Re: jk connector error on post.
I believe I have made some progress now. It looks like this is a problem with mod_perl and mod_jk interacting poorly. I haven't investigated why yet, but mod_perl is calling ap_should_client_block in the authentication phase of Apache, which plays havoc with mod_jk's call in the handler. Mod_jk is in the right here, and I'm 99.9% sure that mod_perl isn't making the call, but rather our code is. So, this is my problem. Thanks to everybody who helped. Ryan On Tue, 18 Jan 2005, William A. Rowe, Jr. wrote: Hi Ryan, fancy meeting up here :) You don't mention, so I'll confirm - you are using mod_jk 1.2.8 (brand new release)? This sure sounds like the (rather obscure) issues with re-piping body content, and taking a simple guess, if you drop back and force http/1.0 (eliminate chunking) the issue goes away? Bill At 06:55 PM 1/17/2005, Ryan Bloom wrote: I am trying to get Apache 1.3 and Tomcat 5.0.27 working together. My site basically works, but any POST request causes the following error with a hex dump: Jan 17, 2005 3:19:23 PM org.apache.jk.common.HandlerRequest invoke SEVERE: Error decoding request java.io.IOException at org.apache.jk.common.JkInputStream.receive(JkInputStream.java:252) at org.apache.jk.common.HandlerRequest.decodeRequest(HandlerRequest.java:500) at org.apache.jk.common.HandlerRequest.invoke(HandlerRequest.java:352) at org.apache.jk.common.ChannelSocket.invoke(ChannelSocket.java:743) at org.apache.jk.common.ChannelSocket.processConnection(ChannelSocket.java:675) at org.apache.jk.common.SocketConnection.runIt(ChannelSocket.java:866) at org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:683) at java.lang.Thread.run(Thread.java:567) I have been unable to get Tomcat to give me more log messages. The web pages are very clear that jk2 is no longer in development, so I really don't want to go to jk2. but I am at a loss as to where to go from here. Can anybody suggest a possible solution? Thanks, Ryan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: jk connector error on post.
Actually, both 1.2.6 and 1.2.8 of mod_jk. What do you mean re-piping body content? I haven't forced http/1.0 yet, I was about to trace through Apache to find the problem. My initial investigation looks like the problem is in Apache somewhere, looks like ap_setup_client_block/ap_should_client_block. Ryan On Tue, 18 Jan 2005, William A. Rowe, Jr. wrote: Hi Ryan, fancy meeting up here :) You don't mention, so I'll confirm - you are using mod_jk 1.2.8 (brand new release)? This sure sounds like the (rather obscure) issues with re-piping body content, and taking a simple guess, if you drop back and force http/1.0 (eliminate chunking) the issue goes away? Bill At 06:55 PM 1/17/2005, Ryan Bloom wrote: I am trying to get Apache 1.3 and Tomcat 5.0.27 working together. My site basically works, but any POST request causes the following error with a hex dump: Jan 17, 2005 3:19:23 PM org.apache.jk.common.HandlerRequest invoke SEVERE: Error decoding request java.io.IOException at org.apache.jk.common.JkInputStream.receive(JkInputStream.java:252) at org.apache.jk.common.HandlerRequest.decodeRequest(HandlerRequest.java:500) at org.apache.jk.common.HandlerRequest.invoke(HandlerRequest.java:352) at org.apache.jk.common.ChannelSocket.invoke(ChannelSocket.java:743) at org.apache.jk.common.ChannelSocket.processConnection(ChannelSocket.java:675) at org.apache.jk.common.SocketConnection.runIt(ChannelSocket.java:866) at org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:683) at java.lang.Thread.run(Thread.java:567) I have been unable to get Tomcat to give me more log messages. The web pages are very clear that jk2 is no longer in development, so I really don't want to go to jk2. but I am at a loss as to where to go from here. Can anybody suggest a possible solution? Thanks, Ryan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
jk connector error on post.
I am trying to get Apache 1.3 and Tomcat 5.0.27 working together. My site basically works, but any POST request causes the following error with a hex dump: Jan 17, 2005 3:19:23 PM org.apache.jk.common.HandlerRequest invoke SEVERE: Error decoding request java.io.IOException at org.apache.jk.common.JkInputStream.receive(JkInputStream.java:252) at org.apache.jk.common.HandlerRequest.decodeRequest(HandlerRequest.java:500) at org.apache.jk.common.HandlerRequest.invoke(HandlerRequest.java:352) at org.apache.jk.common.ChannelSocket.invoke(ChannelSocket.java:743) at org.apache.jk.common.ChannelSocket.processConnection(ChannelSocket.java:675) at org.apache.jk.common.SocketConnection.runIt(ChannelSocket.java:866) at org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:683) at java.lang.Thread.run(Thread.java:567) I have been unable to get Tomcat to give me more log messages. The web pages are very clear that jk2 is no longer in development, so I really don't want to go to jk2. but I am at a loss as to where to go from here. Can anybody suggest a possible solution? Thanks, Ryan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tag files not pooling
The JSP 2.0 specification states that these must not be cached. See section 7.1.5. On Wed, 2003-11-12 at 10:48, jakarta wrote: Is tomcat 5 going to suport tag file pooling as well as tag pooling. When looking at some generated code I notice that it was currently only pooling tags not tag files? Thanks, john - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Ryan Lubke [EMAIL PROTECTED] Sun Microsystems, Inc. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: cvs commit:jakarta-servletapi-5/jsr154/src/share/javax/servlet/http HttpUtils.java
Yes it should. That's twice I've done that now. On Fri, 2002-12-13 at 12:56, Jason Hunter wrote: Shouldn't that be amp; instead? -jh- [EMAIL PROTECTED] wrote: jfarcand2002/12/13 08:47:29 Modified:jsr154/src/share/javax/servlet/http HttpUtils.java Log: Doclets bombs when processing this file (minor fix) Submitted by: Ryan Lubke Revision ChangesPath 1.3 +1 -1 jakarta-servletapi-5/jsr154/src/share/javax/servlet/http/HttpUtils.java Index: HttpUtils.java === RCS file: /home/cvs/jakarta-servletapi-5/jsr154/src/share/javax/servlet/http/HttpUtils.java,v retrieving revision 1.2 retrieving revision 1.3 diff -u -r1.2 -r1.3 --- HttpUtils.java15 Oct 2002 21:39:44 - 1.2 +++ HttpUtils.java13 Dec 2002 16:47:28 - 1.3 @@ -104,7 +104,7 @@ * The query string should be in the form of a string * packaged by the GET or POST method, that is, it * should have key-value pairs in the form ikey=value/i, - * with each pair separated from the next by a character. + * with each pair separated from the next by a apos character. * * pA key can appear more than once in the query string * with different values. However, the key appears only once in -- 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] -- Ryan Lubke [EMAIL PROTECTED] Sun Microsystems, Inc. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: cvs commit:jakarta-servletapi-5/jsr154/src/share/javax/servlet/http HttpUtils.java
You still missed the terminating ';' :) On Fri, 2002-12-13 at 13:01, [EMAIL PROTECTED] wrote: jfarcand2002/12/13 10:01:54 Modified:jsr154/src/share/javax/servlet/http HttpUtils.java Log: amp instead (thanks Jason). Revision ChangesPath 1.4 +1 -1 jakarta-servletapi-5/jsr154/src/share/javax/servlet/http/HttpUtils.java Index: HttpUtils.java === RCS file: /home/cvs/jakarta-servletapi-5/jsr154/src/share/javax/servlet/http/HttpUtils.java,v retrieving revision 1.3 retrieving revision 1.4 diff -u -r1.3 -r1.4 --- HttpUtils.java 13 Dec 2002 16:47:28 - 1.3 +++ HttpUtils.java 13 Dec 2002 18:01:54 - 1.4 @@ -104,7 +104,7 @@ * The query string should be in the form of a string * packaged by the GET or POST method, that is, it * should have key-value pairs in the form ikey=value/i, - * with each pair separated from the next by a apos character. + * with each pair separated from the next by a amp character. * * pA key can appear more than once in the query string * with different values. However, the key appears only once in -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED] -- Ryan Lubke [EMAIL PROTECTED] Sun Microsystems, Inc. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH][JASPER2] Simple patch for one of the resource errormessages.
I think that the following error message is incorrect:
- Scripting elements ( %@, %!, %=, % ) are disallowed here.
Since directives (%@) are not scripting elements.
I've attached a patch to correct the messages.properties.
Index: messages.properties
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/resources/messages.properties,v
retrieving revision 1.60
diff -u -r1.60 messages.properties
--- messages.properties 16 Nov 2002 04:20:10 - 1.60
+++ messages.properties 20 Nov 2002 23:47:23 -
@@ -266,7 +266,7 @@
jsp.error.empty.body.not.allowed=Empty body not allowed for {0}
jsp.error.jspbody.required=Must use jsp:body to specify tag body for {0} if jsp:attribute is used.
jsp.error.jspbody.emptybody.only=The {0} tag can only have jsp:attribute in its body.
-jsp.error.no.scriptlets=Scripting elements ( %@, %!, %=, % ) are disallowed here.
+jsp.error.no.scriptlets=Scripting elements ( %!, %=, % ) are disallowed here.
jsp.error.internal.unexpected_node_type=Internal Error: Unexpected node type encountered
jsp.error.tld.fn.invalid.signature=Invalid syntax for function signature in TLD. Tag Library: {0}, Function: {1}
jsp.error.tld.fn.duplicate.name=Duplicate function name {0} in tag library {1}
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: Bug(?) and suggested solution for chunked transfer-encoding POSTrequest
Remy Maucherat wrote: Adrian Sampaleanu wrote: We are having a problem with Catalina 4.1.12 (I believe we also had issues with previous versions, but can't say for sure) with respect to how POSTed requests are handled if the transfer encoding is chunked. It seems that request parameters in the content are not processed in this case. This is not legal. In HTTP/1.1, FORM encoded data of a POST must have a content-length. You can do the parsing yourself, of course. Rémy See RFC2616, Section 4.4. An excerpt: Messages MUST NOT include both a Content-Length header field and a non-identity transfer-coding. If the message does include a non- identity transfer-coding, the Content-Length MUST be ignored. In looking through RFC2616, I didn't see any special case for POST requests with FORM encoding that either required the use of Content-Length (thereby prohibiting the use of Transfer-Encoding: chunked) or vice versa. If I missed it or if its in a different RFC please post a reference. -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: cvs commit:jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtimePageContextImpl.java
of these.
}
-
+
return this.expressionEvaluator;
}
public void handlePageException(Exception ex)
-throws IOException, ServletException
+throws IOException, ServletException
{
// Should never be called since handleException() called with a
// Throwable in the generated servlet.
@@ -553,7 +554,7 @@
}
public void handlePageException(Throwable t)
-throws IOException, ServletException
+throws IOException, ServletException
{
if (t == null) throw new NullPointerException(null Throwable);
@@ -563,9 +564,11 @@
// Do not set the javax.servlet.error.exception attribute here
// (instead, set in the generated servlet code for the error page)
// in order to prevent the ErrorReportValve, which is invoked as
- // part of forwarding the request to the error page, from
+ // part of forwarding the request to the error page, from
// throwing it if the response has not been committed (the response
// will have been committed if the error page is a JSP page).
+Object
origException=request.getAttribute(javax.servlet.error.exception);
+
request.setAttribute(javax.servlet.jsp.jspException, t);
request.setAttribute(javax.servlet.error.status_code,
new Integer(HttpServletResponse.SC_INTERNAL_SERVER_ERROR));
@@ -578,6 +581,19 @@
} catch (IllegalStateException ise) {
include(errorPageURL);
}
+// The error page could be inside an include.
+
+Object
newException=request.getAttribute(javax.servlet.error.exception);
+
+if( (newException!= null) (newException==origException) ) {
+request.removeAttribute(javax.servlet.error.exception);
+}
+
+// now clear the error code - to prevent double handling.
+request.removeAttribute(javax.servlet.error.status_code);
+request.removeAttribute(javax.servlet.error.request_uri);
+request.removeAttribute(javax.servlet.error.status_code);
+request.removeAttribute(javax.servlet.jsp.jspException);
} else {
// Otherwise throw the exception wrapped inside a ServletException.
@@ -601,19 +617,19 @@
/**
* VariableResolver interface
*/
-public Object resolveVariable( String pName, Object pContext )
+public Object resolveVariable( String pName, Object pContext )
throws ELException
{
// Note: pContext will be going away.
try {
-return PageContextImpl.variableResolver.resolveVariable(
+return PageContextImpl.variableResolver.resolveVariable(
pName, this );
}
catch( org.apache.jasper.runtime.el.jstl.ELException e ) {
throw new ELException( e );
}
}
-
+
/**
* Proprietary method to evaluate EL expressions.
* XXX - This method should go away once the EL interpreter moves
@@ -627,16 +643,16 @@
* @param defaultPrefix Default prefix for this evaluation
* @return The result of the evaluation
*/
-public static Object proprietaryEvaluate( String expression,
+public static Object proprietaryEvaluate( String expression,
Class expectedType, PageContext pageContext,
ProtectedFunctionMapper functionMap, String defaultPrefix )
throws ELException
{
java.util.HashMap funcMap =
(functionMap == null)? null: functionMap.getFnMap();
-
+
try {
-return PageContextImpl.proprietaryEvaluator.evaluate( unknown,
+return PageContextImpl.proprietaryEvaluator.evaluate( unknown,
expression, expectedType, null, pageContext,
funcMap, defaultPrefix );
}
@@ -654,7 +670,7 @@
return null;
}
}
-
+
/*
* fields
*/
--
To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org
For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
--
Ryan Lubke [EMAIL PROTECTED]
Sun Microsystems, Inc.
--
To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org
For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: cvs commit:jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtimePageContextImpl.java
+// no need to synchronize - not a big deal even if we create
// two of these.
}
-
+
return this.expressionEvaluator;
}
public void handlePageException(Exception ex)
-throws IOException, ServletException
+throws IOException, ServletException
{
// Should never be called since handleException() called with a
// Throwable in the generated servlet.
@@ -553,7 +554,7 @@
}
public void handlePageException(Throwable t)
-throws IOException, ServletException
+throws IOException, ServletException
{
if (t == null) throw new NullPointerException(null Throwable);
@@ -563,9 +564,11 @@
// Do not set the javax.servlet.error.exception attribute here
// (instead, set in the generated servlet code for the error page)
// in order to prevent the ErrorReportValve, which is invoked as
- // part of forwarding the request to the error page, from
+ // part of forwarding the request to the error page, from
// throwing it if the response has not been committed (the response
// will have been committed if the error page is a JSP page).
+Object
origException=request.getAttribute(javax.servlet.error.exception);
+
request.setAttribute(javax.servlet.jsp.jspException, t);
request.setAttribute(javax.servlet.error.status_code,
new Integer(HttpServletResponse.SC_INTERNAL_SERVER_ERROR));
@@ -578,6 +581,19 @@
} catch (IllegalStateException ise) {
include(errorPageURL);
}
+// The error page could be inside an include.
+
+Object
newException=request.getAttribute(javax.servlet.error.exception);
+
+if( (newException!= null) (newException==origException) ) {
+request.removeAttribute(javax.servlet.error.exception);
+}
+
+// now clear the error code - to prevent double handling.
+request.removeAttribute(javax.servlet.error.status_code);
+request.removeAttribute(javax.servlet.error.request_uri);
+request.removeAttribute(javax.servlet.error.status_code);
+request.removeAttribute(javax.servlet.jsp.jspException);
} else {
// Otherwise throw the exception wrapped inside a ServletException.
@@ -601,19 +617,19 @@
/**
* VariableResolver interface
*/
-public Object resolveVariable( String pName, Object pContext )
+public Object resolveVariable( String pName, Object pContext )
throws ELException
{
// Note: pContext will be going away.
try {
-return PageContextImpl.variableResolver.resolveVariable(
+return PageContextImpl.variableResolver.resolveVariable(
pName, this );
}
catch( org.apache.jasper.runtime.el.jstl.ELException e ) {
throw new ELException( e );
}
}
-
+
/**
* Proprietary method to evaluate EL expressions.
* XXX - This method should go away once the EL interpreter moves
@@ -627,16 +643,16 @@
* @param defaultPrefix Default prefix for this evaluation
* @return The result of the evaluation
*/
-public static Object proprietaryEvaluate( String expression,
+public static Object proprietaryEvaluate( String expression,
Class expectedType, PageContext pageContext,
ProtectedFunctionMapper functionMap, String defaultPrefix )
throws ELException
{
java.util.HashMap funcMap =
(functionMap == null)? null: functionMap.getFnMap();
-
+
try {
-return PageContextImpl.proprietaryEvaluator.evaluate( unknown,
+return PageContextImpl.proprietaryEvaluator.evaluate( unknown,
expression, expectedType, null, pageContext,
funcMap, defaultPrefix );
}
@@ -654,7 +670,7 @@
return null;
}
}
-
+
/*
* fields
*/
--
To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org
For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
--
Ryan Lubke [EMAIL PROTECTED]
Sun Microsystems, Inc.
--
To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org
For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: Possible bug in RequestDispatcher?
I looked into this a little further. I think the resource the URL points to has a Content-Length of 234 bytes. The Filter uses RequestDispatcher.forward() to return a document that is longer, and that document is getting truncated at about 234 bytes. Could TC be setting the Content-Length header based on the originally requested document but returning the data from the forwarded one? -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net Ryan Hoegg wrote: OK, that clarifies things a little bit. If the problem isn't with RequestDispatcher, I am even more puzzled. I have narrowed the problem down to a particular use case: I have a single filter that calls getFilterContext().getServletContext().getRequestDispatcher(/WEB-INF/templates/form.html).forward(request, response) in the absence of a particular session variable on the request. This filter is mapped to /* and I have a single servlet mapped to /Display Pointing my browser at /webapp/Display : things work as expected. Pointing my browser at /webapp/ : things work as expected Pointing my browser at /webapp/staticfile.html : problem. View Source in my browser shows this: - !DOCTYPE html PUBLIC -//W3C//DTD XHTML 1.0 Transitional//EN http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd; html head title/title /head body form action=j_security_check method=POST pU -- That is the beginning of the html file I am trying to forward() to. It seems like my HttpServletResponse's OutputStream is getting closed prematurely in the exact same place in my forwarded file. So what could it be?! -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net Craig R. McClanahan wrote: In Servlet 2.3 (i.e. Tomcat 4.x), filters are *not* invoked on request dispatcher calls -- only on the original request. In Servlet 2.4 (i.e. Tomcat 5.x) you can optionally configure which filters should be used on forwards, includes, or both, with suitable settings in web.xml. Craig -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Possible bug in RequestDispatcher?
Hi, Before I post a bug report, I wanted to run a peculiar situation by the experts. I am creating some filters that verify that certain variables are in the session before allowing access to any page in my site. There are currently two of them, and each forward the request to a form if a particular session variable is not set and if this is not the form submission. They are using ((HttpServletRequest) request).getRequestDispatcher(/path/to/form1.html).forward(request, response) to accomplish this. Something is broken, and if it's in my code I am missing it. I have these filters mapped to /* and here is the peculiar behavior: 1. If I point my browser at /, I get the desired HTML page. 2. If I point at a servlet I have mapped, I get what looks like an HTML template page. 3. If I point at a static HTML file I have in the / I get a good chunk of the beginning of the HTML page and that's it. My web.xml is simpler than simple, I described it above except for the single servlet mapping. (/Display) I appreciate any insights you can provide! -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: Possible bug in RequestDispatcher?
