tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Larry Isaacs <Larry.Isa...@sas.com>
Subject RE: [Resend] Ready for 3.2b7?
Date Sat, 11 Nov 2000 03:09:24 GMT
Craig,

My last commit gets the default error output back to what we had
before the exception handling patch. Also, thanks for the patch to
PageContextImpl.  I can now look forward to "safely" upgrading to
Tomcat 3.2 Final.

I don't have any other changes in the queue, so I'm fine with
building 3.2b7.  Hope it wasn't already started. :-)

Larry

-----Original Message-----
From: Craig R. McClanahan
To: tomcat-dev@jakarta.apache.org
Sent: 11/10/00 6:22 PM
Subject: Re: [Resend] Ready for 3.2b7?

Larry Isaacs wrote:

> Currently PageContext.handlePageException() uses a
> RequestDisplacher.forward() to invoke the errorPage specified
> for a JSP.  If the response has been committed, you get useless
> output because of the IllegalStateException thrown by the
> forward() call.
>
> Would it be a spec violation to test if the response is committed
> and use include() instead if it is?  I think I have heard that this
> is how others avoid this problem.
>
> All I could find in the spec is that handlePageException()
> should "redirect" the to the error page.  Does this mandate
> that forward() should be used?  If not, I think it would be a
> big plus to make this change in 3.2b7.
>

Well, reducing the number of people who call your extension *definitely*
qualifies as a mission critical fix :-).  That way, you'll have more
time to
work on Tomcat :-) :-).

Seriously, it's a good idea, especially in light of the workaround we
recently committed that causes the response to *always* be committed
before
an include is processed.  I just committed this patch.

>
> Larry
>

Craig



---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org

Mime
View raw message