parser vote

[James M Snell <ja...@us.ibm.com>]

All,

Until we figure out which parser we want for  Axis, I propose that we drop 
JDOM and go with the JAXP interfaces.  This provides us the greatest 
amount of flexibility and the most breathing room.  It doesn't meet our 
requirement for not keeping the entire message object model in memory, but 
I'm wondering if we should reevaluate the importance of that requirement 
for version 1.0.  Until we can hammer it out, we need to get moving on 
this.  Let's take JDOM out, put JAXP in, use whatever JAXP aware parser 
you want to use and go from there. 

In the meantime, some of us should get together with the Xerces guys and 
work with them on performance and interface issues that would make Xerces 
a better tool for Axis to use.  I volunteer for this, anybody else?  We 
should have at least two or three so we can get a good representative 
perspective.

- James Snell
     Software Engineer, Emerging Technologies, IBM
     jasnell@us.ibm.com (online)
     jsnell@lemoorenet.com (offline)

Re: parser vote

[Glen Daniels <gd...@allaire.com>]

James:

I guess I need to look deeper into JAXP.  I thought that JAXP just provided you
with a parser-independent way of getting at DOM or SAX.  If that were true, the
"JDOM vs. JAXP" dichotomy here wouldn't really exist; as I understand it the
JDOM guys are already working towards JAXP compatibility at the bottom layer,
so why punt the object model?  "JDOM vs. DOM" seems like a real question, but
"JDOM vs. JAXP" seems orthogonal to me.

I guess I'm going to abstain on this vote for now, since I don't have the full
picture, but I really like the easy and natural programming model of JDOM.  I'm
at least going to want to use JDOM when dealing directly with XML in handlers I
might write, so I'd like to make sure we don't do anything to preclude it.

More after I get back....

--Glen

----- Original Message -----
From: "James M Snell" <ja...@us.ibm.com>
To: <ax...@xml.apache.org>
Sent: Tuesday, March 27, 2001 3:54 PM
Subject: parser vote


> All,
>
> Until we figure out which parser we want for  Axis, I propose that we drop
> JDOM and go with the JAXP interfaces.  This provides us the greatest
> amount of flexibility and the most breathing room.  It doesn't meet our
> requirement for not keeping the entire message object model in memory, but
> I'm wondering if we should reevaluate the importance of that requirement
> for version 1.0.  Until we can hammer it out, we need to get moving on
> this.  Let's take JDOM out, put JAXP in, use whatever JAXP aware parser
> you want to use and go from there.
>
> In the meantime, some of us should get together with the Xerces guys and
> work with them on performance and interface issues that would make Xerces
> a better tool for Axis to use.  I volunteer for this, anybody else?  We
> should have at least two or three so we can get a good representative
> perspective.
>
> - James Snell
>      Software Engineer, Emerging Technologies, IBM
>      jasnell@us.ibm.com (online)
>      jsnell@lemoorenet.com (offline)
>