On Fri, 20 Jul 2001, Ara Abrahamian wrote:
> Test cases are something we really need, but I don't know really how to
> write *effective tests* for our cases, because of different combinations
> and the fact that it's impacted by what's there in templates and also by
> the code itself.
I think we just need something that we can run a junit test against to
make sure that semanitcs works in the beans... not 100% sure if thats the
best way to go, but there are things that get through compile and dtd
checks, and even deploy, but then misbehave (missing tx instruction or
something)...
cheers
dim
>
> Ara.
>
> > -----Original Message-----
> > From: [EMAIL PROTECTED] [mailto:xdoclet-devel-
> > [EMAIL PROTECTED]] On Behalf Of Andrew Stevens
> > Sent: Wednesday, September 19, 2001 2:26 AM
> > To: [EMAIL PROTECTED]
> > Subject: Re: [Xdoclet-devel] a few bugs in ejb-jar.xml generation
> >
> > A wise old hermit known only as <[EMAIL PROTECTED]> once said:
> >
> > > Wouldn't it be nice if the XDoclet core could validate all generated
> > > XML files against a DTD (or XMLSchema for that matter) once it's
> > > generated? This would be great for testing, and we can use xerces
> (or
> > > any other validating parser) to do the job. We could write JUnit
> test
> > > cases (accompanied with test data which would be @tagged sources)
> that
> > > would ensure that XDoclet is always sound! It would result in *way*
> > > better quality of XDoclet. In my opinion this is more important than
> > > anything else. I can contribute with this if there is agreement on
> it.
> > >
> > > Right now, i'm thinking of adding a method to SubTask.java like
> this:
> > >
> > > validateXml( InputStream xml, InputStream dtd );
> > >
> > > This method could be called by SubTask subclasses after generating
> an
> > > xml file, with the possibility to turn validation on/off through ant
> > > parameters.
> > >
> > > Comments?
> > >
> > > Aslak
> >
> > Why stop at only XML? Anything that's generating a java class (e.g.
> all
> > the other ejbdoclet subtasks) could validate it by trying to compile
> it.
> > Of course, that only ensures the syntax is valid and not that it makes
> > sense, but you could say the same for the DDs. e.g. For a while the
> BMP
> > beans were getting "<persistence-type>Container<persistence-type>"
> > generated, which while wrong is still valid according to the DTD. I
> guess
> > more specific test cases could be written to check things like that,
> > though.
> >
> >
> > Andrew.
> >
> > _______________________________________________
> > Xdoclet-devel mailing list
> > [EMAIL PROTECTED]
> > https://lists.sourceforge.net/lists/listinfo/xdoclet-devel
>
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
>
> _______________________________________________
> Xdoclet-devel mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/xdoclet-devel
>
_______________________________________________
Xdoclet-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/xdoclet-devel
