Re: What's the current theory about derived files in VPATH builds?

On mån, 2010-01-04 at 18:49 -0500, Tom Lane wrote:
> It seems like VPATH building is broken at the moment, at least if you
> are working from a maintainer-clean source tree.
> 
> make -C bootstrap all
> make[3]: Entering directory `/home/postgres/build/src/backend/bootstrap'
> /usr/local/bin/bison  -o bootparse.c /home/postgres/pgsql/src/backend/bootstrap/bootparse.y
> /usr/local/bin/flex  -o'bootscanner.c' /home/postgres/pgsql/src/backend/bootstrap/bootscanner.l
> gcc -O1 -Wall -Wmissing-prototypes -Wpointer-arith -fno-strict-aliasing -g
-I/home/postgres/pgsql/src/backend/bootstrap-I../../../src/include -I/home/postgres/pgsql/src/include
-D_XOPEN_SOURCE_EXTENDED  -c -o bootparse.o bootparse.c
 
> /home/postgres/pgsql/src/backend/bootstrap/bootparse.y:401: bootscanner.c: No such file or directory
> make[3]: *** [bootparse.o] Error 1
> 
> I thought the current plan was that derived files should be generated in
> the build tree (as indeed it seems to be doing) ... but there are
> apparently parts of the Makefiles that are not in sync with this plan.
> Has this just not been tested recently, or am I confused?

This should in principle work.  My guess is an old make version being
confused.