OK, that clarifies things a little bit. If the problem isn't with RequestDispatcher, I am even more puzzled. I have narrowed the problem down to a particular use case: I have a single filter that calls getFilterContext().getServletContext().getRequestDispatcher(/WEB-INF/templates/form.html).forward(request, response) in the absence of a particular session variable on the request. This filter is mapped to /* and I have a single servlet mapped to /Display Pointing my browser at /webapp/Display : things work as expected. Pointing my browser at /webapp/ : things work as expected Pointing my browser at /webapp/staticfile.html : problem. View Source in my browser shows this: - !DOCTYPE html PUBLIC -//W3C//DTD XHTML 1.0 Transitional//EN http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd; html head title/title /head body form action=j_security_check method=POST pU -- That is the beginning of the html file I am trying to forward() to. It seems like my HttpServletResponse's OutputStream is getting closed prematurely in the exact same place in my forwarded file. So what could it be?! -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net Craig R. McClanahan wrote: In Servlet 2.3 (i.e. Tomcat 4.x), filters are *not* invoked on request dispatcher calls -- only on the original request. In Servlet 2.4 (i.e. Tomcat 5.x) you can optionally configure which filters should be used on forwards, includes, or both, with suitable settings in web.xml. Craig -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Tomcat scalability
1. For a project my company is working on we have transactions requirements of 1600 transactions per second. The transactions consists of processing a servlet in Tomcat, doing a database call and then displaying the results to the user so the effective number of transactions Tomcat has to process is actually greater than 1600. Can Tomcat cope with 1600 requests in a second(a 4 processor Sunfire machine will be used)? Based upon our current application architecture(our java application has tomcat running inside it), 1600 request per second means we may have 1600 threads open simultaneously. 2. From one article I read at linux journal, Tomcat 3 didn't scale very well with multiple processors because of JVM issues. -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: Tomcat scalability
One good rule of thumb is not to solve problems that don't exist. Your first task is to set up a server and hit it with something a good 20%-50% more demanding than your expected load. There exist several automated tools to do this. One is JMeter at http://jakarta.apache.org/jmeter/index.html . Then, have a look at your actual performance and work on the bottlenecks that arise. If your application is spending most of its time waiting for database results, cache them. If your application is spending most of its time creating and destroying objects, consider pooling. If your app is choking on serving up 1600 images a second, use a web server such as Apache in front of Tomcat. That will help with static requests (images, static HTML) even with a single Tomcat server doing the servlet work. -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net [EMAIL PROTECTED] wrote: That's a tough question to answer, and pretty much the only way you're going to be able to tell is to try it. Here's a few suggestions of things to think about: The first question you might want to ask is can your database handle 1600 transactions per second? If not (and even if it can) you may want to consider whether some kind of caching could help you if the data is largely static. This could be caching of the data itself, or something as simple as caching the pages themselves, with some scheme to flush the cache if the data changes. If your data is not very static, then this wouldn't be as helpful of course. If it's possible on the OS you are using, I'd be tempted to run 4 copies of Tomcat, and dedicate a processor to each one. It's a lot harder to write code that will run reliably on a MP machine than on a SP one, and it seems that Tomcat has a few issues in that regard. Also, remember, it's unlikely you'll ever have that many threads open at once - if you are actually handling 1600 requests/second then each request is, on average, taking less than a millisecond to complete, so you won't have too many overlapping requests. If you can work out how long it takes to process one request, then you'll have a best-case scenario of how many you can handle. In practice, it will be less of course, due to overhead in handling multiple requests at once. -- Rob 1. For a project my company is working on we have transactions requirements of 1600 transactions per second. The transactions consists of processing a servlet in Tomcat, doing a database call and then displaying the results to the user so the effective number of transactions Tomcat has to process is actually greater than 1600. Can Tomcat cope with 1600 requests in a second(a 4 processor Sunfire machine will be used)? Based upon our current application architecture(our java application has tomcat running inside it), 1600 request per second means we may have 1600 threads open simultaneously. 2. From one article I read at linux journal, Tomcat 3 didn't scale very well with multiple processors because of JVM issues. -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Catalina Ant tasks
Hi, Am I missing something or do we not have Ant tasks for the functionality available through the /admin application? I have an itch, so if not I will be willing to contribute some code. -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
Re: Form Based Authentication, getting login and password
Hi, I am new to the list, so nice to meet you. I found this thread in the archives while researching a way to do something similar. I want to implement an authenticator that uses information in the session to augment the authentication. I have looked into creating a subclass of FormAuthenticator and it seems fairly straightforward. My plan is to override the authenticate() method and: 1. Check if the session is properly initialized (not sure how yet), and if not do something similar to the login form submission piece to collect prerequisite session info. 2. Build the username from the one submitted on the login form and from arbitrary session variables. 3. Hand off the rest of the authentication to FormAuthenticator for Realm based authentication. This strategy would also allow me to build in 3 strikes and you're out functionality as I am inspecting the session before sending the authentication request to the realm. Questions: 1. Is the entire idea wrongheaded for some reason that I am missing? 2. Any gotchas someone would like to point out for me? 3. authenticate() is an awfully long method. Would anyone object to breaking it down into several protected method calls so that I can just override those and leave the rest up to FormAuthenticator? Currently I have to copy a lot of code into my subclass. I will be happy to submit my changes upstream. I have worked a little with the Jakarta community and know the routine... you guys haven't switched to Scarab yet have you? -- Ryan Hoegg ISIS Networks http://www.isisnetworks.net -- To unsubscribe, e-mail: mailto:tomcat-dev-unsubscribe;jakarta.apache.org For additional commands, e-mail: mailto:tomcat-dev-help;jakarta.apache.org
[PATCH][jakarta-servletapi-5/jsr152] Documentation Patch
Update to clarify return values of TagVariableInfo.
- getClassName() will return java.lang.String if
the variable-class element is not specified.
- getDeclare() will return true if the declare
element is not specified.
- getScope() will return NESTED as the scope if
scope is not specified.
Index: TagVariableInfo.java
===
RCS file: /home/cvs/jakarta-servletapi-5/jsr152/src/share/javax/servlet/jsp/tagext/TagVariableInfo.java,v
retrieving revision 1.2
diff -u -r1.2 TagVariableInfo.java
--- TagVariableInfo.java 19 Aug 2002 16:29:51 - 1.2
+++ TagVariableInfo.java 10 Oct 2002 01:02:08 -
@@ -139,7 +139,8 @@
/**
* The body of the lt;variable-classgt; element.
*
- * @return The name of the class of the variable
+ * @return The name of the class of the variable or
+ * 'java.lang.String' if not defined in the TLD.
*/
public String getClassName() {
@@ -149,7 +150,8 @@
/**
* The body of the lt;declaregt; element
*
- * @return Whether the variable is to be declared or not
+ * @return Whether the variable is to be declared or not.
+ * If not defined in the TLD, 'true' will be returned.
*/
public boolean getDeclare() {
@@ -159,7 +161,9 @@
/**
* The body of the lt;scopegt; element
*
- * @return The scope to give the variable.
+ * @return The scope to give the variable. NESTED
+ * scope will be returned if not defined in
+ * the TLD.
*/
public int getScope() {
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH][jakarta-servletapi-5/jsr152] Resubmit of documentation path.
Please use the attached patch instead of the one I previously posted.
Changes
TagVariableInfo
- getClassName() will return java.lang.String if
the variable-class element is not specified.
- getDeclare() will return true if the declare
element is not specified.
- getScope() will return NESTED as the scope if
scope is not specified.
TagLibraryInfo
- Changed the directive from include to taglib in
description of getShortName().
- Minor rewording of getReliableURN for clarity.
Thanks,
-rl
Index: TagLibraryInfo.java
===
RCS file: /home/cvs/jakarta-servletapi-5/jsr152/src/share/javax/servlet/jsp/tagext/TagLibraryInfo.java,v
retrieving revision 1.3
diff -u -r1.3 TagLibraryInfo.java
--- TagLibraryInfo.java 20 Aug 2002 21:08:24 - 1.3
+++ TagLibraryInfo.java 10 Oct 2002 23:38:29 -
@@ -148,7 +148,7 @@
/**
* The preferred short name (prefix) as indicated in the TLD.
* This may be used by authoring tools as the preferred prefix
- * to use when creating an include directive for this library.
+ * to use when creating an taglib directive for this library.
*
* @return the preferred short name for the library
*/
@@ -157,10 +157,9 @@
}
/**
- * The reliable URN indicated in the TLD.
+ * The reliable URN indicated in the TLD (the uri element).
* This may be used by authoring tools as a global identifier
- * (the uri attribute) to use when creating a taglib directive
- * for this library.
+ * to use when creating a taglib directive for this library.
*
* @return a reliable URN to a TLD like this
*/
Index: TagVariableInfo.java
===
RCS file: /home/cvs/jakarta-servletapi-5/jsr152/src/share/javax/servlet/jsp/tagext/TagVariableInfo.java,v
retrieving revision 1.2
diff -u -r1.2 TagVariableInfo.java
--- TagVariableInfo.java 19 Aug 2002 16:29:51 - 1.2
+++ TagVariableInfo.java 10 Oct 2002 23:38:30 -
@@ -139,7 +139,8 @@
/**
* The body of the lt;variable-classgt; element.
*
- * @return The name of the class of the variable
+ * @return The name of the class of the variable or
+ * 'java.lang.String' if not defined in the TLD.
*/
public String getClassName() {
@@ -149,7 +150,8 @@
/**
* The body of the lt;declaregt; element
*
- * @return Whether the variable is to be declared or not
+ * @return Whether the variable is to be declared or not.
+ * If not defined in the TLD, 'true' will be returned.
*/
public boolean getDeclare() {
@@ -159,7 +161,9 @@
/**
* The body of the lt;scopegt; element
*
- * @return The scope to give the variable.
+ * @return The scope to give the variable. NESTED
+ * scope will be returned if not defined in
+ * the TLD.
*/
public int getScope() {
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: cvs commit:jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtimePageContextImpl.java
Jan,
Think it would make more sense if the message was soemthing like
Name or attribute value was null?
On Mon, 2002-09-30 at 19:58, [EMAIL PROTECTED] wrote:
luehe 2002/09/30 16:58:58
Modified:jasper2/src/share/org/apache/jasper/runtime
PageContextImpl.java
Log:
Fixed 13140: Compliance issue - PageContext.setAttribute, under
certain circumstances, fails to throw an NPE when the value provided
is null
Revision ChangesPath
1.23 +16 -3
jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtime/PageContextImpl.java
Index: PageContextImpl.java
===
RCS file:
/home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtime/PageContextImpl.java,v
retrieving revision 1.22
retrieving revision 1.23
diff -u -r1.22 -r1.23
--- PageContextImpl.java30 Sep 2002 23:45:03 - 1.22
+++ PageContextImpl.java30 Sep 2002 23:58:58 - 1.23
@@ -232,11 +232,16 @@
}
public Object getAttribute(String name) {
+
+ if (name == null) throw new NullPointerException(Null name);
return attributes.get(name);
}
public Object getAttribute(String name, int scope) {
+
+ if (name == null) throw new NullPointerException(Null name);
+
switch (scope) {
case PAGE_SCOPE:
return attributes.get(name);
@@ -261,11 +266,19 @@
public void setAttribute(String name, Object attribute) {
+
+ if (name == null || attribute == null)
+ throw new NullPointerException(Null name or attribute value);
+
attributes.put(name, attribute);
}
public void setAttribute(String name, Object o, int scope) {
+
+ if (name == null || o == null)
+ throw new NullPointerException(Null name or attribute value);
+
switch (scope) {
case PAGE_SCOPE:
attributes.put(name, o);
--
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]
RE: can't get context with uri that is sub-dir of current context
anyone? I've re-built tomcat, fixing this so that you can retrieve a sub-context, and everything seems to work ok, but I'm just not familiar enough with the code to know if I've broken something else. Thanks, -Ryan -Original Message- From: Smith, Ryan [mailto:[EMAIL PROTECTED]] Sent: Wednesday, September 25, 2002 7:18 PM To: '[EMAIL PROTECTED]' Subject: can't get context with uri that is sub-dir of current context Can anyone shed some light on why the org.apache.catalina.core.ApplicationContext.getContext(uri) method was written to return the current context whenever the uri passed into it was a sub-dir of the current context's path? For example, calling getContext(/common) from within a context that had as it's path: /, would allways return the context of /. and, calling getContext(/store/common) from within a context that had as it's path: /store, would always return the context of /store anyone? --- Ryan Smith Platform Engineer Sony Online Entertainment (858) 577-3072 [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] --- -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
help....please.....RE: java.lang.NoSuchMethodError after re-building
I know you are all busy, but should I have posted this to the users list?
-Ryan
-Original Message-
From: Smith, Ryan [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 24, 2002 8:06 PM
To: 'Tomcat Developers List'
Subject: java.lang.NoSuchMethodError after re-building
I'm getting the following error when trying to run tc 4, after having built
the current cvs version succesfully:
java.lang.NoSuchMethodError:
javax.management.MBeanServerFactory.findMBeanServer(Ljava/lang/String;)Ljava
/util/List;
at org.apache.commons.modeler.Registry.getServer(Registry.java:245)
at
org.apache.catalina.mbeans.MBeanUtils.createServer(MBeanUtils.java:1666)
at
org.apache.catalina.mbeans.MBeanUtils.clinit(MBeanUtils.java:168)
at
org.apache.catalina.mbeans.GlobalResourcesLifecycleListener.clinit(GlobalR
esourcesLifecycleListener.java:115)
I know I have the jmx jars in place (as you can see from my ant detect
output, and build.properties files I've pasted below).
I just figured maybe someone has ran into this before, thanks for any help
!!
-Ryan
Here's a bit of config:
OS: winxp
jdk: 1.4.1
src: current cvs (as of 7:45pm)
Here's my `ant detect` output:
Buildfile: build.xml
detect:
flags:
flags.display:
[echo] --- Build environment for Catalina ---
[echo] If ${property_name} is displayed, then the property is
not set)
[echo] --- Build options ---
[echo] full.dist=${full.dist}
[echo] build.sysclasspath=${build.sysclasspath}
[echo] compile.debug=on
[echo] compile.deprecation=off
[echo] compile.optimize=on
[echo] --- Ant Flags ---
[echo] style task available (required)=true
[echo] --- JDK ---
[echo] jdk.1.2.present=true
[echo] jdk.1.3.present=true
[echo] jdk.1.4.present=true
[echo] --- Source Dependencies ---
[echo] jtc.home.present=true
[echo] --- Required Libraries ---
[echo] beanutils.present=true
[echo] collections.present=true
[echo] digester.present=true
[echo] jaxp.present=true
[echo] jndi.present=true
[echo] logging.present=true
[echo] regexp.present=true
[echo] servlet.present=true
[echo] --- Optional Libraries ---
[echo] daemon.present=true
[echo] dbcp.present=true
[echo] jaas.present=true
[echo] javamail.present=true
[echo] jmx.present=true
[echo] jsse.present=true
[echo] jta.present=true
[echo] junit.present=${junit.present}
[echo] ldap.present=true
[echo] modeler.present=true
[echo] pool.present=true
[echo] tyrex.present=${tyrex.present}
[echo] --- Required JARs ---
[echo] jndi.jar.present(except JDK 1.3+)=${jndi.jar.present}
[echo] regexp.jar.present=true
[echo] servlet.jar.present=true
[echo] xerces.jar.present(except JDK 1.4+ or
xerces2)=${xerces.jar.present}
[echo] xerces2.jars.present(except JDK 1.4+ or
xerces1)=${xerces2.jars.pres
ent}
[echo] --- Optional JARs ---
[echo] daemon.jar.present=true
[echo] dbcp.jar.present=true
[echo] jaas.jar.present=${jaas.jar.present}
[echo] javamail.jar.present=true
[echo] jdbc20ext.jar.present=${jdbc20ext.jar.present}
[echo] jmx.jar.present=true
[echo] jta.jar.present=true
[echo] junit.jar.present=${junit.jar.present}
[echo] ldap.jar.present=${ldap.jar.present}
[echo] modeler.jar.present=true
[echo] pool.jar.present=true
[echo] tyrex.jar.present=true
[echo] --- Conditional compilation flags ---
[echo] compile.daemon=true
[echo] compile.dbcp=true
[echo] compile.jaas=true
[echo] compile.javamail=true
[echo] compile.jmx=true
[echo] compile.jndi=true
[echo] compile.jsse=true
[echo] compile.jta=true
[echo] compile.junit=${compile.junit}
[echo] compile.ldap=true
[echo] compile.ssi=true
[echo] compile.tyrex=${compile.tyrex}
[echo] --- Distribution flags ---
[echo] copy.daemon.jar=true
[echo] copy.dbcp.jar=true
[echo] copy.jaas.jar=${copy.jaas.jar}
[echo] copy.jdbc20ext.jar=${copy.jdbc20ext.jar}
[echo] copy.javamail.jar=true
RE: help....please.....RE: java.lang.NoSuchMethodError after re-build ing
thanks for the help, started from scratch and had no probs
-Original Message-
From: Amy Roh [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 25, 2002 4:55 PM
To: Tomcat Developers List
Subject: Re: helpplease.RE: java.lang.NoSuchMethodError after
re-build ing
I'm getting the following error when trying to run tc 4, after having
built the current cvs version succesfully:
java.lang.NoSuchMethodError:
javax.management.MBeanServerFactory.findMBeanServer(Ljava/lang/String;
)Ljava
/util/List;
at
org.apache.commons.modeler.Registry.getServer(Registry.java:245)
at
org.apache.catalina.mbeans.MBeanUtils.createServer(MBeanUtils.java:1666)
at
org.apache.catalina.mbeans.MBeanUtils.clinit(MBeanUtils.java:168)
at
org.apache.catalina.mbeans.GlobalResourcesLifecycleListener.clinit(GlobalR
esourcesLifecycleListener.java:115)
From the errors you're receiving, it looks like you're missing proper
jmx or modeler. I'd suggest making sure that you have proper paths
pointing jmx and modeler.
Amy
I know I have the jmx jars in place (as you can see from my ant detect
output, and build.properties files I've pasted below).
I just figured maybe someone has ran into this before, thanks for any
help !!
