Craig, Thanks for your help.
When you say "* You are using a web connector that does its own buffering.", do you mean the Web connector within Tomcat? Isn't tomcat 4.1 use coyote as its web connector? Thanks again. -----Original Message----- From: Craig R. McClanahan [mailto:[EMAIL PROTECTED]] Sent: Tuesday, September 10, 2002 10:48 PM To: Tomcat Users List Subject: RE: HttpResponseBase.sendRedirect vs JetSpeed Possible reasons: * You are opening an HTML element like <table> or <p> and not closing it before the flush, and writing to a browser that does not incrementally render (like Netscape 4.x). * You are using a web connector that does its own buffering. * There is a proxy or something in between. * Bug in 4.1.10 (submit a bug report). Counting on incremental output being visible is a very chancy bet. Craig On Tue, 10 Sep 2002, Samuel Cheung wrote: > Date: Tue, 10 Sep 2002 22:38:15 -0500 > From: Samuel Cheung <[EMAIL PROTECTED]> > Reply-To: Tomcat Users List <[EMAIL PROTECTED]> > To: 'Tomcat Users List' <[EMAIL PROTECTED]> > Subject: RE: HttpResponseBase.sendRedirect vs JetSpeed > > Craig, > Thanks. But when i try to call the flush() method incrementally during the > execution of the doGet(), the client does not display anything. The browser > keeps loading and displayed everything at the end. I am using Tomcat 4.1.10 > for this example. > > Thanks for your help. > Sam > > public void doGet() { > int count = 0; > response.setBufferSize(1); > PrintWriter out = response.getWriter(); > > for (int i= 0; i < 10; i++) { > out.println("<h1> Count: "); > out.println(count++); > out.println("</h1>"); > > System.out.println ("print log count:" + count); > out.flush(); > response.flushBuffer(); > > Thread.sleep(5000); > } > > -----Original Message----- > From: Craig R. McClanahan [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, September 10, 2002 10:28 PM > To: Tomcat Users List > Subject: RE: HttpResponseBase.sendRedirect vs JetSpeed > > > Yes, you can call writer.flush() or response.flushBuffer(). > > See the servlet spec for all the details: > > http://java.sun.com/products/servlet/download.html > > Craig > > > On Tue, 10 Sep 2002, Samuel Cheung wrote: > > > Date: Tue, 10 Sep 2002 18:04:00 -0500 > > From: Samuel Cheung <[EMAIL PROTECTED]> > > Reply-To: Tomcat Users List <[EMAIL PROTECTED]> > > To: 'Tomcat Users List' <[EMAIL PROTECTED]> > > Subject: RE: HttpResponseBase.sendRedirect vs JetSpeed > > > > > > Does the servlet spec allows servlets to flush() the outputStream of the > > HTTP response class? > > Or it will be ignored by the container as well? > > > > Thanks. > > > > > > -----Original Message----- > > From: Craig R. McClanahan [mailto:[EMAIL PROTECTED]] > > Sent: Tuesday, September 10, 2002 5:53 PM > > To: Tomcat Users List > > Subject: Re: HttpResponseBase.sendRedirect vs JetSpeed > > > > > > Per the servlet spec, it is not legal to modify HTTP headers from inside > > an include (either RD.include() or <jsp:include)), and any such attempt > > will be ignored by the container. > > > > You can certainly build Tomcat from source and make the modification > > you're talking about in your copy, but you'll be making your copy > > non-spec-compliant if you do. > > > > Craig > > > > > > On Tue, 10 Sep 2002, Erich Bratton wrote: > > > > > Date: Tue, 10 Sep 2002 17:22:59 -0500 > > > From: Erich Bratton <[EMAIL PROTECTED]> > > > Reply-To: Tomcat Users List <[EMAIL PROTECTED]> > > > To: [EMAIL PROTECTED] > > > Subject: HttpResponseBase.sendRedirect vs JetSpeed > > > > > > > > > Hi, > > > > > > We're working on getting JetSpeed up and running with some portlets > using > > > Velocity and some using JSP. We are seeing behavior where, even using > > Tomcat > > > 4.1 with a bufferSize set to 100k, when a JSP page calls > > > response.sendRedirect(), nothing happens. > > > > > > The problem appears to be in HttpResponseBase, lines 1132-1133, where it > > checks > > > to see if it is being called from a RequestDispatcher.include() and if > so, > > it > > > just returns and silently does nothing. This behavior seems odd, since > > lines > > > 1128-1130 check to see if the response has been committed already, and > if > > so, it > > > throws an Exception. > > > > > > Having said all this, is this a bug? Is there a way to configure this > > behavior > > > from a config file somewhere? Is there a way for us to compile a > version > > of > > > HttpResponseBase with lines 1132-1133 commented out and for us to drop > it > > > somewhere in the Tomcat hierarchy where it will 'override' the default > > Tomcat > > > version of HttpResponseBase? > > > > > > Thanks for any and all ideas, > > > > > > --Erich > > > > > > -- > > > To unsubscribe, e-mail: > > <mailto:[EMAIL PROTECTED]> > > > For additional commands, e-mail: > > <mailto:[EMAIL PROTECTED]> > > > > > > > > > > > > -- > > To unsubscribe, e-mail: > > <mailto:[EMAIL PROTECTED]> > > For additional commands, e-mail: > > <mailto:[EMAIL PROTECTED]> > > > > -- > > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > > > > > > > -- > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > > -- > To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> > > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
