I hacked and slashed and have succeeded in building the wrapper and making a new version of horae. It is getting rather late here so I will have to turn in, but I think the idea of making a different Makefile for the wrapper is a good one. As it didn't seem to be particularly difficult, I assume that I will have no trouble putting one together. I will post a new tarball later tommorrow on sourceforge with a new Mac installer (power pc). This problem with the tarball was new -- there was no conflict before (or was there). The error associated warning message occurred when tar attempted to write the file "atoms" on top of the directory "Atoms". I wonder if the archive order was different before hiding the conflict? Paul On Dec 6, 2006, at 11:59 PM, Bruce Ravel wrote:
On Wednesday 06 December 2006 08:30, Paul Fons wrote:
Alas Bruce, not everyone has a unix without training wheels. The darwin OS is case insensitive. Sort of a good thing in many ways. For instance take the example of "atoms" and "Atoms" in the horae-063 directory. This frozzed tar in the unpacking stage and I didn't realize the conflict until today (I was in Italy until last Friday so didn't get a chance to compile horae until today).
Hmmm... not so trivial. But I'll try to have a trial tarball for you later today.
Can you remove the conflict with uppercase and lowercase "atoms/ Atoms" in the horae directory so this doesn' happen every time. It would also be sort of neat if there were a some sort of "honest" way to compile the ifeffit_wrapper.c source without having to hack the ucky MakeMaker output (a switch perhaps). All this is of course, when you have time. Thanks for the new release.
If I knew how to honestly compile the wrapper, I'd be happy to do so.
Are you able to compile other C code from perl packages "out of the box"? For example, if you downloaded (to suggest an example that horae uses) Compress::Zlib and did the typical perl-y build incantation, does it run to completion without further input? If so, then I can simply remove all darwin-specific code from my makefile.PL. If not, then you need to tell me what to do.
One option is for you to provide me with a makefile that builds ifeffit_wrapper. I'll add that to the tarball and simply have Makefile.PL skip the compilation of ifeffit_wrapper. Then your external makefile can be used to build ifeffit_wrapper. That adds a step to the build incantation, but that's not so bad.
Perhaps better, I can simply skip the build of ifeffit_wrapper for darwin on the assuption that it is already installed or has been installed some other way. That way, the wrapper can be part of the installer package, not part of the tarball update for darwin. That is, in fact, what my Makefile.PL does for debian.
B
-- Bruce Ravel ---------------------------------------------- bravel@anl.gov
Molecular Environmental Science Group, Building 203, Room E-165 MRCAT, Sector 10, Advanced Photon Source, Building 433, Room B007
Argonne National Laboratory phone and voice mail: (1) 630 252 5033 Argonne IL 60439, USA fax: (1) 630 252 9793
My homepage: http://cars9.uchicago.edu/~ravel EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/
participants (1)
-
Paul Fons