-Ryan
Here's a bit of config:
OS: winxp
jdk: 1.4.1
src: current cvs (as of 7:45pm)
Here's my `ant detect` output:
Buildfile: build.xml
detect:
flags:
flags.display:
[echo] --- Build environment for Catalina ---
[echo] If ${property_name} is displayed, then the property is
not set)
[echo] --- Build options ---
[echo] full.dist=${full.dist}
[echo] build.sysclasspath=${build.sysclasspath}
[echo] compile.debug=on
[echo] compile.deprecation=off
[echo] compile.optimize=on
[echo] --- Ant Flags ---
[echo] style task available (required)=true
[echo] --- JDK ---
[echo] jdk.1.2.present=true
[echo] jdk.1.3.present=true
[echo] jdk.1.4.present=true
[echo] --- Source Dependencies ---
[echo] jtc.home.present=true
[echo] --- Required Libraries ---
[echo] beanutils.present=true
[echo] collections.present=true
[echo] digester.present=true
[echo] jaxp.present=true
[echo] jndi.present=true
[echo] logging.present=true
[echo] regexp.present=true
[echo] servlet.present=true
[echo] --- Optional Libraries ---
[echo] daemon.present=true
[echo] dbcp.present=true
[echo] jaas.present=true
[echo] javamail.present=true
[echo] jmx.present=true
[echo] jsse.present=true
[echo] jta.present=true
[echo] junit.present=${junit.present}
[echo] ldap.present=true
[echo] modeler.present=true
[echo] pool.present=true
[echo] tyrex.present=${tyrex.present}
[echo] --- Required JARs ---
[echo] jndi.jar.present(except JDK 1.3+)=${jndi.jar.present}
[echo] regexp.jar.present=true
[echo] servlet.jar.present=true
[echo] xerces.jar.present(except JDK 1.4+ or
xerces2)=${xerces.jar.present}
[echo] xerces2.jars.present(except JDK 1.4+ or
xerces1)=${xerces2.jars.pres
ent}
[echo] --- Optional JARs ---
[echo] daemon.jar.present=true
[echo] dbcp.jar.present=true
[echo] jaas.jar.present=${jaas.jar.present}
[echo] javamail.jar.present=true
[echo] jdbc20ext.jar.present=${jdbc20ext.jar.present}
[echo] jmx.jar.present=true
[echo] jta.jar.present=true
[echo] junit.jar.present=${junit.jar.present}
[echo] ldap.jar.present=${ldap.jar.present}
[echo] modeler.jar.present=true
[echo] pool.jar.present=true
[echo] tyrex.jar.present=true
[echo] --- Conditional compilation flags ---
[echo] compile.daemon=true
[echo] compile.dbcp=true
[echo] compile.jaas=true
[echo] compile.javamail=true
[echo] compile.jmx=true
[echo] compile.jndi=true
[echo] compile.jsse=true
[echo] compile.jta=true
[echo] compile.junit=${compile.junit}
[echo] compile.ldap=true
[echo] compile.ssi=true
[echo] compile.tyrex=${compile.tyrex}
[echo] --- Distribution flags ---
[echo] copy.daemon.jar=true
[echo] copy.dbcp.jar=true
[echo
can't get context with uri that is sub-dir of current context
Can anyone shed some light on why the org.apache.catalina.core.ApplicationContext.getContext(uri) method was written to return the current context whenever the uri passed into it was a sub-dir of the current context's path? For example, calling getContext(/common) from within a context that had as it's path: /, would allways return the context of /. and, calling getContext(/store/common) from within a context that had as it's path: /store, would always return the context of /store anyone? --- Ryan Smith Platform Engineer Sony Online Entertainment (858) 577-3072 [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] ---
Unable to Retrieve another Context from within the root context (by design, or bug?)
I'm wondering if this was by design or if it's a bug. Running Tomcat 4.1.10 I am unable to retrieve the context of another webapp from within the root context. I've got two contexts setup, one with path=/common, and another with path=. From within the context with path=, there is a jsp which attempts to retrieve the path=/common context: servletContext.getContext(/common); Unfortunately, in the org.apache.catalina.core.ApplicationContext.getContext(String uri) method, the code always returns the current context whenever the uri passed contains as it's prefix the path of the current context (context.getPath()). So, given that the root contexts path=, it gets a / appended to it, and that new string is will *always* be the prefix of any uri that is passed into the getContext(String uri) method, there is no way to ever retrieve a context for another app from within the root context. So my question is, is that by design, or is it a bug? Thanks, -Ryan --- Ryan Smith Platform Engineer Sony Online Entertainment (858) 577-3072 [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] ---
java.lang.NoSuchMethodError after re-building
I'm getting the following error when trying to run tc 4, after having built
the current cvs version succesfully:
java.lang.NoSuchMethodError:
javax.management.MBeanServerFactory.findMBeanServer(Ljava/lang/String;)Ljava
/util/List;
at org.apache.commons.modeler.Registry.getServer(Registry.java:245)
at
org.apache.catalina.mbeans.MBeanUtils.createServer(MBeanUtils.java:1666)
at
org.apache.catalina.mbeans.MBeanUtils.clinit(MBeanUtils.java:168)
at
org.apache.catalina.mbeans.GlobalResourcesLifecycleListener.clinit(GlobalR
esourcesLifecycleListener.java:115)
I know I have the jmx jars in place (as you can see from my ant detect
output, and build.properties files I've pasted below).
I just figured maybe someone has ran into this before, thanks for any help
!!
-Ryan
Here's a bit of config:
OS: winxp
jdk: 1.4.1
src: current cvs (as of 7:45pm)
Here's my `ant detect` output:
Buildfile: build.xml
detect:
flags:
flags.display:
[echo] --- Build environment for Catalina ---
[echo] If ${property_name} is displayed, then the property is
not set)
[echo] --- Build options ---
[echo] full.dist=${full.dist}
[echo] build.sysclasspath=${build.sysclasspath}
[echo] compile.debug=on
[echo] compile.deprecation=off
[echo] compile.optimize=on
[echo] --- Ant Flags ---
[echo] style task available (required)=true
[echo] --- JDK ---
[echo] jdk.1.2.present=true
[echo] jdk.1.3.present=true
[echo] jdk.1.4.present=true
[echo] --- Source Dependencies ---
[echo] jtc.home.present=true
[echo] --- Required Libraries ---
[echo] beanutils.present=true
[echo] collections.present=true
[echo] digester.present=true
[echo] jaxp.present=true
[echo] jndi.present=true
[echo] logging.present=true
[echo] regexp.present=true
[echo] servlet.present=true
[echo] --- Optional Libraries ---
[echo] daemon.present=true
[echo] dbcp.present=true
[echo] jaas.present=true
[echo] javamail.present=true
[echo] jmx.present=true
[echo] jsse.present=true
[echo] jta.present=true
[echo] junit.present=${junit.present}
[echo] ldap.present=true
[echo] modeler.present=true
[echo] pool.present=true
[echo] tyrex.present=${tyrex.present}
[echo] --- Required JARs ---
[echo] jndi.jar.present(except JDK 1.3+)=${jndi.jar.present}
[echo] regexp.jar.present=true
[echo] servlet.jar.present=true
[echo] xerces.jar.present(except JDK 1.4+ or
xerces2)=${xerces.jar.present}
[echo] xerces2.jars.present(except JDK 1.4+ or
xerces1)=${xerces2.jars.pres
ent}
[echo] --- Optional JARs ---
[echo] daemon.jar.present=true
[echo] dbcp.jar.present=true
[echo] jaas.jar.present=${jaas.jar.present}
[echo] javamail.jar.present=true
[echo] jdbc20ext.jar.present=${jdbc20ext.jar.present}
[echo] jmx.jar.present=true
[echo] jta.jar.present=true
[echo] junit.jar.present=${junit.jar.present}
[echo] ldap.jar.present=${ldap.jar.present}
[echo] modeler.jar.present=true
[echo] pool.jar.present=true
[echo] tyrex.jar.present=true
[echo] --- Conditional compilation flags ---
[echo] compile.daemon=true
[echo] compile.dbcp=true
[echo] compile.jaas=true
[echo] compile.javamail=true
[echo] compile.jmx=true
[echo] compile.jndi=true
[echo] compile.jsse=true
[echo] compile.jta=true
[echo] compile.junit=${compile.junit}
[echo] compile.ldap=true
[echo] compile.ssi=true
[echo] compile.tyrex=${compile.tyrex}
[echo] --- Distribution flags ---
[echo] copy.daemon.jar=true
[echo] copy.dbcp.jar=true
[echo] copy.jaas.jar=${copy.jaas.jar}
[echo] copy.jdbc20ext.jar=${copy.jdbc20ext.jar}
[echo] copy.javamail.jar=true
[echo] copy.jmx.jar=true
[echo] copy.jndi.jar=${copy.jndi.jar}
[echo] copy.jta.jar=true
[echo] copy.ldap.jar=${copy.ldap.jar}
[echo] copy.logging.jar=true
[echo] copy.modeler.jar=true
[echo
Re: Spec question: RE BUG 12052
On Thu, 2002-08-29 at 11:13, Craig R. McClanahan wrote: On Thu, 29 Aug 2002, Bill Barker wrote: FWIW, I agree that the 2.4 servlet-spec and the CGI-spec are out of sync here. If I actually thought that any members of the JCP were subscribed to this list, I'd think to ask for clarification before 2.4 went final. :) The way to ask would be to send feedback to [EMAIL PROTECTED]. I've forwarded the thread to the spec lead for 2.4 and it has been brought up to the expert group. However, as Craig mentioned, the best way to ask is using the email address he listed. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
RE: Spec question: RE BUG 12052
On Wed, 2002-08-28 at 17:05, Ignacio J. Ortega wrote: De: Ryan Lubke [mailto:[EMAIL PROTECTED]] Enviado el: 28 de agosto de 2002 20:29 Para: Tomcat Developers List Asunto: RE: Spec question: RE BUG 12052 The port MUST be the one in the Host Header if one is present,and should be present if the request is HTTP1.1 compliant, What if the Host header is supplied, but the value is empty. This seems legal per section 14.23 of the HTTP/1.1 RFC? 1) For me after ( another ) reading of rfc2616, from the point you named and following the references given there, i've found that in 5.2 seems to say that a empty Host: header must be responded by 400 because empty it is not a valid Host name.. But an empty host header does have a meaning, i.e. the target resource is being identified via an IP address and not a host name. I'm reading this that if the host, probably a virtual host on the server , can't be determined, then return the 400. 2) Checked Apache2 and it gives a 200 when issuing a GET / with empty Host:, at it replies a 200. I think your problem is related to issuing a 30X after a request with empty Host: header?, well it's really a border case... One never can assure if a Location header will have correct information, because is mandated to follow a filled Host hdr if present, sending a 301 with a guessed Host name in the Location, shouldnt be the worst solution to the problem.. I dont know how to make apache2 (anyone?) issue a unconditional redirect (301) after a request, but i suspect that if the Host Header is empty in such a request, it will try to form a correct Location header from the information it has at hand ( ServerName and his own local port), with Firewalls and NATs possibily an incorrect one? maybe.. In a vote, i would vote to make Tomcat issue a 400 in case of a empty Host header.. Is this such a good idea? Take a look at the following: http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7441 Apache originally would return a 400 on an empty Host header. The modified the behavior to what you're currently seeing. The interpretation could be incorrect, but I would be more inclined to follow a solid HTTP server implementation such as Apache. Saludos , Ignacio J. Ortega -- 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]
Re: HTTP Host Request header and TC Connectors]
On Wed, 2002-08-28 at 17:32, Bill Barker wrote: - Original Message - From: Ryan Lubke [EMAIL PROTECTED] To: tcdev [EMAIL PROTECTED] Sent: Wednesday, August 28, 2002 9:43 AM Subject: [Fwd: HTTP Host Request header and TC Connectors] By the way the quote was pulled from section 14.23 of RFC 2616. = Hi, Looking for a little input from the HTTP gurus here. Given the following: If the requested URI does not include an Internet host name for the service being requested, then the Host header field MUST be given with an empty value. So, I'm looking for other interpretations of what the above means. My interpretation at this point is the serviced targeted by the request URI is identified via an IP address vs a host name, that the Host request header will be sent but with an empty value. Does anyone agree/disagree? The reason I ask is that if an empty Host header is sent to Tomcat, and a redirect is sent back, the value of the Location header is useless, i.e. http:///index.jsp. My reading of 14.23 says that this is exactly what should happen, since the only (valid) way that this could happen is if the user originally requested http:///. Since the client was capable of resolving that request, it should be able to resolve the value of the Location header. Actually, the client doesn't request http:///. Example below. Trying 127.0.0.1... Connected to 127.0.0.1. Escape character is '^]'. GET / HTTP/1.1 Host: HTTP/1.1 302 Moved Temporarily Content-Type: text/html Date: Wed, 28 Aug 2002 23:01:29 GMT Transfer-Encoding: chunked Location: http:///index.html Server: Apache Tomcat/4.0.4 (HTTP/1.1 Connector) Perhaps I missed your point. I'm trying to determine if this is a problem with the client implementation's interpretation of the spec, or a problem with Tomcat. Thanks, -rl -- 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]
[Patch][Tomcat 4/5] Small issue with ErrorDispatcherValve
Given the following:
web.xml with error-page entries for the following Exception types:
- IllegalStateException
- ServletException
If a Servlet generates a RuntimeException, it is wrapped by a
ServletException. Since there is an entry for ServletException
in the web.xml, the actual root cause (in this case, an
IllegalStateException) was never unwrapped, thus leading to unexpected
behavior.
I made a simple modification to the logic so that if the throwable
passed to the ErrorDispatcherValve is a ServletException, it will
attempt to get the root cause. If the root cause is null, then proceed
with the error page lookup with the ServletException as the throwable
object, otherwise, use the root cause for the error page lookup.
I've tested this using my own tests as well as those that Watchdog 4.0
has for this area as well, and all seems OK.
If I missed something here, please let me know. I'm struggling to stay
awake here :)
-rl
PS. I've attached some very simple patches for some javadoc warnings I
noticed during the build process.
Index: ErrorDispatcherValve.java
===
RCS file: /home/cvs/jakarta-tomcat-catalina/catalina/src/share/org/apache/catalina/valves/ErrorDispatcherValve.java,v
retrieving revision 1.1.1.1
diff -u -r1.1.1.1 ErrorDispatcherValve.java
--- ErrorDispatcherValve.java 18 Jul 2002 16:47:42 - 1.1.1.1
+++ ErrorDispatcherValve.java 14 Aug 2002 03:04:47 -
@@ -212,20 +212,20 @@
*/
protected void throwable(Request request, Response response,
Throwable throwable) {
-
Context context = request.getContext();
if (context == null)
return;
-
+
Throwable realError = throwable;
-ErrorPage errorPage = findErrorPage(context, realError);
-if ((errorPage == null) (realError instanceof ServletException)) {
+
+if (realError instanceof ServletException) {
realError = ((ServletException) realError).getRootCause();
-if (realError != null)
-errorPage = findErrorPage(context, realError);
-else
+if (realError == null) {
realError = throwable;
-}
+}
+}
+
+ErrorPage errorPage = findErrorPage(context, realError);
if (errorPage != null) {
response.setAppCommitted(false);
@@ -237,7 +237,7 @@
sreq.setAttribute(Globals.ERROR_MESSAGE_ATTR,
throwable.getMessage());
sreq.setAttribute(Globals.EXCEPTION_ATTR,
- throwable);
+ realError);
Wrapper wrapper = request.getWrapper();
if (wrapper != null)
sreq.setAttribute(Globals.SERVLET_NAME_ATTR,
@@ -246,7 +246,7 @@
sreq.setAttribute(Globals.EXCEPTION_PAGE_ATTR,
((HttpServletRequest) sreq).getRequestURI());
sreq.setAttribute(Globals.EXCEPTION_TYPE_ATTR,
- throwable.getClass());
+ realError.getClass());
if (custom(request, response, errorPage)) {
try {
sresp.flushBuffer();
Index: Node.java
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/compiler/Node.java,v
retrieving revision 1.23
diff -u -r1.23 Node.java
--- Node.java 31 Jul 2002 21:42:27 - 1.23
+++ Node.java 14 Aug 2002 03:06:57 -
@@ -291,7 +291,7 @@
}
/**
- * @ return The enclosing root to this root. Usually represents the
+ * @return The enclosing root to this root. Usually represents the
* page that includes this one.
*/
public Root getParentRoot() {
Index: BodyContentImpl.java
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/runtime/BodyContentImpl.java,v
retrieving revision 1.4
diff -u -r1.4 BodyContentImpl.java
--- BodyContentImpl.java 10 Jun 2002 20:12:27 - 1.4
+++ BodyContentImpl.java 14 Aug 2002 03:07:57 -
@@ -518,7 +518,7 @@
* Note: this is after evaluation!! There are no scriptlets,
* etc in this stream.
*
- * @returns the value of this BodyJspWriter as a Reader
+ * @return the value of this BodyJspWriter as a Reader
*/
public Reader getReader() {
return new CharArrayReader (cb, 0, nextChar);
@@ -529,7 +529,7 @@
* Note: this is after evaluation!! There are no scriptlets,
* etc in this stream.
*
- * @returns the value of the BodyJspWriter as a String
+ * @return the value of the BodyJspWriter as a String
*/
public String getString() {
return new String(cb, 0, nextChar);
--
To unsubscribe, e-mail: mailto:[EMAIL
[Fwd: HEAD brach of jasper]
Hi, Has anyone been able to get any JSP's running on the HEAD brach of tomcat 4 using jasper (not jasper2). I consistently receive: java.lang.NullPointerException at org.apache.jasper.compiler.TldLocationsCache.processJars(TldLocationsCache.java:203) at org.apache.jasper.compiler.TldLocationsCache.(TldLocationsCache.java:139) at org.apache.jasper.EmbededServletOptions.(EmbededServletOptions.java:414) at org.apache.jasper.servlet.JspServlet.init(JspServlet.java:128) but looking at the line 203 of TldLocationsCache, I don't see why this occurs. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH - Jasper2] Minor compliance fix and some resource bundle keyfixes
Hi.
Two fixes:
1. JspDocumentParser - Enforce the fact that relative taglib URI's
(those that start with /) are prefixed by urn:jsptld:.
See section 5.3.5 of the 1.2 spec, as well as the examples on
page 92.
I've added the failure message to messages.properties and
messages_es.properties. Can someone point me to some online
tool that can do the necessary conversion for the
messages_ja.properties?
2. During testing of the above patch using Watchdog, some resource
bundle key lookups failed.
I've corrected the keys in Validator.java based on those found in
the resource bundles.
