On Fri, 16 Jul 2004, Mark Taylor wrote:
> On Thu, 15 Jul 2004, Tim Jenness wrote:
>
> > Any chance that the itcl parts of ccdpack can be conditionally compiled so
> > that we can use ccdpack in the pipeline at least? [presumaing that makemos
> > doesn't need itcl!]
>
> That seems like a reasonable request - though I don't think it should
> be all that hard to get itcl building on OS X, it would make the whole
> thing more robust for unknown platforms. I'll have a go.
Well I've attempted this, and I think it's probably working, but
I'm unable to do much of a test because the tclsys install
targets are broken on my machine (as per symlink-related problems
that Norman says he's going to look at next week). I thought I
was going to be able to work round this but it seems that I can't
without doing most or all of a make world, and I don't have available
the elapsed time that would require.
What I believe happens is that if itcl is installed, CCDPACK will
build and install as before, and if it's not, CCDPACK will build and
install but XREDUCE will fail (ungracefully) if run.
The other CCDPACK-related A-tasks (PAIRNDF, CCDALIGN, IDICURS)
ought to work either way, though that certainly needs to be tested.
I'm on holiday all next week (no email!) back 26 July. If it's still
in the same state when I get back I'll test and (if necessary) fix it,
though if anyone wants to have a go in the mean time they're more than
welcome.
Mark
PS it occurred to me another approach to getting itcl to build on
OS X is to work out why it's calling TCL_SHLIB_LD and replace it
with something else. After all, Tcl and Tk manage to do without
it, and itcl probably doesn't have more esoteric build requirements
than they do. This would probably require editing
tclsys/itcl/{itcl,itk}/Makefile.in and re-running whatever version
of autoconf (or does it use automake? not sure) is required to
regenerate the Makefiles. This would possibly be more robust than
tweaking tcl.m4.
--
Mark Taylor Starlink Programmer Physics, Bristol University, UK
[log in to unmask] +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/
|