[Monetdb-developers] Re: [ monetdb-Bugs-1274819 ] "type error: no variant of function ... accepts the the giv"

Jens Teubner jens.teubner at in.tum.de
Sun Aug 28 17:29:46 CEST 2005


(It starts getting a communication thread now, no longer a bug track. So
I'll reply via email, with CC to the monetdb-developers list.)

> Re: The testsuite. I like a small test suite on my machine,
> so that I can check that tools on my box did nothing
> unexpected during the
> build.  Its also the best way to see what behavior can be
> expected to work right out of the box.  Are the nightly test
> scripts available
> online?

Of course are those test scripts available. Just type `make check'.

Be aware, though, that those test cases take their time. Also note that
there are currently some tests in that do not work as expected. So you
will not get a 100% pass.

The `make check' run will print a summary after processing. You will
also find a directory that contains the detailed test results marked up
with HTML (directory `mTests'). Tests with unexpected outcome are
highlighted in red; you will see the diffs between actual and expected
outcome.

Results of the nightly test runs, btw., are available on the MonetDB
TestWeb website: http://monetdb.cwi.nl/Development/TestWeb/index.html.

> Re: The usecases. They might be nice just to check
> cross-implementation consistency.  E.g. have the nightly
> test queries been ran by a different xquery processor?

Most test cases are tested ``by hand'', sometimes by comparing against
other implementations. At least in theory, all of them are cross-checked
by hand.

> > (well, you may redefine a function with same name, but
> > different number of arguments).
> Wow, I didn't know that.  I learn something new everyday. 

:-)

Regards from Munich

Jens

-- 
Jens Teubner
Technische Universitaet Muenchen, Department of Informatics
D-85748 Garching, Germany
Tel: +49 89 289-17259     Fax: +49 89 289-17263

"Tea or coffee?"
"Coffee, without cream, please."
"It will be without milk, we have no cream."




More information about the developers-list mailing list