Thanks,
-rl
Index: Validator.java
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/compiler/Validator.java,v
retrieving revision 1.10
diff -u -r1.10 Validator.java
--- Validator.java 20 Jun 2002 23:48:23 - 1.10
+++ Validator.java 27 Jun 2002 23:21:56 -
@@ -138,14 +138,14 @@
if (language.equals(attr)) {
if (languageSeen)
- err.jspError(n, jsp.error.language.multiple);
+ err.jspError(n, jsp.error.page.multiple.language);
languageSeen = true;
if (!java.equalsIgnoreCase(value))
err.jspError(n, jsp.error.language.nonjava);
pageInfo.setLanguage(value);
} else if (extends.equals(attr)) {
if (extendsSeen)
- err.jspError(n, jsp.error.extends.multiple);
+ err.jspError(n, jsp.error.page.multiple.extends);
extendsSeen = true;
pageInfo.setExtends(value);
/*
@@ -158,7 +158,7 @@
n.addImport(value);
} else if (contentType.equals(attr)) {
if (contentTypeSeen)
- err.jspError(n, jsp.error.contentType.multiple);
+ err.jspError(n, jsp.error.page.multiple.contenttypes);
contentTypeSeen = true;
pageInfo.setContentType(value);
} else if (session.equals(attr)) {
@@ -173,7 +173,7 @@
err.jspError(n, jsp.error.session.invalid);
} else if (buffer.equals(attr)) {
if (bufferSeen)
- err.jspError(n, jsp.error.buffer.multiple);
+ err.jspError(n, jsp.error.page.multiple.buffer);
bufferSeen = true;
if (none.equalsIgnoreCase(value))
@@ -192,7 +192,7 @@
}
} else if (autoFlush.equals(attr)) {
if (autoFlushSeen)
- err.jspError(n, jsp.error.autoFlush.multiple);
+ err.jspError(n, jsp.error.page.multiple.autoflush);
autoFlushSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setAutoFlush(true);
@@ -202,7 +202,7 @@
err.jspError(n, jsp.error.autoFlush.invalid);
} else if (isthreadSafe.equals(attr)) {
if (isThreadSafeSeen)
- err.jspError(n, jsp.error.isThreadSafe.multiple);
+ err.jspError(n, jsp.error.page.multiple.threadsafe);
isThreadSafeSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setThreadSafe(true);
@@ -212,7 +212,7 @@
err.jspError(n, jsp.error.isThreadSafe.invalid);
} else if (isErrorPage.equals(attr)) {
if (isErrorPageSeen)
- err.jspError(n, jsp.error.isErrorPage.multiple);
+ err.jspError(n, jsp.error.page.multiple.iserrorpage);
isErrorPageSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setIsErrorPage(true);
@@ -222,7 +222,7 @@
err.jspError(n, jsp.error.isErrorPage.invalid);
} else if (errorPage.equals(attr)) {
if (errorPageSeen)
- err.jspError(n, jsp.error.errorPage.multiple);
+ err.jspError(n, jsp.error.page.multiple.errorpage);
errorPageSeen = true;
pageInfo.setErrorPage(value);
} else if (info.equals(attr)) {
@@ -231,7 +231,7 @@
infoSeen = true;
} else if (pageEncoding.equals(attr)) {
if (pageEncodingSeen)
- err.jspError(n, jsp.error.pageEncoding.multiple);
+ err.jspError(n, jsp.error.page.multiple.pageencoding);
pageEncodingSeen = true;
pageInfo.setPageEncoding(value);
}
Index: JspDocumentParser.java
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/compiler/JspDocumentParser.java,v
retrieving revision 1.3
diff -u -r1.3 JspDocumentParser.java
--- JspDocumentParser.java 20 Jun 2002 23:02:16 - 1.3
+++ JspDocumentParser.java 27 Jun 2002 23:04:08 -
@@ -434,6 +434,15 @@
// get the uri
String uri = attrs.getValue(i);
+
+// make sure that a relative path is prefixed
+// with urn:jsptld:path
+if (uri.charAt(0) == '/') {
+throw new JasperException(
+err.getString(jsp.error.xml.taglib.uri.not_prefixed,
+ uri));
+}
+
if (uri.startsWith(URN_JSPTLD)) {
// uri value is of the form urn:jsptld:path
uri = uri.substring(URN_JSPTLD.length());
Index: messages_es.properties
===
RCS file: /home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/resources/messages_es.properties,v
retrieving
[Fwd: [PATCH - Jasper2] Minor compliance fix and some resourcebundle key fixes]
Reposting messages.patch
-Forwarded Message-
From: Ryan Lubke [EMAIL PROTECTED]
To: tcdev [EMAIL PROTECTED]
Subject: [PATCH - Jasper2] Minor compliance fix and some resource bundle key fixes
Date: 27 Jun 2002 19:30:55 -0500
Hi.
Two fixes:
1. JspDocumentParser - Enforce the fact that relative taglib URI's
(those that start with /) are prefixed by urn:jsptld:.
See section 5.3.5 of the 1.2 spec, as well as the examples on
page 92.
I've added the failure message to messages.properties and
messages_es.properties. Can someone point me to some online
tool that can do the necessary conversion for the
messages_ja.properties?
2. During testing of the above patch using Watchdog, some resource
bundle key lookups failed.
I've corrected the keys in Validator.java based on those found in
the resource bundles.
Thanks,
-rl
Index: Validator.java
===
RCS file:
/home/cvs/jakarta-tomcat-jasper/jasper2/src/share/org/apache/jasper/compiler/Validator.java,v
retrieving revision 1.10
diff -u -r1.10 Validator.java
--- Validator.java 20 Jun 2002 23:48:23 - 1.10
+++ Validator.java 27 Jun 2002 23:21:56 -
@@ -138,14 +138,14 @@
if (language.equals(attr)) {
if (languageSeen)
- err.jspError(n, jsp.error.language.multiple);
+ err.jspError(n, jsp.error.page.multiple.language);
languageSeen = true;
if (!java.equalsIgnoreCase(value))
err.jspError(n, jsp.error.language.nonjava);
pageInfo.setLanguage(value);
} else if (extends.equals(attr)) {
if (extendsSeen)
- err.jspError(n, jsp.error.extends.multiple);
+ err.jspError(n, jsp.error.page.multiple.extends);
extendsSeen = true;
pageInfo.setExtends(value);
/*
@@ -158,7 +158,7 @@
n.addImport(value);
} else if (contentType.equals(attr)) {
if (contentTypeSeen)
- err.jspError(n, jsp.error.contentType.multiple);
+ err.jspError(n, jsp.error.page.multiple.contenttypes);
contentTypeSeen = true;
pageInfo.setContentType(value);
} else if (session.equals(attr)) {
@@ -173,7 +173,7 @@
err.jspError(n, jsp.error.session.invalid);
} else if (buffer.equals(attr)) {
if (bufferSeen)
- err.jspError(n, jsp.error.buffer.multiple);
+ err.jspError(n, jsp.error.page.multiple.buffer);
bufferSeen = true;
if (none.equalsIgnoreCase(value))
@@ -192,7 +192,7 @@
}
} else if (autoFlush.equals(attr)) {
if (autoFlushSeen)
- err.jspError(n, jsp.error.autoFlush.multiple);
+ err.jspError(n, jsp.error.page.multiple.autoflush);
autoFlushSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setAutoFlush(true);
@@ -202,7 +202,7 @@
err.jspError(n, jsp.error.autoFlush.invalid);
} else if (isthreadSafe.equals(attr)) {
if (isThreadSafeSeen)
- err.jspError(n, jsp.error.isThreadSafe.multiple);
+ err.jspError(n, jsp.error.page.multiple.threadsafe);
isThreadSafeSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setThreadSafe(true);
@@ -212,7 +212,7 @@
err.jspError(n, jsp.error.isThreadSafe.invalid);
} else if (isErrorPage.equals(attr)) {
if (isErrorPageSeen)
- err.jspError(n, jsp.error.isErrorPage.multiple);
+ err.jspError(n, jsp.error.page.multiple.iserrorpage);
isErrorPageSeen = true;
if (true.equalsIgnoreCase(value))
pageInfo.setIsErrorPage(true);
@@ -222,7 +222,7 @@
err.jspError(n, jsp.error.isErrorPage.invalid);
} else if (errorPage.equals(attr)) {
if (errorPageSeen)
- err.jspError(n, jsp.error.errorPage.multiple);
+ err.jspError(n, jsp.error.page.multiple.errorpage);
errorPageSeen = true;
pageInfo.setErrorPage(value);
} else if (info.equals(attr)) {
@@ -231,7 +231,7 @@
infoSeen = true;
} else if (pageEncoding.equals(attr)) {
if (pageEncodingSeen
Re: 5.0 proposal
Watchdog is an official TCK Actually, Watchdog is not an official TCK (just wanted to make this clear). It's nothing more than a test suite that uses the same test source. However, I do agree that performance analysis in not the goal of Watchdog. -rl -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: watchdog, servletapi, karma
Christopher, You can send me the Watchdog patches and I'll look them over. Thanks, -rl On Wed, 2002-06-12 at 12:01, Christopher K. St. John wrote: GOMEZ Henri wrote: No, I saw a problem here since servlet-api-4.0 is the OFFICIAL servlet 2.3/jsp1.2 API and may not change without (wide) notice. Yes, but there's a difference between changing the API (which is the job of the servlet spec jsr team) and fixing bugs in the implementation classes. I'd be perfectly happy if somebody else would just fix the bug, but nobody seems to be maintaining the code at the moment. -- Christopher St. John [EMAIL PROTECTED] DistribuTopia http://www.distributopia.com -- 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]
[Fwd: [PATCH] TC 4.0/Coyote locale parsing issue]
Did some checking and the TC 4.0 HTTP/1.1 and HTTP/1.0 connectors have
the same problem (o.a.c.c.http and o.a.c.c.http10).
I've modified the connectors in question and ran the same set of tests
that I ran against Coyote and all seems well. Locales with variants are
properly handled.
I've attached patches for all 3 connectors.
-rl
-Forwarded Message-
From: Ryan Lubke [EMAIL PROTECTED]
To: tcdev [EMAIL PROTECTED]
Subject: [PATCH] TC 4.0/Coyote locale parsing issue
Date: 25 Mar 2002 13:35:28 -0500
Hi,
When dealing with preferred locales from a client (Accept-Language),
the CoyoteRequest class (org.apache.coyote.tomcat4) doesn't correctly
handle locales with variants (i.e. en-IE-EURO).
RFC 2616 states that the following for language-range of
Accept-Language:
language-range = ( ( 1*8ALPHA *( - 1*8ALPHA ) ) | * )
So it seems that variants coming in from a client are legal (provided
I've read the BNF properly)
The attached patch seems to correct the issue.
-rl
Index: CoyoteRequest.java
===
RCS file:
/home/cvs/jakarta-tomcat-connectors/coyote/src/java/org/apache/coyote/tomcat4/CoyoteRequest.java,v
retrieving revision 1.11
diff -u -r1.11 CoyoteRequest.java
--- CoyoteRequest.java 19 Mar 2002 20:34:41 - 1.11
+++ CoyoteRequest.java 25 Mar 2002 14:34:54 -
@@ -1987,17 +1987,27 @@
// Extract the language and country for this entry
String language = null;
String country = null;
+String variant = null;
int dash = entry.indexOf('-');
if (dash 0) {
language = entry;
country = ;
+variant = ;
} else {
language = entry.substring(0, dash);
country = entry.substring(dash + 1);
+int vDash = country.indexOf('-');
+if (vDash 0) {
+String cTemp = country.substring(0, vDash);
+variant = country.substring(vDash + 1);
+country = cTemp;
+} else {
+variant = ;
+}
}
// Add a new Locale to the list of Locales for this quality level
-Locale locale = new Locale(language, country);
+Locale locale = new Locale(language, country, variant);
Double key = new Double(-quality); // Reverse the order
ArrayList values = (ArrayList) locales.get(key);
if (values == null) {
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Index: HttpProcessor.java
===
RCS file:
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/connector/http10/HttpProcessor.java,v
retrieving revision 1.8
diff -u -r1.8 HttpProcessor.java
--- HttpProcessor.java 18 Mar 2002 07:15:40 - 1.8
+++ HttpProcessor.java 27 Mar 2002 15:17:59 -
@@ -463,12 +463,19 @@
while (le.hasMoreElements()) {
String language = (String)le.nextElement();
String country = ;
+String variant = ;
int countryIndex = language.indexOf('-');
if (countryIndex -1) {
country = language.substring(countryIndex + 1).trim();
language = language.substring(0, countryIndex).trim();
+int vDash = country.indexOf(-);
+if (vDash 0) {
+String cTemp = country.substring(0, vDash);
+variant = country.substring(vDash + 1);
+country = cTemp;
+}
}
-request.addLocale(new Locale(language, country));
+request.addLocale(new Locale(language, country, variant));
}
}
} else if (match.equals(cookie)) {
Index: HttpProcessor.java
===
RCS file:
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/connector/http/HttpProcessor.java,v
retrieving revision 1.45
diff -u -r1.45 HttpProcessor.java
--- HttpProcessor.java 18 Mar 2002 07:15:40 - 1.45
+++ HttpProcessor.java 27 Mar 2002 15:17:39 -
@@ -499,17 +499,27 @@
// Extract the language and country for this entry
String language = null;
String country = null;
+String variant = null;
int dash = entry.indexOf('-');
if (dash 0) {
language = entry;
country = ;
+variant = ;
} else {
language = entry.substring(0, dash);
country = entry.substring(dash + 1);
+int vDash = country.indexOf
[PATCH] TC 4.0/Coyote locale parsing issue
Hi,
When dealing with preferred locales from a client (Accept-Language),
the CoyoteRequest class (org.apache.coyote.tomcat4) doesn't correctly
handle locales with variants (i.e. en-IE-EURO).
RFC 2616 states that the following for language-range of
Accept-Language:
language-range = ( ( 1*8ALPHA *( - 1*8ALPHA ) ) | * )
So it seems that variants coming in from a client are legal (provided
I've read the BNF properly)
The attached patch seems to correct the issue.
-rl
Index: CoyoteRequest.java
===
RCS file:
/home/cvs/jakarta-tomcat-connectors/coyote/src/java/org/apache/coyote/tomcat4/CoyoteRequest.java,v
retrieving revision 1.11
diff -u -r1.11 CoyoteRequest.java
--- CoyoteRequest.java 19 Mar 2002 20:34:41 - 1.11
+++ CoyoteRequest.java 25 Mar 2002 14:34:54 -
@@ -1987,17 +1987,27 @@
// Extract the language and country for this entry
String language = null;
String country = null;
+String variant = null;
int dash = entry.indexOf('-');
if (dash 0) {
language = entry;
country = ;
+variant = ;
} else {
language = entry.substring(0, dash);
country = entry.substring(dash + 1);
+int vDash = country.indexOf('-');
+if (vDash 0) {
+String cTemp = country.substring(0, vDash);
+variant = country.substring(vDash + 1);
+country = cTemp;
+} else {
+variant = ;
+}
}
// Add a new Locale to the list of Locales for this quality level
-Locale locale = new Locale(language, country);
+Locale locale = new Locale(language, country, variant);
Double key = new Double(-quality); // Reverse the order
ArrayList values = (ArrayList) locales.get(key);
if (values == null) {
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
RE: JDK 1.31??
I would just like to state that as I Covalent employee, I disavow all knowledge of this completely. :-( Ryan -- Ryan Bloom [EMAIL PROTECTED] 645 Howard St. [EMAIL PROTECTED] San Francisco, CA -Original Message- From: Pier Fumagalli [mailto:[EMAIL PROTECTED]] Sent: Wednesday, March 20, 2002 10:15 AM To: Tomcat Developers List Cc: Mark Goddard; Julian Mitchell; Isabella Pighi; Stefano Mazzocchi; Bellintani Luca; Federico Barbieri Subject: FW: JDK 1.31?? I believe this _deserves_ greater attention than webmaster... Pearls like those come once every year... Pier -- Forwarded Message From: /* NAME CONCEALED */ Date: Wed, 20 Mar 2002 11:48:15 -0500 To: [EMAIL PROTECTED] Subject: JDK 1.31?? Hello, I purchased a server package from Covalent (over a year ago) that includes an installation of Tomcat. It doesn't work and all I can get out of Covalent is: Upgrade your jdk to 1.31 and then modify the Tomcat.sh to reflect the new JAVA_HOME and start the server with the covalent script. It might as well be Chinese. I'm running Red Hat Linux 7.2. I checked all the Tomcat and Apache references and I can't find anything that says JDK 1.31. What are they talking about? A Java JDK? I can't find anything that says JDK 1.31. Please include a copy of this message with your reply. -- End of Forwarded Message -- To unsubscribe, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] For additional commands, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
RE: JDK 1.31??
Granted, but we could have provided more detailed instructions. Ryan -- Ryan Bloom [EMAIL PROTECTED] 645 Howard St. [EMAIL PROTECTED] San Francisco, CA -Original Message- From: Pier Fumagalli [mailto:[EMAIL PROTECTED]] Sent: Wednesday, March 20, 2002 10:38 AM To: Tomcat Developers List Cc: 'Mark Goddard'; 'Julian Mitchell'; 'Isabella Pighi'; 'Stefano Mazzocchi'; 'Bellintani Luca'; 'Federico Barbieri' Subject: Re: JDK 1.31?? It was not against Covalent... But rather against the user... Your instructions were concise and accurate... Even my mother could have followed them... If the guy doesn't have a clue of what a Java JDK 1.3.1 is, well, then he'd better not even touch Tomcat :) :) :) Pier Ryan Bloom [EMAIL PROTECTED] wrote: I would just like to state that as I Covalent employee, I disavow all knowledge of this completely. :-( Ryan -- Ryan Bloom [EMAIL PROTECTED] 645 Howard St. [EMAIL PROTECTED] San Francisco, CA -Original Message- From: Pier Fumagalli [mailto:[EMAIL PROTECTED]] Sent: Wednesday, March 20, 2002 10:15 AM To: Tomcat Developers List Cc: Mark Goddard; Julian Mitchell; Isabella Pighi; Stefano Mazzocchi; Bellintani Luca; Federico Barbieri Subject: FW: JDK 1.31?? I believe this _deserves_ greater attention than webmaster... Pearls like those come once every year... Pier -- Forwarded Message From: /* NAME CONCEALED */ Date: Wed, 20 Mar 2002 11:48:15 -0500 To: [EMAIL PROTECTED] Subject: JDK 1.31?? Hello, I purchased a server package from Covalent (over a year ago) that includes an installation of Tomcat. It doesn't work and all I can get out of Covalent is: Upgrade your jdk to 1.31 and then modify the Tomcat.sh to reflect the new JAVA_HOME and start the server with the covalent script. It might as well be Chinese. I'm running Red Hat Linux 7.2. I checked all the Tomcat and Apache references and I can't find anything that says JDK 1.31. What are they talking about? A Java JDK? I can't find anything that says JDK 1.31. Please include a copy of this message with your reply. -- End of Forwarded Message -- To unsubscribe, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] For additional commands, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] -- To unsubscribe, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] For additional commands, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] -- To unsubscribe, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] For additional commands, e-mail: mailto:tomcat-dev- [EMAIL PROTECTED] -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: Watchdog problems
Hi, I have an idea for a fix. I'll have this implemented today. -rl On Thu, 2002-01-17 at 06:56, Remy Maucherat wrote: Hi, I have some problems to run the Watchdog tests on mod_webapp and I think there is a problem in watchdog: For example: +++ [java] [gtest] Unable to find the expected header: 'Content-Type: text/plain' in the server's response. [java] [gtest] The following headers were received: [java] [gtest] HEADER - Date: Thu, 17 Jan 2002 11:22:24 GMT [java] [gtest] HEADER - Server: Apache/2.0.31-dev (Unix) mod_ssl/3.0a0 OpenSSL/0.9.6b [java] [gtest] HEADER - Connection: close [java] [gtest] HEADER - Content-Type: text/plain; charset=ISO-8859-1 [java] [gtest] FAIL GET /servlet-tests/Include_1Test HTTP/1.0 +++ I DO think that mod_webapp sends a valid Content-Type but watchdog does not like it! Any comments? I agree. ISO-8859-1 is the default encoding, but obviously it can be explicitely specified. Watchdog is rather not flexible, so I don't know how easy it would be to fix this. Remy -- 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]
Re: [VOTES] Tomcat 4.0.2 beta 1
Hi, Did you rebuild/redploy and test with the latest Watchdog? -rl On Wed, 2001-12-19 at 11:16, jean-frederic clere wrote: Hi, I have noted that 4 of the jsp tests fail: jsp-tests/jsp/tagext/TagSupport/positiveDoStartTagXML.jsp jsp-tests/jsp/tagext/TagSupport/positiveDoStartTag.jsp jsp-tests/jsp/tagext/TagSupport/positiveDoStartTagXML.jsp jsp-tests/jsp/tagext/TagSupport/positiveDoStartTag.jsp It sounds like a tag library is missing (in my test environment) the line where the exception comes from is TagLibraryInfo tlibInfo=tagInfo.getTagLibrary(); From TestTag.java (of jakarta-watchdog-4.0). The WARP part needs a little review... Cheers Jean-frederic Remy Maucherat wrote: Hi, After some delay, I'd like to release the first beta of Tomcat 4.0.2 this week (the sooner, the better, so I plan to package the release as soon as this vote is considered approved). This release has already been approved, and although there are some issues which will need to be addressed before 4.0.2 final is released, there is no open showstopper problems. A few additional fixes may be made before beta 1 is released. The list of changes and fixes is available at: http://cvs.apache.org/viewcvs.cgi/~checkout~/jakarta-tomcat-4.0/Attic/RELEAS E-NOTES-4.0.2-B1.txt?rev=1.1.2.4 ballot [ ] +1: Make the release [ ] -1: I'm opposed to the release until the following issues are fixed: /ballot There is also a new feature that has been added in the HEAD branch which could be worth porting: the instance pooling for STM servlets. I wrote the code as an experiment, but it appears to be working very well (I didn't notice any thread safety issues during load testing). ballot [ ] +1: Add the feature to the 4.0 branch [ ] -1: Don't add the feature, because: /ballot Remy -- 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]
[PATCH] Tomcat 4.0 BUILDING.txt
Hi,
After going through the exercise of building TC, I noticed
one issue with the instructions contained in BUILDING.txt.
