Re: [gentoo-dev] QA question wrg. GRP

[Spider]

begin  quote
On Thu, 5 Feb 2004 21:57:42 +0100
Paul de Vrieze <pauldv@g.o> wrote:

> On Thursday 05 February 2004 21:34, Spider wrote:
> > begin  quote
> > On Thu, 5 Feb 2004 20:58:01 +0100
> > Can you explain more? When would LIBVER be set? by whom? (developer?
> > no thanks.. :P ) and woudn't there need to be one LIBVER per .so
> > file that a package installs?
>
> In most cases we could probably derive it automatically from the
> soname of a library, in special cases LIBVER would need to be set.
> Basically the idea breaks through the one LIBVER per .so by needing a
> developer to define a different LIBVER when an incompatible library
> change is introduced in the package (in any file) (the contents of
> LIBVER don't matter). In some cases this will be automatically
> detected, in some cases not, in which the developer needs to fix this.
>


I'm still not sure I actually understand how this would hang about.
LIBVER would be created for the package at post-compile time, and would
refer to its own contents... right?

This could work, but in cases like :
foolib.so.0 -> foolib.so.0.5.4

Which is LIBVER?    0, or 0.5.4 ?   since if applications link to .so.0 
the binaries will break if LIBVER is 0.5.4 (due to dependencies)  but if
it provides both, and a program links to the hard minor (0.5.4)  ,
LIBVER would break if it was "0" ...

This falls on the fact that all packages don't necessarily link to the
same LIBVER ,  some may link to .so.0, others may be pickier and link to
`foolib.so.0.5.4`  ..  Not a pretty situation.

<SNIP>

>
> That would be "enhanced rpm style", which would probably work too. It
> would however not solve the problem of determining which versions are
> actually compatible (one could use sonames for that)

No, that wouldn't be solved , but you'd at least have a "safety net"
that catches you when you fall down to binaries. IE  installing
"foopack" won't be allowed if the contents wouldn't work (well, override
flags, but I think that would be a bad thing. simply not having a
--force would avoid some user-generated problems here ;)



>
> > and yes. its dirty. its rpmish.    and I'd love to see a better
> > thing. however, its better than the thing we have currently.
>
> Anything is better than nothing (which we have now) ;-)

hehe :)

 
> > Any ideas?
> Don't have binary packages ;-)

not very practical IMO.  computers aren't that fast, yet. (and with
software like KDE, Mozilla, qt, OpenOffice.. May never be) 

//Spider

-- 
begin  .signature
This is a .signature virus! Please copy me into your .signature!
See Microsoft KB Article Q265230 for more information.
end