axis-java-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Glen Daniels" <gdani...@macromedia.com>
Subject Re: [GUMP] Function Test Failure - Axis
Date Sat, 09 Jun 2001 19:26:56 GMT
My $0.02:

I'd suggest distinguishing between "sample tests" and "functional tests".  The
former are the tests for the sample applications, and will require network
connectivity in most cases.  The latter are the "black box" tests for the
system, which should not require a network connection to run successfully.  I
expect the functional tests will grow beyond the point where we want to include
every test as a full-blown sample application.

We might even break the sample tests down into local-samples and remote-samples
so it's easier to run more tests locally.

--Glen

----- Original Message -----
From: "Sam Ruby" <rubys@us.ibm.com>
To: <axis-dev@xml.apache.org>
Sent: Friday, June 08, 2001 10:29 PM
Subject: Re: [GUMP] Function Test Failure - Axis


> Rob Jellinghaus wrote:
> >
> > Yes, this makes sense.  I think the functional tests should not reference
> > external servers which may or may not be up (a la the xmltoday quote
> > service).  The functional tests should only make localhost connections.
> > Otherwise, we will continue to get spurious failures such as this.  What
> do
> > you all think?  (And yes, this is my bad for adding the doTestIBM method
> in
> > the first place...)
>
> I'm not of one mind on this issue.  I will admit to a significant
> temptation to rip this out.  But on the other hand, this is a sample that
> we ship, and if it is unreliable or stops working, it behooves us to to
> know this early.
>
> This actually is the second XML service we have used.  There now are SOAP
> servers out there which provide XML stock quotes.  We should be
> interoperable with them...so it might be nice to have a demo WebService
> which invokes another web service...
>
> Here's a simple one: http://www.xmethods.net/detail.html?id=2
>
> > You mentioned that you had patched FileReader's inability to handle the
> > returned AxisFault.  Are you planning to check that in?  (or did I
> mishear
> > you?)
>
> I made a number of improvements, previously it was looping over the same
> file every 200 milliseconds...
>
> Those changes have been checked in, but do not yet appear to be sufficient.
>
> - Sam Ruby
>


Mime
View raw message