The download URL referenced in the BUILDING.txt file to obtain
JAXP 1.1 is now stale.
Attached is a possible patch to update the BUILDING.txt
file.
NOTE: This release of JAXP doesn't include a jaxp.jar, so
the jaxp.jar variable within the build.properties isn't
really needed, but doesn't appear to affect the build process
if it's left in.
Index: BUILDING.txt
===
RCS file: /home/cvspublic/jakarta-tomcat-4.0/BUILDING.txt,v
retrieving revision 1.15
diff -u -r1.15 BUILDING.txt
--- BUILDING.txt2001/12/01 07:32:59 1.15
+++ BUILDING.txt2001/12/14 19:46:44
@@ -57,17 +57,18 @@
available, which will be used to actually perform the build.
-(2) Download and Install the JAXP/1.1 Reference Implementation
+(2) Download and Install the Java XML Pack Binary Distribution
-* Download a binary distribution of JAXP 1.1 (Final Version) from:
+* Download a binary distribution of the Java XML Pack:
-http://java.sun.com/xml/download.html
+http://java.sun.com/xml/downloads/javaxmlpack.html
* Unpack the binary distribution into a convenient location so that the
- JAXP/1.1 release resides in its own directory (conventionally named
- jaxp-1.1. For the purposes of the remainder of this document, the
- symbolic name ${jaxp.home} is used to refer to the full pathname of
- the release directory.
+ Java XML Pack release resides in its own directory (the JAXP libraries
+ reside in the jaxp-1.1.3 subdirectory). For the purposes of the
+ remainder of this document, the symbolic name ${jaxp.home} is used
+ to refer to the full pathname to the jaxp-1.1.3 subdirectory of the
+ Java XML Pack release directory.
* Make the xalan.jar file of this distribution available to Ant (so that
it can be used with the style tag) by copying it to ${ant.home}/lib.
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog 4.0 -- New Filter tests + JSP web.xml fix for DTDreference
Hello,
In the attached tarball are source and goldenfiles for some new
Filter tests. Unpack the tarball in the src directory of Watchdog.
Also attached are patches for the servlet-gtest.xml and the web.xml for
the servlet tests.
Also attached is a patch for the jsp web.xml to correct the DTD
reference in the DOCTYPE element.
Comments welcome.
-rl
filter.tar.gz
Description: GNU Zip compressed data
Index: servlet-gtest.xml
===
RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/conf/servlet-gtest.xml,v
retrieving revision 1.19
diff -u -r1.19 servlet-gtest.xml
--- servlet-gtest.xml 2001/07/20 23:07:56 1.19
+++ servlet-gtest.xml 2001/12/03 20:25:15
@@ -10,9 +10,27 @@
!-- Watchdog Servlet tests --
- !-- javax_servlet_GenericServlet tests --
target name=gtestservlet-test
+
+ !-- javax_servlet_Filter tests --
+
+ gtest request=GET /servlet-tests/DoFilterTest HTTP/1.0
+ debug=0 host=${host} port=${port}
+ goldenFile=${wgdir}/javax_servlet/Filter/DoFilterTest.html
+ assertion=The doFilter method of the Filter is called by the container
+each time a request/response pair is passed through the stack due to a client request
+for the Servlet in the stack. Java Servlet Specification v2.3, Sec. 14
+ testStrategy=Client attempts to access a servlet and the filter
+configured for that servlet should be invoked.
+ /
+
+ gtest request=GET /servlet-tests/InitFilterConfigTest HTTP/1.0
+ debug=0 host=${host} port=${port}
+ goldenFile=${wgdir}/javax_servlet/Filter/InitFilterConfigTest.html
+ assertion=The container calls this method when the Filter is
+instantiated and passes in a FilterConfig object. Java Servlet Specification v2.3,
+Sec. 14
+ testStrategy=Client attempts to access a servlet and the filter
+configured for that servlet.
+ /
+
+ !-- javax_servlet_GenericServlet tests --
+
gtest request=GET /servlet-tests/DestroyTest HTTP/1.0
debug=0 host=${host} port=${port}
returnCode=200
Index: web.xml
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/server/servlet-tests/WEB-INF/web.xml,v
retrieving revision 1.3
diff -u -r1.3 web.xml
--- web.xml 2000/12/22 18:59:14 1.3
+++ web.xml 2001/12/03 20:25:43
@@ -1,27 +1,85 @@
?xml version=1.0 encoding=ISO-8859-1?
!DOCTYPE web-app
-PUBLIC -//Sun Microsystems, Inc.//DTD Web Application 2.2//EN
-http://java.sun.com/j2ee/dtds/web-app_2_2.dtd;
+PUBLIC -//Sun Microsystems, Inc.//DTD Web Application 2.3//EN
+http://java.sun.com/dtd/web-app_2_3.dtd;
web-app
context-param
- param-name
- OS
- /param-name
- param-value
- ULTRASPARC
- /param-value
+ param-name
+ OS
+ /param-name
+ param-value
+ ULTRASPARC
+ /param-value
/context-param
context-param
- param-name
- EDITOR
- /param-name
- param-value
- VI
- /param-value
+ param-name
+ EDITOR
+ /param-name
+ param-value
+ VI
+ /param-value
/context-param
+!-- Filters --
+filter
+filter-name
+ServletMappedDoFilter_Filter
+/filter-name
+display-name
+ServletMappedDoFilter_Filter
+/display-name
+filter-class
+tests.javax_servlet.Filter.DoFilter_Filter
+/filter-class
+init-param
+param-name
+attribute
+/param-name
+param-value
+tests.javax_servlet.Filter.DoFilter_Filter.SERVLET_MAPPED
+/param-value
+/init-param
+/filter
+filter
+filter-name
+ServletMappedInitFilter_Filter
+/filter-name
+display-name
+ServletMappedInitFilter_Filter
+/display-name
+filter-class
+tests.javax_servlet.Filter.InitFilter_Filter
+/filter-class
+init-param
+param-name
+attribute
+/param-name
+param-value
+tests.javax_servlet.Filter.InitFilter_Filter.SERVLET_MAPPED
+/param-value
+/init-param
+ /filter
+ filter-mapping
+ filter-name
+ ServletMappedDoFilter_Filter
+ /filter-name
+ servlet-name
+ DoFilterTest
+ /servlet-name
+ /filter-mapping
+ filter-mapping
+ filter-name
+ ServletMappedInitFilter_Filter
+ /filter-name
+ servlet-name
+ InitFilterConfigTest
+/servlet-name
+ /filter-mapping
+
+ !-- END
[PATCH] Watchdog-4.0 -- New Tests tests.javax_servlet.Error
New tests for verification of error page handling by the container.
The tests validate the use of dynamic and static resources as error
pages and validate that the appropriate request parameters are set
depending on the circumstances when the error resource is dynamic.
The attached .gz file can be untarred in the root directory of the
watchdog distribution and all new files will be put into the proper
directories. Also attached are patches to the servlet-tests web
application's web.xml, and the servlet-gtest.xml file.
Comments are welcome.
-rl
javax_servlet_Error.tar.gz
Description: GNU Zip compressed data
Index: servlet-gtest.xml
===
RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/conf/servlet-gtest.xml,v
retrieving revision 1.19
diff -u -r1.19 servlet-gtest.xml
--- servlet-gtest.xml 2001/07/20 23:07:56 1.19
+++ servlet-gtest.xml 2001/11/30 20:23:12
@@ -586,6 +586,49 @@
debug=0 host=${host} port=${port}
goldenFile=${wgdir}/javax_servlet/UnavailableException/UnavailableException_Constructor2Test.html
/
+ !-- javax_servlet_Error tests --
+
+gtest request=GET /servlet-tests/ServletToServletErrorPageTest HTTP/1.0
+debug=0 host=${host} port=${port}
+goldenfile=${wgdir}/javax_servlet/Error/ServletToServletErrorPageTest.html
+testName=ServletToServletErrorPageTest
+assertion=Servlet error handlers can be installed for specific Exception
+types.
+testStrategy=The target servlet will generage a
+java.lang.ArithmeticException. The Servlet error page should display non-null values
+for status_code, exception_type, message, exception, request_uri, and servlet_name
+/
+
+gtest request=GET /servlet-tests/ServletToServletError501PageTest HTTP/1.0
+debug=0 host=${host} port=${port}
+
+goldenfile=${wgdir}/javax_servlet/Error/ServletToServletError501PageTest.html
+testName=ServletToServletError501PageTest
+assertion=Servlet error handlers can be installed for specific HTTP status
+codes.
+testStrategy=The target servlet will class HttpServletResponse.sendError(
+int status, String message ). The Servlet error page should display non-null values
+for status_code, message, request_uri, and servlet_name
+/
+
+gtest request=GET /servlet-tests/ServletToJSPErrorPageTest HTTP/1.0
+debug=0 host=${host} port=${port}
+goldenfile=${wgdir}/javax_servlet/Error/ServletToJSPErrorPageTest.html
+testName=ServletToJSPErrorPageTest
+assertion=JSP error handlers can be installed for specific Exception types.
+testStrategy=The target servlet will generage a
+java.lang.ArrayIndexOutOfBoundsException. The JSP error page should display non-null
+values for status_code, exception_type, message, exception, request_uri, and
+servlet_name
+/
+
+
+gtest request=GET /servlet-tests/ServletToJSPError502PageTest HTTP/1.0
+debug=0 host=${host} port=${port}
+goldenfile=${wgdir}/javax_servlet/Error/ServletToJSPError502PageTest.html
+testName=ServletToJSPError502PageTest
+assertion=JSP error handlers can be installed for specific HTTP status
+codes.
+testStrategy=The target servlet will generage a
+java.lang.ArrayIndexOutOfBoundsException. The JSP error page should display non-null
+values for status_code, message, request_uri, and servlet_name
+/
+
+
+gtest request=GET /servlet-tests/ServletToHTMLErrorPageTest HTTP/1.0
+debug=0 host=${host} port=${port}
+goldenfile=${wgdir}/javax_servlet/Error/ServletToHTMLErrorPageTest.html
+testName=ServletToHTMLErrorPageTest
+assertion=Static HTML page error handlers can be installed for specific
+Exception types.
+testStrategy=The target servlet will generage a
+java.lang.NumberFormatException. The static HTML error page should be returned to
+the client.
+/
!-- javax_servlet_http_Cookie tests --
Index: web.xml
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/server/servlet-tests/WEB-INF/web.xml,v
retrieving revision 1.3
diff -u -r1.3 web.xml
--- web.xml 2000/12/22 18:59:14 1.3
+++ web.xml 2001/11/30 20:23:35
@@ -242,6 +242,71 @@
/servlet-class
/servlet
+ !-- Error --
+
+ servlet
+servlet-name
+ServletToServletErrorPageTest
+/servlet-name
+servlet-class
+tests.javax_servlet.Error.ServletToServletErrorPageTestServlet
+/servlet-class
+ /servlet
+
+ servlet
+servlet-name
+ServletToServletError501PageTest
+/servlet-name
+servlet-class
+tests.javax_servlet.Error.ServletToServletError501PageTestServlet
+/servlet-class
+ /servlet
+
+ servlet
+servlet-name
+JSPErrorPage
+/servlet-name
+jsp-file
+/tests/javax_servlet/Error/JSPErrorPage.jsp
+/jsp-file
[PATCH] Watchdog-4.0 - Correction/cleanup of jsp_precompile tests
I've cleaned up the jsp_precompile tests in Watchdog-4.0 in response to
Bug 5169.
- removed the goldenfile checks in the jsp-gtest.xml for those
precompile tests that used them.
- removed commented and invalid tests from that section
- added new functionality to GTest to assert cases where no
response body is expected from the server (precompile tests
are such a case).
- cleanup the assertion and test strategy text
The following goldenfiles should be removed from the repository.
${wgdir}/misc/precompilation/response.html
${wgdir}/misc/precompilation/no_parameter.html
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java 2001/09/28 04:09:56 1.3
+++ GTest.java 2001/11/29 23:31:03
@@ -29,6 +29,7 @@
// Expected response
boolean magnitude=true;
boolean exactMatch=false;
+boolean expectResponseBody=true;
// Match the body against a golden file
String goldenFile;
// Match the body against a string
@@ -163,6 +164,19 @@
exactMatch=Boolean.valueOf( exact ).booleanValue();
}
+/**
+ * codesetExpectResponseBody/code will indicate
+ * whether or not a response body is expected from the
+ * server for a particular test. In some cases, no
+ * body should be returned to the client. This allows
+ * that assertion.
+ *
+ * @param b a codeboolean/code value
+ */
+public void setExpectResponseBody( boolean b ) {
+expectResponseBody = b;
+}
+
/** Set the port as int - different name to avoid confusing ant
*/
public void setPortInt(int i) {
@@ -356,7 +370,7 @@
if( responseMatch != null ) {
// check if we got the string we wanted
- if( responseBody == null ) {
+ if( responseBody == null expectResponseBody ) {
System.out.println(ERROR: got no response, expecting +
responseMatch);
return false;
}
@@ -366,6 +380,11 @@
System.out.println(responseBody );
}
}
+
+if ( !expectResponseBody responseBody != null ) {
+System.out.println( ERROR: Received a response body from the server where
+none was expected );
+return false;
+}
// compare the body
if( goldenFile==null) return responseStatus;
Index: jsp-gtest.xml
===
RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/conf/jsp-gtest.xml,v
retrieving revision 1.9
diff -u -r1.9 jsp-gtest.xml
--- jsp-gtest.xml 2001/08/09 17:35:13 1.9
+++ jsp-gtest.xml 2001/11/29 23:39:02
@@ -15,63 +15,38 @@
!-- Precompilation tests --
gtest request=GET
/jsp-tests/jsp/misc/precompilation/precompile.jsp?jsp_precompile HTTP/1.0
debug=0 host=${host} port =${port}
+ expectResponseBody=false
returncode=200
- testName=precompile.jsp
- assertion=Test that the request is not sent to the JSP
- testStrategy=No HTML page received when jsp_precompile has no value
- /
+ testName=precompileNoValueTest
+ assertion=If the jsp_precompile request parameter has no value, the
+request will not be delivered to the target JSP page. JavaServer Pages Specification
+v1.2, Sec. 8.4.2
+ testStrategy=Validate that no response body is returned when
+jsp_precompile has no value
+ /
- gtest request=GET
/jsp-tests/jsp/misc/precompilation/precompile.jsp?jsp_precompile=falseamp;test=insignificant
HTTP/1.0
- debug=0 host=${host} port =${port}
- goldenFile=${wgdir}/misc/precompilation/response.html
- testName=precompile.jsp
- assertion=Test that the request is forwarded properly when
jsp_precompile=false
- testStrategy=The value of variable passed in the request is displayed if
the request goes to the JSP
- /
-
gtest request=GET
/jsp-tests/jsp/misc/precompilation/precompile.jsp?jsp_precompile=false HTTP/1.0
- debug=0 host=${host} port =${port}
- goldenFile=${wgdir}/misc/precompilation/no_parameter.html
- testName=precompile.jsp
- assertion=Test that the request is forwarded properly when
jsp_precompile=false
- testStrategy=The request should be processed
- /
-
-!--
- gtest request=GET
/jsp-tests/jsp/misc/precompilation/precompile.jsp?jsp_precompile=%22false%22 HTTP/1.0
debug=0 host=${host} port =${port}
- goldenFile=${wgdir}/misc/precompilation/no_parameter.html
- testName=precompile.jsp
- assertion=Test that the request is forwarded properly when
jsp_precompile=%22false%22
- testStrategy=The request should be processed
- /
---
+
[PATCH] Tomcat 4.0 -- Spelling correction in JspServlet class
Simple spelling correction in org.apache.jasper.servlet.JspServlet Index: JspServlet.java === RCS file: /home/cvspublic/jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/servlet/JspServlet.java,v retrieving revision 1.23 diff -u -r1.23 JspServlet.java --- JspServlet.java 2001/11/26 22:55:43 1.23 +++ JspServlet.java 2001/11/28 17:20:56 @@ -424,7 +424,7 @@ else if (value.equals(false)) return (false);// ?jsp_precompile=false else -throw new ServletException(Cannott have request parameter + +throw new ServletException(Cannot have request parameter + Constants.PRECOMPILE + set to + value); -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Proposed Fix for Bugzilla 5169
Synopsis of 5169: JavaServer Pages Specification 1.2, Section JSP.8.4.2: The jsp_precompile parameter may have no value, or may have the values true or false. In all cases, the request should not be delivered to the JSP page. With Tomcat, if the jsp_precompile parameter is false, the request is delivered to the target JSP page. This is a violation of the spec. The issue seems to lie in org/apache/jasper/servlet/JspServlet.java. During the processing of the request parameters, if the jsp_precompile value evaluated to false, the preCompile method would return false. Because of this, the service method in the JspServletWrapper class would not return as the precompile value passed was false. The attached simple patch seems to resolve the issue running what tests I have. Comments welcome. Thanks, -rl Index: JspServlet.java === RCS file: /home/cvspublic/jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/servlet/JspServlet.java,v retrieving revision 1.24 diff -u -r1.24 JspServlet.java --- JspServlet.java 2001/11/28 17:26:08 1.24 +++ JspServlet.java 2001/11/28 21:40:18 @@ -422,7 +422,7 @@ if (value.equals(true)) return (true); // ?jsp_precompile=true else if (value.equals(false)) -return (false);// ?jsp_precompile=false +return (true);// ?jsp_precompile=false else throw new ServletException(Cannot have request parameter + Constants.PRECOMPILE + set to + -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
RE: [VOTE] Tomcat 4.0.2 Release Plan
Hi Andrew, Take a look at the following URL: http://jakarta.apache.org/site/getinvolved.html On Wed, 2001-11-28 at 19:10, Andrew Bruno wrote: How can this kind of documentation be inclcuded? http://www.vacodi.com/howto/tomcat/iisnt/ I am happy to do some extra work (fix some content, etc..), but what standard and process do I follow? if there is one? How can it be submitted by me? or anyone else is welcome to grab it, and put in CVS, docs, etc Basically you would create a documentation patch of some sort from the anon CVS tree which you could submit to the developer list. A committer monitoring the list would apply the patch to the CVS tree if appropriate. If it were included in the next release it would help a lot of people, and hopefully reduce some traffic on the tomcat-user mailing list. Regards, Andrew Hope this helps. -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog-4.0 -- GetMethod_HEADTest patch
Hi,
This test, the way it is currently written, expects that whent he client
invokes the HTTP HEAD method, that a response-body is returned. This is
incorrect. The test currently passes due to Tomcat Bugzilla 5126.
