Ralf Hemmecke wrote:
> 
> First. I'm sorry for my last mail. I wrote it while in a seminar. :(
> 
> > Well, the point is that the only affected file is 'warm.data':
> > in 'clean' we want to remove all files produced during build,
> > but preserve file present in original tree.  So, we do not
> > care what exactly is matched by shell glob.  We care that
> > files we want to delete match and no original source files
> > match.
> 
> Ah, that's different. Actually, we shouldn't use globs at all. What if 
> I've put in the build directory some file. I wouldn't want it to be 
> removed by "make clean". "make clean" should only remove what "make" 
> generated, not more.
> 
> I would be quite in favour to fix this and would be willing to invest 
> time in it.
> 

I do not think this is feasible.  Our build system depends on
naming convention.  For example, there is no *.o files in the
build tree and all such files are generated.  'warm.data' is
a specific case because on one hand the '.data' extension is
reasonable for this file, OTOH clisp creates files with .data
extension which needs to be removed by clean.  I do not think
it makes sense to try to modify build system so that you can
add a random .data file without risk of this file beeing deleted
by clean.  Simply, we have too many real problems to work on
to waste time on problems which are only potential one (and
of quite low probability to turn real).

-- 
                              Waldek Hebisch
[email protected] 

-- 
You received this message because you are subscribed to the Google Groups 
"FriCAS - computer algebra system" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/fricas-devel?hl=en.

Reply via email to