[csw-maintainers] zlib package in /testing

[James Lee]

On 05/04/09, 10:42:57, "Roger Håkansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] zlib package in /testing:

> > There is no need for an RPATH as it doesn't use any CSW libs.

> Is it really a problem if it has RPATH set? (and its correctly set)
> "Unsetting" (in general, not just for this package) it would mean that
> each gar package (which don't use any CSW libs) would have to have a
> LD_OPTION rewrite rule, since gar sets it by default?

> I have no problem doing this, but I'm just thinking if were creating
> unnecessary work.

It's not correct as no CSW libs are opened.

It's not a "problem" but if it's the result of the build system it
suggests we are generally making settings with no thought.  The
current batch of odd RPATHs only causes the link loader to look for
a directory that doesn't exist, but this is less wasted effort than
what happens with $ISALIST which typically looks at all the 64 bit
locations before finding that matching 32 lib.


How difficult is it to do right?  Perhaps the recent batch of odd
paths are telling me that people are using a "sausage machine"
approach to building (throw anything in the top and just wind the
handle).  We are dangerously close to the "I'd like to do the right
thing but the build system won't let me".  The value CSW should add
*is* to take the time to get things right - your individual effort
shared for the benefit of many.



>  > Please set an SONAME.
>  >
> > $ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
> > [4] SONAME          libz.so.1
> >

> Fixed.

Thank you.




James.