I've modified the test so that no response output is written via a
writer and instead will use sendError( int, string ), to send the status
of the test. I had to add a new instance variable and setter method to
GTest so that the response message could be configured by the
servlet-gtest.xml file, and be checked appropriately by the
checkResponse() method of GTest.
NOTE: The goldenfile
${wgdir}/javax_servlet_http/HttpServletRequest/GetMethod_HEADTest.html
should be removed from the repository.
Attached are the patches necessary to correct this test.
Comments welcome.
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java 2001/09/28 04:09:56 1.3
+++ GTest.java 2001/11/27 15:19:57
@@ -37,6 +37,7 @@
Hashtable expectHeaders;
// Match request line
String returnCode=;
+String returnCodeMsg = ;
// Actual response
String responseLine;
@@ -221,6 +222,10 @@
this.returnCode=s;
}
+ public void setReturnCodeMsg( String s ) {
+ this.returnCodeMsg = s;
+ }
+
public void setHeaders( String s ) {
requestHeaders=new Hashtable();
parseHeader( s, requestHeaders );
@@ -309,7 +314,13 @@
// If returnCode doesn't match
if( request.indexOf( HTTP/1. ) -1) {
- boolean match= ( responseLine!=null responseLine.indexOf(returnCode)
-1);
+boolean match = false;
+if ( returnCodeMsg == ) {
+ match = ( responseLine != null responseLine.indexOf( returnCode )
+-1 );
+} else {
+match = ( responseLine != null responseLine.indexOf( returnCode ) -1
+ responseLine.indexOf( returnCodeMsg ) -1 );
+}
if( match != testCondition ) {
responseStatus = false;
System.out.println(ERROR in: + request);
Index: GetMethod_HEADTestServlet.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/server/servlet-tests/WEB-INF/classes/tests/javax_servlet_http/HttpServletRequest/GetMethod_HEADTestServlet.java,v
retrieving revision 1.1
diff -u -r1.1 GetMethod_HEADTestServlet.java
--- GetMethod_HEADTestServlet.java 2000/10/16 22:38:56 1.1
+++ GetMethod_HEADTestServlet.java 2001/11/27 14:52:17
@@ -80,20 +80,16 @@
public void service (HttpServletRequest request, HttpServletResponse response)
throws ServletException, IOException {
- PrintWriter out = response.getWriter();
String method = request.getMethod();
if(method!=null) {
if(method.trim().equalsIgnoreCase(HEAD))
{
- out.println(GetMethod_HEADTest test PASSED);
+ response.sendError( HttpServletResponse.SC_OK,
+GetMethod_HEADTest PASSED);
return;
}
}
- out.println(GetMethod_HEADTest test FAILED);
- out.println( BRgetMethod did not return proper method type);
- out.println(Actual Return Value : + method );
- out.println(Expected Return Value : HEAD );
+response.sendError( HttpServletResponse.SC_INTERNAL_SERVER_ERROR,
+GetMethod_HEADTest FAILED );
}
}
Index: servlet-gtest.xml
===
RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/conf/servlet-gtest.xml,v
retrieving revision 1.19
diff -u -r1.19 servlet-gtest.xml
--- servlet-gtest.xml 2001/07/20 23:07:56 1.19
+++ servlet-gtest.xml 2001/11/27 15:14:13
@@ -739,7 +739,7 @@
gtest request=HEAD /servlet-tests/GetMethod_HEADTest HTTP/1.0
debug=0 host=${host} port=${port}
-
goldenFile=${wgdir}/javax_servlet_http/HttpServletRequest/GetMethod_HEADTest.html /
+ returnCode=200 returnCodeMsg=GetMethod_HEADTest PASSED /
gtest request=POST /servlet-tests/GetMethod_POSTTest HTTP/1.0
debug=0 host=${host} port=${port}
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog 4.0 - Fix for Bugzilla 3527
Classes referenced by TLD are not in the repository. As stated in the bug, some containers are very strict about having all classes available that are referenced in a TLD. The included patches removed the unused references from the TLDs. Comments welcome. -rl Index: tld_uri.tld === RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/server/jsp-tests/WEB-INF/tlds/tld_uri.tld,v retrieving revision 1.2 diff -u -r1.2 tld_uri.tld --- tld_uri.tld 2000/11/30 23:53:58 1.2 +++ tld_uri.tld 2001/11/20 12:34:42 @@ -16,22 +16,7 @@ description A simple tab library for testing computation of TLD Resource Path /description -validator - validator-classtags.TagLibraryValidator.Validator/validator-class -init-param -param-nameusa/param-name -param-valueWashington/param-value -descriptionCapital/description -/init-param -init-param -param-nameindia/param-name -param-valueNew Delhi/param-value -descriptionCapital/description -/init-param -/validator - - !-- general validator tag -- tag nametldtag/name tag-classtags.tld_resource_path.tld_URI/tag-class Index: TestLib.tld === RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/server/jsp-tests/WEB-INF/tlds/TestLib.tld,v retrieving revision 1.8 diff -u -r1.8 TestLib.tld --- TestLib.tld 2000/12/01 00:19:25 1.8 +++ TestLib.tld 2001/11/20 12:34:30 @@ -201,24 +201,6 @@ /tag tag -namepropertyEditor/name -tag-classtags.PropertyEditorTag/tag-class -description -A tag that Uses ProprtyEditorSupport class -/description - -attribute - nameurl/name - requiredfalse/required -/attribute - -attribute - namelocale/name - requiredfalse/required -/attribute -/tag - -tag nameiterator/name tag-classtags.SimpleIteratorTag/tag-class description @@ -310,21 +292,5 @@ requiredtrue/required /attribute /tag - -tag -namethirdWinner/name -tag-classtags.TagSupport.thirdPrizeTag/tag-class -description -A tag that sets the third prize winner and also returns it to a -nested tag -/description -attribute - namewinner/name - requiredtrue/required -/attribute -/tag - - - /taglib -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog 4.0 - Fix for Bugzilla 3904
GTest currently verifies response headers in a case sensative
manner. This patch resolves the issue.
Fix is based of suggested fix in Bug report.
Comments welcome.
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java 2001/09/28 04:09:56 1.3
+++ GTest.java 2001/11/19 21:36:00
@@ -15,6 +15,11 @@
// derived from Jsp
public class GTest extends Task {
+
+private static final String ZEROS = ;
+private static final int SHORTPADSIZE = 4;
+private static final int BYTEPADSIZE = 2;
+
String prefix=http://localhost:8080/test;;
String host=localhost;
int port=8080;
@@ -376,9 +381,9 @@
boolean cmp=true;
if(exactMatch)
- cmp=compare(responseBody, expResult.toString() );
+ cmp=compare(responseBody.getBytes(), expResult.toString().getBytes() );
else
- cmp=compareWeek( responseBody, expResult.toString());
+ cmp=compareWeak( responseBody, expResult.toString());
if( cmp != testCondition ) {
responseStatus = false;
@@ -521,30 +526,200 @@
}
}
+/*
+ * codecompare/code compares the two byte arrays passed
+ * in to verify that the lengths of the arrays are equal, and
+ * that the content of the two arrays, byte for byte are equal.
+ *
+ * @param fromServer a codebyte[]/code value
+ * @param fromGoldenFile a codebyte[]/code value
+ * @return codeboolean/code true if equal, otherwise false
+ */
+private boolean compare( byte[] fromServer, byte[] fromGoldenFile ) {
+if ( fromServer == null || fromGoldenFile == null ) {
+return false;
+}
+
+/*
+ * Check to see that the respose and golden file lengths
+ * are equal. If they are not, dump the hex and don't
+ * bother comparing the bytes. If they are equal,
+ * iterate through the byte arrays and compare each byte.
+ * If the bytes don't match, dump the hex representation
+ * of the server response and the goldenfile and return
+ * false.
+ */
+
+if ( fromServer.length != fromGoldenFile.length ) {
+StringBuffer sb = new StringBuffer( 50 );
+sb.append( Response and golden file lengths do not match!\n );
+sb.append( Server response length: );
+sb.append( fromServer.length );
+sb.append( \nGoldenfile length: );
+sb.append( fromGoldenFile.length );
+System.out.println( sb.toString() );
+sb = null;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+
+} else {
+
+int i = 0;
+int j = 0;
+
+while ( ( i fromServer.length ) ( j fromGoldenFile.length ) ) {
+if ( fromServer[ i ] != fromGoldenFile[ j ] ) {
+System.out.println( Error at position + ( i + 1 ) );
-// Compare the actual result and the expected result.
-private boolean compare(String str1, String str2) {
- //System.out.println(In compare);
- if ( str1==null || str2==null) return false;
- if ( str1.length() != str2.length() ) {
- System.out.println(Wrong size + str1.length() + + str2.length() );
- return false;
- }
-
-for(int i=0; istr1.length() ; i++ ) {
-if (str1.charAt( i ) != str2.charAt( i ) ) {
- System.out.println(Error at + i + + str1.charAt(1) +
- str2.charAt(i));
-return false;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+}
+
+i++;
+j++;
+}
+}
+
+return true;
+}
+
+/*
+ * codedumpHex/code helper method to dump formatted
+ * hex output of the server response and the goldenfile.
+ *
+ * @param serverResponse a codebyte[]/code value
+ * @param goldenFile a codebyte[]/code value
+ */
+private void dumpHex( byte[] serverResponse, byte[] goldenFile ) {
+StringBuffer outBuf = new StringBuffer( ( serverResponse.length +
+goldenFile.length ) * 2 );
+
+String fromServerString = getHexValue( serverResponse, 0,
+serverResponse.length );
+String fromGoldenFileString = getHexValue( goldenFile, 0, goldenFile.length );
+
+outBuf.append( Hex dump of server response and goldenfile below.\n\n###
+RESPONSE FROM SERVER ###\n );
+outBuf.append( \n );
+outBuf.append( fromServerString );
+outBuf.append(
Re: [PATCH] Watchdog 4.0 - Fix for Bugzilla 3904
Apologies,
I attached the wrong patch in the previous email. The correct
patch for Bugzilla 3904 is now attached.
-rl
On Tue, 2001-11-20 at 09:07, Ryan Lubke wrote:
GTest currently verifies response headers in a case sensative
manner. This patch resolves the issue.
Fix is based of suggested fix in Bug report.
Comments welcome.
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java2001/09/28 04:09:56 1.3
+++ GTest.java2001/11/19 21:36:00
@@ -15,6 +15,11 @@
// derived from Jsp
public class GTest extends Task {
+
+private static final String ZEROS = ;
+private static final int SHORTPADSIZE = 4;
+private static final int BYTEPADSIZE = 2;
+
String prefix=http://localhost:8080/test;;
String host=localhost;
int port=8080;
@@ -376,9 +381,9 @@
boolean cmp=true;
if(exactMatch)
- cmp=compare(responseBody, expResult.toString() );
+ cmp=compare(responseBody.getBytes(), expResult.toString().getBytes() );
else
- cmp=compareWeek( responseBody, expResult.toString());
+ cmp=compareWeak( responseBody, expResult.toString());
if( cmp != testCondition ) {
responseStatus = false;
@@ -521,30 +526,200 @@
}
}
+/*
+ * codecompare/code compares the two byte arrays passed
+ * in to verify that the lengths of the arrays are equal, and
+ * that the content of the two arrays, byte for byte are equal.
+ *
+ * @param fromServer a codebyte[]/code value
+ * @param fromGoldenFile a codebyte[]/code value
+ * @return codeboolean/code true if equal, otherwise false
+ */
+private boolean compare( byte[] fromServer, byte[] fromGoldenFile ) {
+if ( fromServer == null || fromGoldenFile == null ) {
+return false;
+}
+
+/*
+ * Check to see that the respose and golden file lengths
+ * are equal. If they are not, dump the hex and don't
+ * bother comparing the bytes. If they are equal,
+ * iterate through the byte arrays and compare each byte.
+ * If the bytes don't match, dump the hex representation
+ * of the server response and the goldenfile and return
+ * false.
+ */
+
+if ( fromServer.length != fromGoldenFile.length ) {
+StringBuffer sb = new StringBuffer( 50 );
+sb.append( Response and golden file lengths do not match!\n );
+sb.append( Server response length: );
+sb.append( fromServer.length );
+sb.append( \nGoldenfile length: );
+sb.append( fromGoldenFile.length );
+System.out.println( sb.toString() );
+sb = null;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+
+} else {
+
+int i = 0;
+int j = 0;
+
+while ( ( i fromServer.length ) ( j fromGoldenFile.length ) ) {
+if ( fromServer[ i ] != fromGoldenFile[ j ] ) {
+System.out.println( Error at position + ( i + 1 ) );
-// Compare the actual result and the expected result.
-private boolean compare(String str1, String str2) {
- //System.out.println(In compare);
- if ( str1==null || str2==null) return false;
- if ( str1.length() != str2.length() ) {
- System.out.println(Wrong size + str1.length() + + str2.length() );
- return false;
- }
-
-for(int i=0; istr1.length() ; i++ ) {
-if (str1.charAt( i ) != str2.charAt( i ) ) {
- System.out.println(Error at + i + + str1.charAt(1) +
-str2.charAt(i));
-return false;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+}
+
+i++;
+j++;
+}
+}
+
+return true;
+}
+
+/*
+ * codedumpHex/code helper method to dump formatted
+ * hex output of the server response and the goldenfile.
+ *
+ * @param serverResponse a codebyte[]/code value
+ * @param goldenFile a codebyte[]/code value
+ */
+private void dumpHex( byte[] serverResponse, byte[] goldenFile ) {
+StringBuffer outBuf = new StringBuffer( ( serverResponse.length +
goldenFile.length ) * 2 );
+
+String fromServerString = getHexValue( serverResponse, 0,
serverResponse.length );
+String fromGoldenFileString
[PATCH] Tomcat 4.0.1- Proposed fix for Bugzilla 4609
Hi,
Thought I'd give a shot at patching a bug (4609) I logged.
The basics of the bug is that an IOException is not thrown
if out.close() is called from within a JSP page and
subsequent calls to write() or println(), etc. are made.
The solution I have affects two classes:
org.apache.jasper.runtime.JspWriterImpl
org.apache.jasper.compiler.JspParseEventListener
The modification to JspWriterImpl was simple:
-Added a new boolean instance variable called 'closed'
-When JspWriter.close() is called, set 'closed' to true.
-Modified the JspWriter.ensureOpen() method. If
'closed' is true, or response is null, throw the IOException.
I felt the modification to JspParseEventListener.generateFooter()
was necessary as the page code generated:
snip
} catch (Throwable t) {
if (out != null out.getBufferSize() != 0)
out.clearBuffer();
if (pageContext != null) pageContext.handlePageException(t);
}
/snip
So in this case, now that the IOException is thown, we go into the
throwable, but the call to out.clearBuffer() generates its own
IOException. The problem with this is that the stacktrace becomes
inaccurate. Showing the call to clearBuffer() as the top-most call
on the stack.
So the change I introduced would generate the following code:
snip
if (out != null out.getBufferSize() != 0)
try {
out.clearBuffer();
} catch (java.io.IOException ioe) {
if (t instanceof java.io.IOException) {
if (!(t.getMessage().equals(ioe.getMessage( {
t = ioe;
}
}
}
/snip
So, here if out.clearBuffer() happens to throw an IOException,
check to see if the throwable that brought us to this point
in the code is an IOException as well. If it is, see if
the messages are the same, if they aren't, then there
is a new IO issue and set ioe to the throwable which will
be handled by pageContext.handlePageException. Otherwise
the catch becomes a no-op and pageContext.handlePageException
will use the original throwable. Hope that made sense.
Anyway, comments are definately welcome. Still getting familiar
with the code.
Thanks,
-rl
Index: JspParseEventListener.java
===
RCS file:
/home/cvspublic/jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/compiler/JspParseEventListener.java,v
retrieving revision 1.35
diff -u -r1.35 JspParseEventListener.java
--- JspParseEventListener.java 2001/11/02 19:36:09 1.35
+++ JspParseEventListener.java 2001/11/20 22:29:40
@@ -378,7 +378,23 @@
writer.pushIndent();
writer.println(if (out != null out.getBufferSize() != 0));
writer.pushIndent();
+writer.println( try { );
+writer.pushIndent();
writer.println(out.clearBuffer(););
+writer.popIndent();
+writer.println(} catch (java.io.IOException ioe) {);
+writer.pushIndent();
+writer.println(if (t instanceof java.io.IOException) {);
+writer.pushIndent();
+writer.println(if (!(t.getMessage().equals(ioe.getMessage( {);
+writer.pushIndent();
+writer.println(t = ioe;);
+writer.popIndent();
+writer.println(});
+writer.popIndent();
+writer.println(});
+writer.popIndent();
+writer.println(});
writer.popIndent();
writer.println(if (pageContext != null) pageContext.handlePageException(t););
writer.popIndent();
Index: JspWriterImpl.java
===
RCS file:
/home/cvspublic/jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/runtime/JspWriterImpl.java,v
retrieving revision 1.1
diff -u -r1.1 JspWriterImpl.java
--- JspWriterImpl.java 2000/08/12 00:52:12 1.1
+++ JspWriterImpl.java 2001/11/20 18:10:11
@@ -99,6 +99,7 @@
protected static int defaultCharBufferSize = Constants.DEFAULT_BUFFER_SIZE;
protected boolean flushed = false;
+protected boolean closed = false;
public JspWriterImpl() {
super( defaultCharBufferSize, true );
@@ -223,6 +224,7 @@
if (out != null)
out.close();
out = null;
+closed = true;
//cb = null;
}
}
@@ -236,7 +238,7 @@
/** check to make sure that the stream has not been closed */
protected void ensureOpen() throws IOException {
- if (response == null)
+ if ( closed || response == null)
throw new IOException(Stream closed);
}
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog-4.0 - False positive from GetRealPathTestServlet
Simple patch, but
server/servlet-tests/WEB-INF/classes/tests/javax_servlet/ServletContext/GetRealPathTestServlet.java
Yeilds a false positive when the test is run.
Attached is a simple patch to rectify.
More to follow later
Index: GetRealPathTestServlet.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/server/servlet-tests/WEB-INF/classes/tests/javax_servlet/ServletContext/GetRealPathTestServlet.java,v
retrieving revision 1.1
diff -r1.1 GetRealPathTestServlet.java
90c90
if(path!=null) {
---
if(realPath!=null) {
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog-4.0 -- tag_int.jsp tests functionality notmentioned in detail in spec
Hi,
The following test:
server/jsp-tests/jsp/tagext/LifeCycle/tag_int.jsp
Is testing functionality not really explained clearly in the
Specification.
Test code:
%--
Test for defining an int variable.
