jakarta-watchdog-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Victor Duarte" <vdua...@blueisle.com>
Subject Re: web.xml
Date Thu, 13 Jul 2000 16:44:32 GMT
Hey guys,

Please take me off this list. It's not for me.

Thanks,

Victor

----- Original Message ----- 
From: Craig R. McClanahan <Craig.McClanahan@eng.sun.com>
To: Rousseau, John <JRousseau@silverstream.com>
Cc: <watchdog-dev@jakarta.apache.org>
Sent: Thursday, July 13, 2000 12:17 PM
Subject: Re: web.xml


> Hi John,
> 
> "Rousseau, John" wrote:
> 
> > Hi!
> >
> > I'm new to Watchdog, but I wanted to use the tests to validate our
> > servlet 2.2 implementation at SilverStream. I couldn't find an
> > archive of this list, and I hope this hasn't been asked before.
> >
> 
> The welcome message you received when you subscribed has some available
> commands for retrieving older messages and message indexes -- send an
> empty mail to "watchdog-dev-info@jakarta.apache.org" to get a new copy.
> 
> But you haven't really missed much -- it's been very quiet :-).
> 
> >
> > I was looking at the web.xml deployment descriptor from
> > servlet-tests.war and noticed that all of the elements were written
> > like this:
> >
> >     <param-name>
> >             OS
> >     </param-name>
> >
> > In XML, the value of param-name is then "\n\t\t\tOS\n" instead of
> > "OS". We could easily chop each string we get out of the deployment
> > descriptor, but this would prevent us from encoding a real tab or
> > space in the future.
> >
> > According to XML, if the intended value is "OS", the element should
> > be written as:
> >
> >     <param-name>OS</param-name>
> >
> > Is this an oversight, or am I missing something?
> >
> 
> At least for Tomcat (and I believe for some of the other containers) we
> trim leading and trailing whitespace from values retrieved from web.xml
> before interpreting them, so that either way of writing the parameter
> you used above would come up with the same result.
> 
> Looking at the spec, I don't see that behavior defined ... this is
> probably worth a clarification question to the spec lead for the servlet
> spec <Danny.Coward@eng.sun.com>.  This is something that should
> definitely be defined one way or the other, and the timing is great to
> make sure that whatever the answer is gets in to the upcoming servlet
> 2.3 spec as well.
> 
> I'll forward the question to Danny and ask him to reply to both of us.
> 
> >
> > Thanks!
> > -John
> 
> Craig McClanahan
> 


Mime
View raw message