Tom Lane wrote:
> I think we've already found out that the opinions *aren't* equal.
> So far the score seems to be:
> Red Hat: will use relative symlink
> Solaris: will use hardwired path in program
> ???: will use absolute symlink
> Peter's patch supports only the third case, which no packager has yet
> committed to use (though I suppose Debian might like to). It just
> gets in the way for the other two cases.
We can certainly try to develop a solution that addresses the
requirements of various camps.
So far, the requirements I've heard are:
1. Symlinks must be relative.
2. Symlink must have a different name from a previous directory.
3. make check still has to work.
I have not heard the requirement
4. Symlinks must not be used; we must hardwire all paths.
although I'm willing to discuss it if there is some rationale given
(other than, "my previous patch did this").
FWIW, hardwiring plus an environment variable would seem to address all
currently known und unknown requirements, and is not so totally
different from solutions to previous, related problems.
--
Peter Eisentraut
http://developer.postgresql.org/~petere/