--%
%@ taglib uri=/TestLib.tld prefix=x %
%! public static Integer increment(Integer i) {
if (i != null)
return new Integer(i.intValue()+1);
return new Integer(0);
}
%
%! public static int valueof(Integer i) {
if (i != null)
return i.intValue();
return 0;
}
%
x:define id=i scope=page life=nested
(1) i was %= i %; % i = increment(i); % i is now %= valueof(i) %
/x:define
x:define id=i scope=page life=nested
(2) i was %= i %; % i = increment(i); % i is now %= valueof(i) %
/x:define
x:define id=i scope=page life=at_begin
(3) i was %= i %; % i = increment(i); % i is now %= valueof(i) %
/x:define
(4) i was %= i %; % i = increment(i); % i is now %= valueof(i) %
**
GOLDENFILE OUTPUT:
(1) i was null; i is now 0
(2) i was 0; i is now 1
(3) i was 0; i is now 1
(4) i was 0; i is now 1
**
Issue:
In (3) since the life of the variable is described as at_begin the
value
should also be available via the PageContext.The value of i is not
synchronized
with the pageContext after the end of the define tag.
Hence in (4) i starts afresh with 0.
Recommendation:
The attached patch excludes the tag_int tests from running until a
clarification is made against the specification.
Index: jsp-gtest.xml
===
RCS file: /home/cvspublic/jakarta-watchdog-4.0/src/conf/jsp-gtest.xml,v
retrieving revision 1.9
diff -u -r1.9 jsp-gtest.xml
--- jsp-gtest.xml 2001/08/09 17:35:13 1.9
+++ jsp-gtest.xml 2001/11/19 19:53:47
@@ -1585,6 +1585,7 @@
/
--
+!--
gtest request=GET /jsp-tests/jsp/tagext/LifeCycle/tag_int.jsp HTTP/1.0
debug=0 host=${host} port=${port}
goldenFile=${wgdir}/tagext/LifeCycle/tag_int.html
@@ -1592,6 +1593,7 @@
assertion=Test for defining an int variable, specified in the Java Server
Pages Specification v1.2, Sec 6.4.7
testStrategy=testing LifeCycle process
/
+--
!--
gtest request=GET /jsp-tests/jsp/tagext/LifeCycle/tag_nonempty_body_1.jsp
HTTP/1.0
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Watchdog-4.0 - GTest patch for whitspace comparison.
Hi,
After reading through GTest, and the JSP/Servlet-Gtest.xml, none of the
tests set the exactMatch property to true to do whitespace comparison
between the response from the server, and the contents of the
goldenfile. This patch, will make GTest do a byte for byte comparison
of the response and goldenfile. If any length or byte differeces
appear, it will dump a hex representation (ascii on the right side, hex
on the left) to help the test writer figure out what's wrong. Make's
it easier in my opinion anyway :).
I plan to start going through and correcting the whitespace issues as
I come across them, so there will probably be more patches to come down
the pipe.
Anyway, attached is the patch to GTest (GTest.patch.txt) as well as
sample output (sample.txt) of what this code produces when exactMatch is
set to true.
Anyone care to comment?
Thanks,
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java 2001/09/28 04:09:56 1.3
+++ GTest.java 2001/11/19 21:36:00
@@ -15,6 +15,11 @@
// derived from Jsp
public class GTest extends Task {
+
+private static final String ZEROS = ;
+private static final int SHORTPADSIZE = 4;
+private static final int BYTEPADSIZE = 2;
+
String prefix=http://localhost:8080/test;;
String host=localhost;
int port=8080;
@@ -376,9 +381,9 @@
boolean cmp=true;
if(exactMatch)
- cmp=compare(responseBody, expResult.toString() );
+ cmp=compare(responseBody.getBytes(), expResult.toString().getBytes() );
else
- cmp=compareWeek( responseBody, expResult.toString());
+ cmp=compareWeak( responseBody, expResult.toString());
if( cmp != testCondition ) {
responseStatus = false;
@@ -521,30 +526,200 @@
}
}
+/*
+ * codecompare/code compares the two byte arrays passed
+ * in to verify that the lengths of the arrays are equal, and
+ * that the content of the two arrays, byte for byte are equal.
+ *
+ * @param fromServer a codebyte[]/code value
+ * @param fromGoldenFile a codebyte[]/code value
+ * @return codeboolean/code true if equal, otherwise false
+ */
+private boolean compare( byte[] fromServer, byte[] fromGoldenFile ) {
+if ( fromServer == null || fromGoldenFile == null ) {
+return false;
+}
+
+/*
+ * Check to see that the respose and golden file lengths
+ * are equal. If they are not, dump the hex and don't
+ * bother comparing the bytes. If they are equal,
+ * iterate through the byte arrays and compare each byte.
+ * If the bytes don't match, dump the hex representation
+ * of the server response and the goldenfile and return
+ * false.
+ */
+
+if ( fromServer.length != fromGoldenFile.length ) {
+StringBuffer sb = new StringBuffer( 50 );
+sb.append( Response and golden file lengths do not match!\n );
+sb.append( Server response length: );
+sb.append( fromServer.length );
+sb.append( \nGoldenfile length: );
+sb.append( fromGoldenFile.length );
+System.out.println( sb.toString() );
+sb = null;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+
+} else {
+
+int i = 0;
+int j = 0;
+
+while ( ( i fromServer.length ) ( j fromGoldenFile.length ) ) {
+if ( fromServer[ i ] != fromGoldenFile[ j ] ) {
+System.out.println( Error at position + ( i + 1 ) );
-// Compare the actual result and the expected result.
-private boolean compare(String str1, String str2) {
- //System.out.println(In compare);
- if ( str1==null || str2==null) return false;
- if ( str1.length() != str2.length() ) {
- System.out.println(Wrong size + str1.length() + + str2.length() );
- return false;
- }
-
-for(int i=0; istr1.length() ; i++ ) {
-if (str1.charAt( i ) != str2.charAt( i ) ) {
- System.out.println(Error at + i + + str1.charAt(1) +
- str2.charAt(i));
-return false;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+}
+
+i++;
+j++;
+}
+}
+
+return true;
+}
+
+/*
+ * codedumpHex/code helper method to dump formatted
+ * hex output of the server response and the goldenfile.
+ *
+ * @param
Re: [PATCH] Watchdog-4.0 - GTest patch for whitspace comparison.
Forgot to mention, I changed the method name : compareWeek, to
compareWeak.
On Mon, 2001-11-19 at 17:04, Ryan Lubke wrote:
Hi,
After reading through GTest, and the JSP/Servlet-Gtest.xml, none of the
tests set the exactMatch property to true to do whitespace comparison
between the response from the server, and the contents of the
goldenfile. This patch, will make GTest do a byte for byte comparison
of the response and goldenfile. If any length or byte differeces
appear, it will dump a hex representation (ascii on the right side, hex
on the left) to help the test writer figure out what's wrong. Make's
it easier in my opinion anyway :).
I plan to start going through and correcting the whitespace issues as
I come across them, so there will probably be more patches to come down
the pipe.
Anyway, attached is the patch to GTest (GTest.patch.txt) as well as
sample output (sample.txt) of what this code produces when exactMatch is
set to true.
Anyone care to comment?
Thanks,
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java2001/09/28 04:09:56 1.3
+++ GTest.java2001/11/19 21:36:00
@@ -15,6 +15,11 @@
// derived from Jsp
public class GTest extends Task {
+
+private static final String ZEROS = ;
+private static final int SHORTPADSIZE = 4;
+private static final int BYTEPADSIZE = 2;
+
String prefix=http://localhost:8080/test;;
String host=localhost;
int port=8080;
@@ -376,9 +381,9 @@
boolean cmp=true;
if(exactMatch)
- cmp=compare(responseBody, expResult.toString() );
+ cmp=compare(responseBody.getBytes(), expResult.toString().getBytes() );
else
- cmp=compareWeek( responseBody, expResult.toString());
+ cmp=compareWeak( responseBody, expResult.toString());
if( cmp != testCondition ) {
responseStatus = false;
@@ -521,30 +526,200 @@
}
}
+/*
+ * codecompare/code compares the two byte arrays passed
+ * in to verify that the lengths of the arrays are equal, and
+ * that the content of the two arrays, byte for byte are equal.
+ *
+ * @param fromServer a codebyte[]/code value
+ * @param fromGoldenFile a codebyte[]/code value
+ * @return codeboolean/code true if equal, otherwise false
+ */
+private boolean compare( byte[] fromServer, byte[] fromGoldenFile ) {
+if ( fromServer == null || fromGoldenFile == null ) {
+return false;
+}
+
+/*
+ * Check to see that the respose and golden file lengths
+ * are equal. If they are not, dump the hex and don't
+ * bother comparing the bytes. If they are equal,
+ * iterate through the byte arrays and compare each byte.
+ * If the bytes don't match, dump the hex representation
+ * of the server response and the goldenfile and return
+ * false.
+ */
+
+if ( fromServer.length != fromGoldenFile.length ) {
+StringBuffer sb = new StringBuffer( 50 );
+sb.append( Response and golden file lengths do not match!\n );
+sb.append( Server response length: );
+sb.append( fromServer.length );
+sb.append( \nGoldenfile length: );
+sb.append( fromGoldenFile.length );
+System.out.println( sb.toString() );
+sb = null;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+
+} else {
+
+int i = 0;
+int j = 0;
+
+while ( ( i fromServer.length ) ( j fromGoldenFile.length ) ) {
+if ( fromServer[ i ] != fromGoldenFile[ j ] ) {
+System.out.println( Error at position + ( i + 1 ) );
-// Compare the actual result and the expected result.
-private boolean compare(String str1, String str2) {
- //System.out.println(In compare);
- if ( str1==null || str2==null) return false;
- if ( str1.length() != str2.length() ) {
- System.out.println(Wrong size + str1.length() + + str2.length() );
- return false;
- }
-
-for(int i=0; istr1.length() ; i++ ) {
-if (str1.charAt( i ) != str2.charAt( i ) ) {
- System.out.println(Error at + i + + str1.charAt(1) +
-str2.charAt(i));
-return false;
+// dump the hex representation of the byte arrays
+dumpHex( fromServer, fromGoldenFile );
+
+return false;
+}
+
+i
[PATCH] - Watchdog 4.0 - 3 New HttpServletResponse tests + Patchesto support new tests.
Hi,
I've added 3 tests to javax_servlet_http/HttpServletResponse to
validate that headers set after sendError() or sendRedirect are
called are ignored by the container (this was an issue at one point).
The 3 new tests are:
SendErrorIgnoreHeaderTestServlet.java
SendError_StringIgnoreHeaderTestServlet.java
SendRedirectIgnoreHeaderTestServlet.java
These should be placed in
${WATCHDOG{/src/server/servlet-tests/WEB-INF/classes/tests/javax_servlet_http/HttpServletResponse
To support these tests, I've added a new method to GTest,
setUnexpectedHeaders (set in the servlet or jsp gtest.xml like so:
unexpectedHeaders=key:value ).
This will setup a hashtable of headers to compare against the server
response. The logic to checkResponse has been updated accordingly.
More to come.
Comments welcome.
-rl
Index: GTest.java
===
RCS file:
/home/cvspublic/jakarta-watchdog-4.0/src/tools/org/apache/tomcat/task/GTest.java,v
retrieving revision 1.3
diff -u -r1.3 GTest.java
--- GTest.java 2001/09/28 04:09:56 1.3
+++ GTest.java 2001/11/19 23:54:31
@@ -35,6 +35,10 @@
String responseMatch;
// the response should include the following headers
Hashtable expectHeaders;
+
+// the response should not include the following headers
+Hashtable unexpectedHeaders;
+
// Match request line
String returnCode=;
@@ -197,10 +201,21 @@
//Ramesh
public void setExpectHeaders( String s ) {
- this.expectHeaders=new Hashtable();
- getHeaderDetails( s, expectHeaders );
+this.expectHeaders=new Hashtable();
+getHeaderDetails( s, expectHeaders );
- //parseHeader( s, expectHeaders );
+//parseHeader( s, expectHeaders );
+}
+
+ /**
+ * codesetUnexpectedHeaders/code sets an ArrayList
+ * with headers that are not expected to be found in a response
+ *
+ * @param s a codeString/code value
+ */
+public void setUnexpectedHeaders( String s ) {
+this.unexpectedHeaders = new Hashtable();
+getHeaderDetails( s, unexpectedHeaders );
}
public void setResponseMatch( String s ) {
@@ -329,30 +344,46 @@
if( expectHeaders != null ) {
// Check if we got the expected headers
if(headers==null) {
- System.out.println(ERROR no response header, expecting header);
+System.out.println(ERROR no response header, expecting header);
}
Enumeration e=expectHeaders.keys();
while( e.hasMoreElements()) {
- String key=(String)e.nextElement();
- String value=(String)expectHeaders.get(key);
- String respValue=(String)headers.get(key);
- if( respValue==null || respValue.indexOf( value ) 0 ) {
- System.out.println(ERROR expecting header + key + : +
- value + GOT: + respValue+ HEADERS( +
headers + ));
- if ( resultOut != null )
- {
- expectedString = expectedHeader + key + :+ value
+ /expectedHeader\n;
- actualString = actualHeader+ key + : + respValue
+ /actualHeader\n;
- resultOut.write(expectedString.getBytes() );
- resultOut.write(actualString.getBytes() );
+String key=(String)e.nextElement();
+String value=(String)expectHeaders.get(key);
+String respValue=(String)headers.get(key);
+if( respValue==null || respValue.indexOf( value ) 0 ) {
+System.out.println(ERROR expecting header + key + : +
+ value + GOT: + respValue+ HEADERS( +
+headers + ));
+if ( resultOut != null )
+{
+expectedString = expectedHeader + key + :+ value +
+/expectedHeader\n;
+actualString = actualHeader+ key + : + respValue +
+/actualHeader\n;
+resultOut.write(expectedString.getBytes() );
+resultOut.write(actualString.getBytes() );
- }
+}
- return false;
- }
+return false;
+}
}
-
}
+
+// check to see if any unexpected headers we're recieved
+if ( unexpectedHeaders != null ) {
+if ( headers != null ) {
+Enumeration e = unexpectedHeaders.keys();
+while( e.hasMoreElements() ) {
+String key = (String) e.nextElement();
+String value = (String) unexpectedHeaders.get( key );
+String respValue = (String) headers.get( key );
+if ( respValue != null respValue.indexOf( value ) 0 ) {
+System.out.println( ERROR Unexpected header: + key + : +
+value +
+
Re: Performance of tomcat 3.2.3
http://jakarta.apache.org/jmeter/index.html On Fri, 2001-11-16 at 12:16, Michael Jennings wrote: Hi everyone, I'm working on a project for a client that involves a servlet engine component. I suggested tomcat since I am most familiar with it and it seems the most compliant servlet engine around. My client has mentioned concerns about tomcat's performance and now is insisting on using IBM websphere 4.0 My suspicion is that tomcat's performance will be fine for this application, but I would like to have some hard numbers to back up my claim. Does anyone know of any kind of tool or code out there that can measure http and https performance? Thanks in advance. -Mike Jennings -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Status-Code not set for Error pages when Exception is thrown
Hi, Say a user has a custom error page for status codes of 500 when an uncaught exception occurs. The request parameter, javax.servlet.error.status_code, is not set, so if the error page tries to access this req. parameter, null will be returned. Attached is a patch that I've tested in my env and seems to do the trick. -rl Index: ErrorDispatcherValve.java === RCS file: /home/cvspublic/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorDispatcherValve.java,v retrieving revision 1.6 diff -u -r1.6 ErrorDispatcherValve.java --- ErrorDispatcherValve.java 2001/11/13 00:42:24 1.6 +++ ErrorDispatcherValve.java 2001/11/16 23:56:55 @@ -229,6 +229,8 @@ response.setAppCommitted(false); ServletRequest sreq = request.getRequest(); ServletResponse sresp = response.getResponse(); +sreq.setAttribute(Globals.STATUS_CODE_ATTR, + new +Integer(HttpServletResponse.SC_INTERNAL_SERVER_ERROR)); sreq.setAttribute(Globals.ERROR_MESSAGE_ATTR, throwable.getMessage()); sreq.setAttribute(Globals.EXCEPTION_ATTR, -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: [VOTE] New commiter Ryan Bloom
On Monday 10 September 2001 10:23, Christopher Cain wrote: Thanks for the vote of confidence everybody. I wasn't expecting such a fast turn-around time on this. I'll try not to muck things up too badly. Ryan +1 Welcome, Slick! - Christopher /** * Pleurez, pleurez, mes yeux, et fondez vous en eau! * La moitié de ma vie a mis l'autre au tombeau. *---Corneille */ GOMEZ Henri wrote: I would like to propose Ryan Bloom as commiter in Tomcat, and particulary on jakarta-tomcat-connector. Ryan is one of the dev leader in Apache 2.0 and contributed many patch for both mod_jk and mod_webapp, showing us that connectors avoid politics :) Vote, please - Henri Gomez ___[_] EMAIL : [EMAIL PROTECTED](. .) PGP KEY : 697ECEDD...oOOo..(_)..oOOo... PGP Fingerprint : 9DF8 1EA8 ED53 2F39 DC9B 904A 364F 80E6 -- __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: [VOTE] Removal of mod_jk for Apache 2.0 from jakarta-tomcat f or Tomcat 3.3
On Monday 10 September 2001 14:05, Pier Fumagalli wrote: [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: On Mon, 10 Sep 2001, Pier Fumagalli wrote: GOMEZ Henri [EMAIL PROTECTED] wrote: Ryan to became more than just a contributer : This is the third time we agree on something in less than 24 hours. This implies that either I'm getting old, or just plain silly... Now, if you could agree on merging mod_webapp and mod_jk, that would be something... Slowww down... :) If mod_jk wants to start using APR, I believe we're talking, otherwise, I'm done with cross-platform porting, I live it to Ryan Oh no you don't. I did the cross-platform stuff. I wrote APR to get awar from it. Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: [VOTE] Removal of mod_jk for Apache 2.0 from jakarta-tomcat f or Tomcat 3.3
On Monday 10 September 2001 14:51, [EMAIL PROTECTED] wrote: On Mon, 10 Sep 2001, Pier Fumagalli wrote: This is the third time we agree on something in less than 24 hours. This implies that either I'm getting old, or just plain silly... Now, if you could agree on merging mod_webapp and mod_jk, that would be something... Slowww down... :) If mod_jk wants to start using APR, I believe we're talking, otherwise, I'm done with cross-platform porting, I live it to Ryan :) Mod_jk will use APR - that's certain. The only question is when and how to do the transition without affecting the stability of the code. Having an APR1.0 out is one of the requirements - I don't think we can release mod_jk, even from j-t-c, with dependencies on un-released library. There are already 2 proposals for how to do that - one with preserving the current common as a temporary solution, until we make sure it works with IIS/NES, and the other with removing the common utils and hoping things will work with IIS/NES. Right now the APR/common is not the main itch - it'll become pretty soon, at least for me ( I need mmap for the new connector ) I'm actually right now working on the thread locks for Windows, and then I am going to start agitating for an APR release. We should have APR 1.0 out the door soon-ish. I am hoping to have it released sometime in the next month or two. :-) Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: [VOTE] Removal of mod_jk for Apache 2.0 from jakarta-tomcat f or Tomcat 3.3
On Monday 10 September 2001 15:22, Pier Fumagalli wrote: Ryan Bloom [EMAIL PROTECTED] wrote: MMAP is the other scary stuff in APR, the new code (without Ralph's libmm) it no more than one month old... I need it for load balancing, but I want to double check with the guys in CA next week and see what they tell me before publishing anything.. Actually, MMAP has been in APR for a long time, it is just shared memory that is new. My bad... :) And, the shared memory code has been stressed in Apache for the last month. Also the shared memory code is basically just the important stuff from the MM library. My last status was 2 weeks ago when I last saw David, and he said wait for a little longer still... So, can I assume that little longer is over? I think so. The problem back then, was that he needed APR to be available before he could do anything with apr-util, but we were cleaning APR before trying to clean apr-util, and it just didn't work. That was a BeOS-only problem though. Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: [VOTE] Removal of mod_jk for Apache 2.0 from jakarta-tomcat f or Tomcat 3.3
On Monday 10 September 2001 16:15, Christopher Cain wrote: Gomez Henri wrote: [snip] PS: Something goes crasy these days, on tomcat list, what do you think about this Pier (known as my worst enemy :) Something is indeed a little bizarre on the list today, mon ami. Maybe because Craig isn't here to keep us in line =) a) There are now four key people seriously discussing a partial merge of mod_jk and mod_webapp b) Henri and Pier are agreeing on all sorts of things c) The 3.3 guys are congratulating the 4.0 guys and vice-versa d) There have been about 15 votes in last 24 hours e) _I'm_ the short-tempered one, for a change The only reason I know for sure that the world is still spinning is that Jon is still vetoing things ;-) Well, I'm new to the list, but I like to veto things too. Somebody point me at something I can veto... :-) Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: cvs commit: jakarta-tomcat-connectors/webapp/apache-2.0 mod_w ebapp.c
On Thursday 06 September 2001 18:06, Justin Erenkrantz wrote: On Fri, Sep 07, 2001 at 01:52:50AM +0100, Pier Fumagalli wrote: Since there is no official release of httpd-2.0, I am confident that nobody will actually care unless they're testing something... And by testing good practice says get the HEAD of CVS and try it out yourself if it doesn't work. Development happens on head and if someone forks off for a little while to do some porting to some weird operating systems, at the end those changes will have to be integrated with HEAD, so, there's no point in sticking with backward compatibility AT THIS POINT. +1 to what Pier said. (Ryan may chip in here too...) If you are using anything other than HEAD for httpd-2.0, I don't care much about any bug reports you may have. The odds are that it'd either be fixed or broken in a different manner. =) -- justin Don't try to mainain backwards compat in 2.0. We aren't even trying yet. Every release of 2.0 other than head has a serious problem, either a memory leek, or a security hole, or it can crash the machine. IF your not using HEAD of httpd-2.0, you have a problem. :-) Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
[PATCH] only parse workers.properties once
This is an improvement for mod_jk, to only parse workers.properties once
on platforms that use fork(). Most platforms will not require us to read
the workers.properties file in the child process, because they process it in
the parent. Windows is the only platform I know of today that does require
us to read it twice.
