plain C when you exchange the
>> position of 2 variables in a structure.
> 
> Yes - but very, very few C API's use structures.  In the whole of
> the C standard library there are maybe two.  Things like the
> internals of a FILE are typically hidden.
> 

There are many C API's which use structures I'll gladly admit I know
little about C++ but I can dream in C-code and this simply is not true.

> You just can't do that in C++ and yet still get the benefits of
> a class interface.
> 

<snip>

> No - this is how a C++ library has to be maintained.
> 
> There aren't a whole lot of heavily C++ libraries out there with the
> class objects exposed.   Those that do suffer this problem.
> 
Thats untrue, see QT/KDE for example or the STL or many others which all
work fine with shared-objects and C++.

But as I already said:
>> lets end this with agreeing on the fact that we disagree.

I think thats all there is to it. You are not willing to guarantee a
stable ABI, then downstream (Fedora or whatever) should not depend on a
stable ABI and thus I won't. As said before I've used the full plib
version/release as soname, so if a newer plib gets build it will have a
different soname, guaranteeing proper working (or a linker error giving
a clear message about the problem).

All I'm asking now (and all thats keeping this discussion going) is that
you do not defend your stance with clearly false statements like the
ones above.

<snip>

> If someone comes to us with mysterious errors our reaction will be:
> 
>    "Oh dear - your copy of PLIB has been installed incorrectly
>     by Fedora (or whoever) - I suggest you uninstall it and put
>     it back correctly.  If you still have a problem - come back
>     to us."
> 
> ...we've seen these kinds of fiasco's many times before from the
> RedHat guys - both C++ and Mesa have been grabbed from CVS repositories
> in an unfinished or unreleased state and stuffed into their distro.
> I'd hoped that PLIB wouldn't suffer the same fate...but there isn't
> a good defense of that for an OpenSource library team.

1) I'm not RedHat but a volunteer contributer
2) The C++ compiler and Mesa things are off topic
3) Debian (which is known for its stability) has been doing this for
ages so solely targeting Fedora for this isn't helping this discussion,
instead you should be glad that I'm trying to work this out with you
instead of shipping modified packages without ever letting you know
which Debian appearantly is doing.

Regards,

Hans