tech-pkg archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: design of SUBST_STAGE=post-install behavior?



On Sun, Sep 13, 2009 at 07:26:38PM -0600, Brook Milligan wrote:
> Joerg Sonnenberger writes:
>  > Using substs in the post-install phase sounds wrong. What is using that?
> 
> As much as this design is problematic, I do not wish to stray too far
> from it, because updates will be that much more difficult.  As a
> result, it seems like post-install substitutions within $PREFIX are
> required.

I do not like such actions in the *-install part because it is very easy
to screw that part up and change things that shouldn't be changed. It
also has a good change of leaving around garbage unintended. Often
enough the vendor scripts also ignore other things like permissions,
mtime etc, which is why the installation is often done to a local
directory under WRKDIR first and the result copied from that.

>  > Try make reinstall.
> 
> Does this do something other than 'make deinstall; make install'?

Yes, it cleans up the cookie and e.g. makes sure that the PLIST is
rebuilt.

Joerg


Home | Main Index | Thread Index | Old Index