This also has the advantage, that the workers.properties file does not need
to be readable by the user that the web server is running as.
Ryan
Index: mod_jk.c
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/jk/native/apache-2.0/mod_jk.c,v
retrieving revision 1.23
diff -u -d -b -w -u -r1.23 mod_jk.c
--- mod_jk.c2001/09/06 17:00:54 1.23
+++ mod_jk.c2001/09/06 21:54:01
@@ -1214,6 +1214,7 @@
return overrides;
}
+#ifdef WIN32
/** Standard apache callback, initialize jk.
*/
static void jk_child_init(apr_pool_t *pconf,
@@ -1224,6 +1225,7 @@
init_jk( pconf, conf, s );
}
+#endif
/** Initialize jk, using worker.properties.
We also use apache commands ( JkWorker, etc), but this use is
@@ -1340,7 +1342,9 @@
{
ap_hook_handler(jk_handler, NULL, NULL, APR_HOOK_MIDDLE);
ap_hook_post_config(jk_post_config,NULL,NULL,APR_HOOK_MIDDLE);
+#ifdef WIN32
ap_hook_child_init(jk_child_init,NULL,NULL,APR_HOOK_MIDDLE);
+#endif
ap_hook_translate_name(jk_translate,NULL,NULL,APR_HOOK_FIRST);
#if (MODULE_MAGIC_NUMBER_MAJOR 20010808)
ap_hook_map_to_storage(jk_map_to_storage, NULL, NULL, APR_HOOK_MIDDLE);
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
Re: Build system for mod_webapp and Apache 2.0
On Wednesday 05 September 2001 04:13, jean-frederic clere wrote: If you don't remove the --disable-shared, then libtool 1.4 can't compile mod_webapp as a dynamic library. It really sucks, but complain to the libtool developers. That flg must go, or we break anybody who tries to use the latest version of libtool. :-( Ryan Hi, I have committed it. Note that the --disable-shared MUST NOT be removed otherwise libtool uses the dynamic libraries and we want it to use the static ones. Cheers Jean-frederic __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: Build system for mod_webapp and Apache 2.0
On Wednesday 05 September 2001 09:57, jean-frederic clere wrote: Ryan Bloom wrote: On Wednesday 05 September 2001 04:13, jean-frederic clere wrote: If you don't remove the --disable-shared, then libtool 1.4 can't compile mod_webapp as a dynamic library. I just want a DSO file... - On my Linux if I remove the --disable-shared to the APR, libtool complains when building mod_webapp.so: something like that we use dynamic library we have to install them... (That is libtool 1.3.5) Are you suggesting to use dynamic libapr and libwebapp? I don't really have a good suggestion. I am just telling you that with --disable-shared, libtool 1.4 CAN NOT create DSOs. It sucks, but that is libtool for ya. It sounds like we may have to figure out which version of libtool the user has, and adjust our build rules appropriately. :-( Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Patch for mod_jk and Apache 2.0
This patch allows mod_jk to successfully serve pages against Apache 2.0
and Tomcat 4.0. This is required after the changes to abstract out location
and directory walks from the core request processing. This was created by
Will Rowe and myself.
Ryan
Index: mod_jk.c
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/jk/native/apache-2.0/mod_jk.c,v
retrieving revision 1.18
diff -u -d -b -w -u -r1.18 mod_jk.c
--- mod_jk.c2001/09/04 10:57:03 1.18
+++ mod_jk.c2001/09/06 00:17:28
@@ -1333,13 +1333,22 @@
return DECLINED;
}
+static int jk_map_to_storage(request_rec *r)
+{
+if (apr_table_get(r-notes, JK_WORKER_ID)) {
+r-filename = apr_filename_of_pathname(r-uri);
+return OK;
+}
+return DECLINED;
+}
+
static void jk_register_hooks(apr_pool_t *p)
{
ap_hook_handler(jk_handler, NULL, NULL, APR_HOOK_MIDDLE);
ap_hook_post_config(jk_post_config,NULL,NULL,APR_HOOK_MIDDLE);
ap_hook_child_init(jk_child_init,NULL,NULL,APR_HOOK_MIDDLE);
ap_hook_translate_name(jk_translate,NULL,NULL,APR_HOOK_FIRST);
-
+ap_hook_map_to_storage(jk_map_to_storage, NULL, NULL, APR_HOOK_MIDDLE);
}
module AP_MODULE_DECLARE_DATA jk_module =
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
Re: cvs commit: jakarta-tomcat-connectors/webapp/apache-2.0 Makefile.in
On Friday 31 August 2001 10:32, Pier Fumagalli wrote: [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: jfclere 01/08/29 07:29:29 Modified:webapp configure.in Added: webapp/apache-2.0 Makefile.in Log: Copy/add configure logic for Apache-2.0 from mod_jk. Revision ChangesPath 1.23 +12 -3 jakarta-tomcat-connectors/webapp/configure.in APXS is broken in Apache 2.0... We should rely on APRVARS... I actually fixed APXS this last weekend, and I have used it many times now to build mod_webapp. :-) Ryan __ Ryan Bloom [EMAIL PROTECTED] Covalent Technologies [EMAIL PROTECTED] --
Re: Build system for mod_webapp and Apache 2.0
On Friday 31 August 2001 10:36, Pier Fumagalli wrote:
+TARGET=${SRCDIR}/apache-2.0
+makefile=apache-2.0/Makefile
+AC_MSG_RESULT([setting target module to... ${TARGET}])
+ ]
+)
+AC_SUBST(TARGET)
+
+dnl
Wasn't APXS for 2.0 completely broken? That's what you said last time :)
It was. I fixed it. :)
LOCAL_HEADER([Configuring APR])
LOCAL_FILTEREXEC(
- [./configure --enable-static --disable-shared --disable-threads],
+ [./configure --enable-static --disable-threads],
[APR configure])
if ${TEST} ${ret} -ne 0
then
Hmm... I'd like to rely on the APR code coming with Apache 2.0 instead of
building our local copy with Apache 2.0... (I'll see what I can figure out,
hoping that 2.0 works on my OS/X box :)
Oh, if that's what you want to do, I can make that happen. Give me five minutes,
and you'll have a patch. :-)
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
Re: Build system for mod_webapp and Apache 2.0
LOCAL_HEADER([Configuring APR])
LOCAL_FILTEREXEC(
- [./configure --enable-static --disable-shared --disable-threads],
+ [./configure --enable-static --disable-threads],
[APR configure])
if ${TEST} ${ret} -ne 0
then
Hmm... I'd like to rely on the APR code coming with Apache 2.0 instead of
building our local copy with Apache 2.0... (I'll see what I can figure
out, hoping that 2.0 works on my OS/X box :)
Oh, if that's what you want to do, I can make that happen. Give me five
minutes, and you'll have a patch. :-)
Sorry, I got called into a meeting, so this took a little bit longer than I wanted it
to. Here is a patch that stops us from building APR, and we just use the one
from Apache 2.0.
Index: Makefile.in
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/webapp/Makefile.in,v
retrieving revision 1.16
diff -u -d -b -w -u -r1.16 Makefile.in
--- Makefile.in 2001/08/31 10:26:24 1.16
+++ Makefile.in 2001/08/31 20:36:29
@@ -70,11 +70,11 @@
config.log \
config.status
-all: apr-all local-all
+all: @APR_BUILD@ local-all
install: local-install
-clean: apr-clean local-clean
+clean: @APR_CLEAN@ local-clean
distclean: clean
@$(ECHO)
Index: configure.in
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/webapp/configure.in,v
retrieving revision 1.24
diff -u -d -b -w -u -r1.24 configure.in
--- configure.in2001/08/31 10:26:24 1.24
+++ configure.in2001/08/31 20:36:29
@@ -210,15 +210,21 @@
APA=`${GREP} STANDARD20 ${APXS}`
if ${TEST} -z $APA ; then
TARGET=${SRCDIR}/apache-1.3
+ APR_BUILD=apr-all
+ APR_CLEAN=apr-clean
makefile=apache-1.3/Makefile
else
TARGET=${SRCDIR}/apache-2.0
+ APR_BUILD=
+ APR_CLEAN=
makefile=apache-2.0/Makefile
fi
AC_MSG_RESULT([setting target module to... ${TARGET}])
]
)
AC_SUBST(TARGET)
+AC_SUBST(APR_BUILD)
+AC_SUBST(APR_CLEAN)
dnl --
dnl Process the --with-apr=... command line argument
@@ -226,10 +232,16 @@
AC_MSG_CHECKING([APR directory])
AC_ARG_WITH(apr,
[ --with-apr=DIR where the APR sources can be found.],
- APRDIR=${withval},
- APRDIR=${SRCDIR}/apr
+ APRDIR=${withval}
)
+if ${TEST} -z ${APA} -a -z ${APRDIR} ; then
+ APRDIR={SRCDIR}/apr
+else
+ APRDIR=${APXS_PREFIX}
+fi
+
+if ${TEST} -z ${APA} ; then
if ${TEST} ! -d ${APRDIR} ; then
AC_MSG_ERROR([Cannot find APR sources directory \${APRDIR}\])
fi
@@ -267,6 +279,9 @@
AC_MSG_ERROR([APR configure script terminated with error code ${ret}])
fi
cd ${curdir}
+else
+ AC_MSG_RESULT(Using APR from Apache build)
+fi
dnl --
dnl Retrieving APR default values
@@ -275,19 +290,29 @@
if ${TEST} ! -f ${APRDIR}/APRVARS
then
+ if ${TEST} ${APA} ; then
+LIBDIR=/lib
+ else
AC_MSG_RESULT([error])
AC_MSG_ERROR([Cannot find APR defaults in ${APRDIR}/APRVARS])
fi
+else
+ LIBDIR=
+fi
if ${TEST} ! -x ${APRDIR}/libtool
then
+ if ${TEST} ${APA} ; then
+LIBTOOL=${APXS_PREFIX}/build/libtool
+ else
AC_MSG_RESULT([error])
AC_MSG_ERROR([Cannot find APR libtool binary])
+ fi
else
LIBTOOL=${APRDIR}/libtool
fi
-eval `cat ${APRDIR}/APRVARS`
+eval `cat ${APRDIR}${LIBDIR}/APRVARS`
AC_SUBST(CC)
AC_SUBST(CPP)
AC_SUBST(SHELL)
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
Re: Build system for mod_webapp and Apache 2.0
On Thursday 30 August 2001 01:10, jean-frederic clere wrote:
Ryan Bloom wrote:
Hi everybody, I'm new to the list, but for anybody who doesn't know me,
I have been pretty active on httpd 2.0. I decided to port mod_webapp
last week, and Pier committed the code for me, but the build system isn't
there, so I am posting the build system now.
This is currently working, although there seems to be some strangeness
when Tomcat responds to mod_webapp. I'll be looking into that over the
next few days. I hope to be more active as time permits. :-)
The Makefile.in should be placed in
jakarta-tomcat-connectors/webapp/apache-2.0
Hi Ryan,
I committed some code some hours before... I will try to merge the two.
I am detecting Apache version using apxs thru the following code:
+++
dnl test apache version (from mod_jk)
$RM -rf test
$APXS -n test -g
APA=`grep STANDARD20 test/mod_test.c`
if ${TEST} -z $APA ; then
TARGET=${SRCDIR}/apache-1.3
makefile=apache-1.3/Makefile
else
TARGET=${SRCDIR}/apache-2.0
makefile=apache-2.0/Makefile
fi
+++
Any comment? - I am using the same in mod_jk -
Doesn't this needlessly create a new directory on the machine? And, it doesn't
clean that dir until the next time you try to configure. At the very least,
we should execute the '$RM -rf test' command after the grep. It would actually
be possible to just do 'grep STANDARD20 $APXS'. This is because
the perl script actually has a copy of the module in it. That is probably the
cleanest solution.
I have also noted strange things mod_webapp always returns 500.
It seems the request is mapped (wam_match) correctly but the handler
(wam_invoke) is not called.
What is the problem you have detected?
It looks like when Tomcat returns a 302 through mod_webapp, we are
redirected to a strange port. I haven't even begun to look at it yet, but I
am hoping to have time either today or tomorrow.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
Build system for mod_webapp and Apache 2.0
Hi everybody, I'm new to the list, but for anybody who doesn't know me,
I have been pretty active on httpd 2.0. I decided to port mod_webapp last
week, and Pier committed the code for me, but the build system isn't there, so
I am posting the build system now.
This is currently working, although there seems to be some strangeness when
Tomcat responds to mod_webapp. I'll be looking into that over the next few days.
I hope to be more active as time permits. :-)
The Makefile.in should be placed in jakarta-tomcat-connectors/webapp/apache-2.0
Thanks,
Ryan
Index: webapp/Makefile.in
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/webapp/Makefile.in,v
retrieving revision 1.15
diff -u -d -b -w -u -r1.15 Makefile.in
--- webapp/Makefile.in 2001/08/06 22:21:52 1.15
+++ webapp/Makefile.in 2001/08/29 23:51:41
@@ -62,6 +62,7 @@
LOCALDIRS = @TGTDIRS@ @TARGET@
APRDIR = @APRDIR@
+INSTALLDIR = @TARGET@
CFGS = @CONFIGFILES@ \
@SRCDIR@/lib/pr_warp_defs.h \
@@ -71,6 +72,8 @@
all: apr-all local-all
+install: local-install
+
clean: apr-clean local-clean
distclean: clean
@@ -128,6 +131,19 @@
$(ECHO) Cleaning up $${DIR}... ; \
cd $${DIR} ; \
$(MAKE) clean ; \
+ RET=$$? ; \
+ cd $(SRCDIR) ; \
+ if test $${RET} != 0 ; then \
+ exit $${RET} ; \
+ fi ; \
+ done
+
+local-install:
+ @for DIR in $(INSTALLDIR) ; do \
+ $(ECHO) ; \
+ $(ECHO) Compiling sources in $${DIR}... ; \
+ cd $${DIR} ; \
+ $(MAKE) install ; \
RET=$$? ; \
cd $(SRCDIR) ; \
if test $${RET} != 0 ; then \
Index: webapp/configure.in
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/webapp/configure.in,v
retrieving revision 1.22
diff -u -d -b -w -u -r1.22 configure.in
--- webapp/configure.in 2001/08/06 22:48:45 1.22
+++ webapp/configure.in 2001/08/29 23:51:41
@@ -214,6 +214,41 @@
AC_SUBST(TARGET)
dnl --
+dnl Process the --with-apxs2[=FILE] command line argument
+dnl --
+AC_ARG_WITH(apxs2,
+ [ --with-apxs2[=FILE] build a shared Apache 2.0.x module.],
+ [
+if ${TEST} -n ${TARGET}
+then
+ AC_MSG_ERROR([target already defined as ${TARGET}])
+fi
+
+APXS=${withval}
+
+if ${TEST} -z ${APXS} ; then APXS=yes ; fi
+if ${TEST} ${APXS} = yes
+then
+ AC_PATH_PROG(APXS,apxs,${PATH})
+fi
+
+AC_MSG_CHECKING([if apxs is working])
+${APXS} -q CC 21 1/dev/null
+if ${TEST} ! $? -eq 0
+then
+ AC_MSG_RESULT([no])
+ AC_MSG_ERROR([apxs is unworkable])
+fi
+AC_MSG_RESULT([ok])
+
+TARGET=${SRCDIR}/apache-2.0
+makefile=apache-2.0/Makefile
+AC_MSG_RESULT([setting target module to... ${TARGET}])
+ ]
+)
+AC_SUBST(TARGET)
+
+dnl --
dnl Process the --with-apr=... command line argument
dnl --
AC_MSG_CHECKING([APR directory])
@@ -253,7 +288,7 @@
LOCAL_HEADER([Configuring APR])
LOCAL_FILTEREXEC(
- [./configure --enable-static --disable-shared --disable-threads],
+ [./configure --enable-static --disable-threads],
[APR configure])
if ${TEST} ${ret} -ne 0
then
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
# = #
# #
# The Apache Software License, Version 1.1 #
# #
# Copyright (c) 1999-2001 The Apache Software Foundation. #
# All rights reserved.#
# #
# = #
# #
# Redistribution and use in source and binary forms, with or without modi- #
# fication, are permitted provided that the following conditions are met: #
# #
# 1. Redistributions of source code must retain the above copyright notice #
#notice, this list of conditions and the following disclaimer
submitting a patch
I have found a bug and fixed it in the Tomcat NT service program. How do I submit it? Do you guys take bug fixes from independent people. I had heard from several people that they had submitted bug fixes and you guys told them to f*** off (not necessarily in those terms). __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/
