The variables cover my needs in this area. I think that actual
implementation will result in some proliferation though because it's
considerably easier two implement a variable with two states in a make
script than one with poly states. :)
Also, in playing around with building a bsd.java.mk, I've already found
that defining variables like:
JAVA_CMD # full path to selected java
JAVAC_CMD # full path to selected javac
JAR_CMD # full path to selected jar
JAVA_LIB # full path of directory holding tools.jar or
become critically useful once you get to attempting an implementation.
Selection Order --
The runtime ordering makes great sense--theoretically. Practically,
it's effect is the same as the current use of dependency on JDK 1.1.8.
I believe it makes sense to implement the full rigor of this selection
process only after Sun and the Foundation have arrived at an agreement
that allows higher-order JDK's and JRE's to be built AND distributed by
FreeBSD. Until that time, rules 3 and 4 should be skipped.
But, we're making progress!
Now, we need some competing implementations so the*details can be
> Just updated the proposal containing the requirements for a
> bsd.java.mk, see:
> Any comments would be highly appreciated! John? Chris? Maxim? Kees
> Ernst de Haan
> EuroNet Internet B.V.
> "Come to me all who are weary and burdened
> and I will give you rest" -- Jesus Christ
jmc || MacroHard -- \
|| the perfection of form over |
----------------------------------|| substance, marketing over |
Web: http://www.veryComputer.com/ || performance, and greed over |
|| design . . . |
Public Key: http://www.veryComputer.com/; |
with "unsubscribe freebsd-java" in the body of the message