Aaron Bannert wrote:
>
> On Tue, Mar 12, 2002 at 05:30:27PM -0000, [EMAIL PROTECTED] wrote:
> > jfclere 02/03/12 09:30:27
> >
> > Modified: . Makefile.in
> > Log:
> > make test fails because there is no subdirectory test in test.
> >
> > Revision Changes Path
> > 1.71 +0 -1 apr/Makefile.in
> >
> > Index: Makefile.in
> > ===================================================================
> > RCS file: /home/cvs/apr/Makefile.in,v
> > retrieving revision 1.70
> > retrieving revision 1.71
> > diff -u -r1.70 -r1.71
> > --- Makefile.in 11 Mar 2002 17:29:06 -0000 1.70
> > +++ Makefile.in 12 Mar 2002 17:30:27 -0000 1.71
> > @@ -123,7 +123,6 @@
> >
> > test: $(TARGET_LIB)
> > (cd test; make clean; make; \
>
> Why don't we just get rid of the above line and keep the next? IOW, why
> must we clean every time we test?
I have noted there is more to do: at least one test needs input.
And having the library as target of the directory is not good.
Probably the following is only acceptable:
test:
(cd test; make)
But of course that means the logic must be added in the makefile of the test
subdirectory.
>
> > - cd test; \
> > for prog in `find . -type f -perm +u+x -name "test*" -print`; do \
> > ./$$prog; \
> > if [ $$? -eq 255 ]; then \
> >
>
> -a
