[From nobody Fri Aug  6 17:07:29 2004
>From - Fri Mar 15 14: 50:25 2002
X-Mozilla-Status2: 00000000
Return-Path: <david-b@pacbell.net>
Received: from mta5.snfc21.pbi.net (mta5.snfc21.pbi.net [206.13.28.241])
	by thor.host4u.net (8.11.6/8.11.6) with ESMTP id g2FK8ir10461
	for <dms@sosnoski.com>; Fri, 15 Mar 2002 14:08:44 -0600
Received: from krypton ([206.170.7.91])
	by mta5.snfc21.pbi.net (iPlanet Messaging Server 5.1 (built May 7
	2001))
	with SMTP id <0GT100CRB6GJFF@mta5.snfc21.pbi.net> for dms@sosnoski.com;
	Fri, 15 Mar 2002 12:05:09 -0800 (PST)
Date: Fri, 15 Mar 2002 12:03:31 -0800
From: David Brownell <david-b@pacbell.net>
Subject: Re: SAX/SAX2 parser reusability
To: Dennis Sosnoski <dms@sosnoski.com>
Cc: David Megginson <david@megginson.com>
Message-id: <020c01c1cc5c$7b8d4d40$6800000a@brownell.org>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 7bit
X-Priority: 3
X-MSMail-priority: Normal
References: <E16lorW-0004pT-00@usw-sf-web1.sourceforge.net>
	<15505.59043.661672.183245@megginson.com>
	<3C924B61.1060803@sosnoski.com>

Dennis --

SAX parsers that can't be re-used after they (or some
handler they call) throw an exception are nonconformant.
Throwing an exception should only abort the current parse.
As it says in the javadoc for XMLReader.parse():

    Once a parse is complete, an application may reuse
    the same XMLReader object, possibly with a different
    input source. ... If a client application wants to terminate
    parsing early, it should throw an exception.

Throwing an exception just causes "early" completion.

I'm not sure what javadoc seems to suggest otherwise;
could you point it out to me?

FWIW, that's the policy I'd expect from instances of any
class whatsoever, unless its interface spec said quite
explicitly that it self-corrupted when throwing exceptions.

- Dave



----- Original Message ----- 
From: "Dennis Sosnoski" <dms@sosnoski.com>
To: "David Brownell" <dbrownell@users.sourceforge.net>
Cc: "David Megginson" <david@megginson.com>
Sent: Friday, March 15, 2002 11:28 AM
Subject: Re: SAX/SAX2 parser reusability


> Hi DavidB,
> 
> This is in regard to a current discussion on the JDOM email list, so I'd 
> like permission to forward your response there. The issue is whether 
> parser reusability is something that can be depended on.
> 
> Thanks,
> 
>   - Dennis
> 
> David Megginson wrote:
> 
> >Dennis Sosnoski writes:
> >
> > > I can see where the documentation might be subject to
> > > different interpretations - the error handling code says the
> > > parser can basically die after throwing a fatal exception.
> > > My own interpretation is that the parser can die *for that
> > > document only*, and must still be usable with other
> > > documents on later calls to parse.
> >
> >That was my intention as well -- that the parser should be reusable
> >for a new parse.  I've copied the current SAX maintainer, David
> >Brownell, so that he can decide whether any documentation
> >clarifications are in order.
> >
> >
> >All the best,
> >
> >
> >David
> >
> 
> 

]