Your changes had better be backwards compatible with the existing <cvs>
task - and support the 'command' attribute too - sounds like you have this
covered. Otherwise you're wasting your time! :) I think if you need
addConfiguredX you're probably doing something funky and unnecessary, but
I'll hold off comment until I see the code.
Erik
----- Original Message -----
From: "stephan beal" <[EMAIL PROTECTED]>
To: "Ant Developers List" <[EMAIL PROTECTED]>
Sent: Sunday, March 24, 2002 12:09 PM
Subject: Re: Cvs <commandline> implementation question...
> On Sunday 24 March 2002 16:52 pm, Erik Hatcher wrote:
> > ----- Original Message -----
> > From: "stephan beal" <[EMAIL PROTECTED]>
> > > Is implementing both createX() and addX() considered evil (or just bad
> > > style)?
> >
> > What would be the point? Only one will get invoked if "X" is the same.
>
> i've come across a use for it, but i'll admit that it could also be done
> another way. i've done Considerable refactoring of AbstractCvsTask, and to
> make sure i get the noexec/quiet/etc options in the command line in the
> correct place, i really need to do that part of their configuration in
> createCommandline(). However, because of the implementation i also need to
do
> some work in addConfiguredCommandline(). The reason for this bit is
backwards
> compatibility: i internally convert the old 'command' into a <command> and
> insert it at the beginning of the Commandline vector. That way it's
treated
> 100% consistently with cdata and <commandline> args.
>
> i'm essentially done with the <commandline> support, but the major problem
> with the implementation right now is that on the default 'command' i can
get
> the args in the right order (-n/-q first), but on <commandline> args, i'm
> getting the objects too late for that (in addConfiguredCommandline()).
i've
> gotta run for a movie in a bit, but after that i'll come back and submit a
> draft patch for anyone who's interested in commenting on the
implementation
> and refactoring.
>
> > And like Adam said, addX is preferred so that subclassed (!!) types can
be
> > substituted by Ant rather than the concrete class in the addX signature.
> > That opens up a lot of cool options.
>
> Understood. :)
>
> ----- stephan
> Generic Universal Computer Guy
> [EMAIL PROTECTED] - http://www.einsurance.de
> Office: +49 (89) 552 92 862 Handy: +49 (179) 211 97 67
> Student: "Master, you must teach me the way of liberation!"
> Master: "Tell me who it is that binds you."
> Student: "No one binds me!"
> Master: "Then why do you seek liberation?"
>
> --
> To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
>
>
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
