> > I am expecting to hear some bleating about this from people whose
> > preferred platforms don't support symlinks ;-). However, if we don't
Well, one option would be to have the low level filesystem storage (md.c?)
routines implement a kind of symlink themselves. Just a file with a special
magic number followed by a path.
I'm normally against reimplementing OS services but symlinks are really a very
simple concept and simple to implement. Especially if you can make a few
simplifying assumptions: they only ever need to appear as the final path
element not as parent directories and tablespaces don't need symlinks pointing
to symlinks. Ideally postgres also doesn't need to implement relative links
either.
--
greg