Re: avoid creating sys/time.h replacement when possible

[Jim Meyering]

Bruno Haible <address@hidden> wrote:
> Jim Meyering wrote:
>>   - use $(SYS_TIME_H) also in the definition of MOSTLYCLEANFILES
>
> This is not a good change. Suppose a user does in the same directory
>   - first, a "./configure; make" for a deficient platform,
>   - then, a "./configure" for a glibc platform
>   - then he notices that he forgot to "make distclean" first, so he does
>     "make distclean",
>   - then, again, "./configure; make" for the glibc platform.

Good point.
Normally, I would not want a Makefile rule to remove something
it could not create.  But this is a little different.
A "clean" rule has the responsibility to remove all things
any _instantiation_ of a Makefile.in might create.

I've reverted that part, and adjusted the preceding ChangeLog entry:

        * modules/sys_time (Makefile.am) [MOSTLYCLEANFILES]: Do use
        sys/time.h here, reverting that part of the previous change:
        <http://thread.gmane.org/gmane.comp.lib.gnulib.bugs/8959>.

Index: modules/sys_time
===================================================================
RCS file: /sources/gnulib/gnulib/modules/sys_time,v
retrieving revision 1.5
diff -u -p -r1.5 sys_time
--- modules/sys_time    28 Jan 2007 18:55:01 -0000      1.5
+++ modules/sys_time    28 Jan 2007 20:12:58 -0000
@@ -28,7 +28,7 @@ sys/time.h: sys_time_.h
              < $(srcdir)/sys_time_.h; \
        } > address@hidden
        mv address@hidden $@
-MOSTLYCLEANFILES += $(SYS_TIME_H) sys/time.h-t
+MOSTLYCLEANFILES += sys/time.h sys/time.h-t

 Include:
 #include <sys/time.h>