2009/11/4 Kelly O'Hair <kelly.oh...@sun.com>: > > I'm basically ok with this, but have one concern. > > There is no longer a default ant value for drops.dir, it is only > set via the Makefile. Maybe that's ok, just not sure. > I think it means that a plain 'ant' command will always download > from the url. Again, maybe that's fine. So I'm ok with the push. >
Ah I see what you mean. I changed things so all the logic was in one place in the Makefile: ifdef ALT_DROPS_DIR DROPS_DIR = $(ALT_DROPS_DIR) else ifdef ALT_JDK_DEVTOOLS_DIR DROPS_DIR = $(ALT_JDK_DEVTOOLS_DIR)/share/jdk7-drops else DROPS_DIR = $(_SLASHJAVA)/devtools/share/jdk7-drops endif endif so you get the same behaviour as before, but as you say, only when running ant via the makefile. Is ant ever run directly in this situation? I've never had reason to do so. Either way, I can add an updated version back to the build.xml. That's not a problem, as we'll never use it. Can you allocate a bug ID so I can push? :) > I've learned to keep ant scripts out of the environment variable > business if at all possible, so leaving any env var logic in > the makefiles may be fine. > > Keeping the ant<->make boundaries simple seems like an essential > thing to do, sigh, but is anything "simple" with ant scripts or > Makefiles... :^( > I know what you mean all too well... :( At one point, I managed to have three different JDKs, all from different vendors, involved in the building of langtools... > -kto > > > Andrew John Hughes wrote: >> >> 2009/10/28 Andrew John Hughes <gnu_and...@member.fsf.org>: >> >> snip.. >> >>> Well spotted; revised version here: >>> >>> http://cr.openjdk.java.net/~andrew/drops/webrev.03/ >> >> Does this look ok to push? > Thanks, -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
