ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James Cook" <>
Subject RE: Anteater... I'm Baaaack...
Date Fri, 15 Dec 2000 04:27:31 GMT
If the only object to DOM is overhead then I think we have a topic for
discussion. As a developer, I have a few XML parsers (Xerces, Sun's parser
that is packaged with JAXP, JDOM) installed, and they all support DOM. I
think we have to remember that this is a developer tool. Why restrict its
simplicity and elegance by throwing out a perfectly acceptable library? I
don't understand the objection, especially when you look at Simeon's elegant
use of DOM in Antidote.

> W3C DOM will never get into core. Waaaaaay to[sic] much overhead.

Is this already a done deal? I don't have voting priveleges, but has this
been voted down before?


> -----Original Message-----
> From: Peter Donald []
> Sent: Thursday, December 14, 2000 8:22 PM
> To:
> Cc:
> Subject: RE: Anteater... I'm Baaaack...
> At 08:15  14/12/00 -0500, James Cook wrote:
> >I see round-tripping as a major impetus behind an Ant 2.0. I
> guess I have to
> >see a list of the shortcomings in Ant 1.2! I haven't experienced too many
> >from a user's perspective.
> The only real shortcomings from a users perspective I know of are
> * you have to use build scripts
> * task registration isn't as easy as it could be
> * classpath management is often painful
> However there are things that would be nice features - JPan discussed a
> while back is one of them as is the idea of "Workspaces" that contain many
> ant-projects.
> >I imagine that we should start to collect a list of these
> shortcomings so we
> >(you) know what needs to be addressed in a new version. I'll start:
> >
> >1. Ant build scripts should be parsed from a variety of valid
> input sources
> >into a DOM structure to facilitate Ant integration into GUI editing
> >environments. (let the flames begin) :-)
> W3C DOM will never get into core. Waaaaaay to much overhead. However some
> form of abstracted DOM would be useful. I proposed a Configuration object
> which is a 1-to-1 mapping with Ants idea of object model. AntEater
> presumably uses nested hashtables. Another good option is JDOM thou no one
> warmed up to that idea ;)
> Cheers,
> Pete
> *-----------------------------------------------------*
> | "Faced with the choice between changing one's mind, |
> | and proving that there is no need to do so - almost |
> | everyone gets busy on the proof."                   |
> |              - John Kenneth Galbraith               |
> *-----------------------------------------------------*

View raw message