[lsb-discuss] LSB conf call notes for 2008-07-23

Wichmann, Mats D mats.d.wichmann at intel.com
Wed Jul 30 07:47:23 PDT 2008


>  Ted: tags?
> Jeff: sent to list.  Ted: what is the goal?  If we want a minimal
> subset that's easy for ISVs to use and portable, many of those tags
>   don't make sense. Jeff: some feature requests, also difficulty in
> building.

There's a difference between adding the tags to the LSB minimal set,
and adding them to the LSB database so that they're recognized and
we can do sane things (in terms of warnings, messages, etc.) when
seeing them.  The suggestion arising out of Jeff's list was the latter,
not the former.

We need to understand what the non-LSB tags mean, so that we're able
to determine whether the use of a particular one in a candidate package
is a problem or not.  If in the course of doing that we identify some
that should be in the spec, well that's not much different than
re-evaluating an included library to see if there are new symbols that
should be added to LSB - just a normal part of maintaining the
spec over time.

> Ted: multiple solutions: document and test tags, writing a
> simple RPM creation tool, post-processor for RPMs to remove the tags.

Let's not get into modifying rpms - that's a bad idea. We
either restrict how they're built, or we get smarter about
what's allowed so we can better handle what comes out of
a native rpmbuild.  This is similar in concept to how we
have to stay on our toes to handle code being generated by
newer native compiler toolchains.




More information about the lsb-discuss mailing list