Let me see if I understand correctly:
In my web.xml I would put:
...
In my web containers proprietary descriptor file I would put:
...
So when I deploy, the container will match up the extra description
information in my proprietary descriptor with the
I am working on a deployment plugin for JBoss IDE and had a question about
the servlet spec (I was reading 2.3). On pages 112-116 it describes an ID
mechanism so that tools can provide additional deployment information. I
don't really understand what it is talking about. Does anyone have an
exam
JDT stands for Java Development Tools and is a subproject of the Eclipse
framework. I looked through the source a bit and it looks like the main
compiler class is org.eclipse.jdt.internal.compiler.Compiler. I found the
source in my plugins directory of Eclipse (no additional download needed)
und
Just use Check Out As... to check out HEAD and name it something like
"JBoss HEAD". Next, just use Check Out As... again and check out the 3.2
branch into something like "JBoss 3.2". That way both branches can be in
the same workspace at the same time.
-Phil
> I think it is better to use the jars from the module output directory,
> cause the exact location under build/output will change from version to
> version.
What build/output will change from version to version?
> Is there any way to make Eclipse create jars?
Go under the file menu to export.
Starting with M5 (I can't remember if it was in M4) you can do a check out
into... which will allow you to choose a source folder to check out into.
If you right click on your project, choose properties and then Java Build
Path there is a source tab where you can add new source directories. Next
If you are interested in XPath check out
http://jakarta.apache.org/commons/jxpath/index.html which has an XPath
implementation that will work with (quote from site) "JavaBeans, Maps,
Servlet contexts, DOM etc, including mixtures thereof. "
-Phil
Right click on an xml file and the popup menu will have an option called
"Run Ant...".
-Phil
[EMAIL PROTECTED]
Dain wrote:
> ServicePort contains a hostName, port, and InetAddress.
Doesn't InetAddress contain the hostName?
-Phil
___
Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.s
Can't we just create a special org.apache.log4j.Category (and later
org.apache.log4j.Logger) and make it so none of the methods do anything.
We could make it so the factory methods simply return our degenerate
Category. This would make the smallest possible log4j distribution for
embedded use.
Peter (Braswell) and I were looking at UDDI implementations for jboss.net
and the best one so far is UDDI4j. It is under the IBM Public License and
we weren't too sure what the differences between licenses are. Is IBM's
license open source enough for JBoss?
-Phil
_
I want to hear your message on "visibility".
-Phil
"marc fleury"
I think that all the extra information will confuse new JBoss users. If a
person is advanced enough to want to debug components within JBoss they
will be advanced enough to edit the log4j properties themselves.
-Phil
Maybe I am missing something, but isn't this whole issue about distributing
a compiler with JBoss so the user doesn't have to download the JDK and edit
a configuration file to point to it?
If that is the case, can't we distribute jikes?
-Phil
I think that the rationale behind making all your fields private is that
nobody can break design and access those fields. If you have a specific
design reason why a field needs to be accessed directly then you can open
up your field to allow access. Better yet is to control access through an
ac
15 matches
Mail list logo
