ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Bodewig <>
Subject Re: Backwards incompatible change to Javac
Date Mon, 25 Mar 2002 09:33:51 GMT
On Fri, 22 Mar 2002, Diane Holt <> wrote:

> Now that <javac> has a 'compiler' attribute, couldn't you allow that
> to take a full path to the compiler when fork="yes"?

We could, yes.  Then the rules would be

* compiler is the name of the compiler adapter to use unless fork has
been specified.

* compiler is the name of the executable if fork is used.

Which would again be overloading an attribute with two meanings.  Put
that together with 

<compilerarg ... compiler="extJavac" />

which you'd have to use in forked execution ...

I think a new attribute would be better.  I'll implement it that way
and we can merge them to a single attribute if this should turn out to
be consensus.


To unsubscribe, e-mail:   <>
For additional commands, e-mail: <>

View raw message