 |
 |
Norman, It also fails on updating a system from the last release. I will have to think of a way around this, but I defiantly need a fixed makefile for the current and previous release versions of AUTOASTROM. Steve. -----Original Message----- From: Starlink development [mailto:[log in to unmask]] On Behalf Of Norman Gray Sent: 18 June 2004 12:53 To: [log in to unmask] Subject: Re: USSC219 rc2 Peter and Steve, On 2004 Jun 18 , at 12.03, Peter W. Draper wrote: > thought I'd try to install this one as a complete update of rc1, rather > than a fresh install. This fails when cleaning the deinstalled > AUTOASTROM. > See the attached log. Blast -- the Curse of the Autoastrom Makefile strikes again. These crop up because the autoastrom makefile is a horrible, horrible, mess, since it tries to configure and build (and install, deinstall and clean) astrom, catlib and match all at the same time, and in an unprincipled way. This is not the first time Steve has had trouble with it (contra Steve, I don't remember seeing this particular error before, but I've possibly just erased the memory). A fix is problematic since I started moving autoastrom and astrom to the Starlink repository a couple of months ago (at last! -- now it will make sense, and not be quite so fragile). They're mostly done, but catlib isn't, because it threw up some problems which are probably common in thirdparty applications. And packaging will be a puzzle, sigh. I have to resolve those before the CVS week, but the upshot is that this isn't distributable from the RAL repository, and to avoid (otherwise inevitable) confusion I removed autoastrom from my local repository. I can get the repository back of course, but I'd really rather not, since it would be pretty expensive in time. Thus if I fix this, I'll just do so by unpacking the distribution tarball and hacking away at the makefile within it (I think I've spotted where the problem is). That is, do you (youse) think we can just leave this error, if everything else works? It's only a clean target, and while I'd rather it didn't happen, I doubt many folk would even notice. See you, Norman -- ---------------------------------------------------------------------- Norman Gray / Physics and Astronomy, University of Glasgow, UK http://www.astro.gla.ac.uk/users/norman/