Return-Path: 
 <axis-dev-return-4474-apmail-xml-axis-dev-archive=xml.apache.org@xml.apache.org>
Delivered-To: apmail-xml-axis-dev-archive@xml.apache.org
Received: (qmail 19029 invoked by uid 500); 28 Nov 2001 12:14:41 -0000
Mailing-List: contact axis-dev-help@xml.apache.org; run by ezmlm
Precedence: bulk
Reply-To: axis-dev@xml.apache.org
list-help: <mailto:axis-dev-help@xml.apache.org>
list-unsubscribe: <mailto:axis-dev-unsubscribe@xml.apache.org>
list-post: <mailto:axis-dev@xml.apache.org>
Delivered-To: mailing list axis-dev@xml.apache.org
Received: (qmail 19019 invoked from network); 28 Nov 2001 12:14:41 -0000
Message-ID: <012c01c17806$1815bfe0$6b00a8c0@ne.mediaone.net>
From: "Glen Daniels" <gdaniels@macromedia.com>
To: <dims@yahoo.com>,
	<axis-dev@xml.apache.org>
References: <20011128120253.22329.qmail@web12803.mail.yahoo.com>
Subject: Re: Serializing faults, unmatched <init>s, and offline testing...
Date: Wed, 28 Nov 2001 07:13:28 -0500
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N

Also FYI - typically you're caring about the whole <init> thing in the
context of serializing a Fault out to XML using a SerializationContext.  The
SerializationContext.writeSafeString() method is a convenience method
already in place which calls xmlEncodeString() for you.

+1 for better unit test coverage!

--Glen

P.S.  XMLP meeting goes well so far, I hope to have a little time tonight
for some more Axis work.

----- Original Message -----
From: "Davanum Srinivas" <dims@yahoo.com>
To: <axis-dev@xml.apache.org>
Sent: Wednesday, November 28, 2001 7:02 AM
Subject: Re: Serializing faults, unmatched <init>s, and offline testing...


> Sam,
>
> FYI. I had a problem with <init>. To make a long story short it was due to
a problem in
> NonBlockingBufferedInputStream.java. Checked in a fix yesterday.
>
> Thanks,
> dims
>
> --- Sam Ruby <rubys@us.ibm.com> wrote:
> > Serializing faults used to work.  Given the amount of change in this
area
> > recently, I gather that they stopped working at some point.  This points
to
> > a failing in our test coverage, in which we can all share the blame
(myself
> > included).
> >
> > I saw in IRC yesterday that somebody was having a problem with invalid
XML
> > being produced - unmatched <init> tags.  This typically is caused by
stack
> > tracebacks being present in a fault with at least one method in the list
> > being a constructor.  The fix for this is to call
XMLUtils.xmlEncodeString
> > ().  Again, this used to be done properly.
> >
> > The design of the tests which support getQuote("IBM") rely on faults
being
> > correctly serialized in order to support offline testing.  Of course
having
> > a functional test fail only says that *something* is wrong, leaving a
> > significant debug task.  For this reason, having focused unit tests are
> > preferred.
> >
> > - Sam Ruby
> >
>
>
> =====
> Davanum Srinivas - http://jguru.com/dims/
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
> http://geocities.yahoo.com/ps/info1
>