Thanks Hasan, this makes things clearer! It seems its a problem with the two static initializers, apparently only the first of them is executed when running all tests together, then we have a Null key in the HashMap. Very Strange behaviour, the compile doesn't complain about two static initializers, but at runtime sometimes one is left out.
Cheers, reto On Sun, Feb 14, 2010 at 7:56 PM, Hasan <[email protected]> wrote: > ok, it seems the problem lies in this line > > typeConverterMap.put(byteArrayType, new ByteArrayConverter()); > > if you comment out this line and the respective test > it will build without error. > but what's wrong with that line? > > hasan > > Hasan wrote: >> >> Hmmm... weird.... the individual tests are ok... >> i debug the test, but can't find errors yet >> >> hasan >> >> Reto Bachmann-Gmuer wrote: >>> >>> I'm not getting it: >>> >>> java.lang.NullPointerException at >>> >>> org.apache.clerezza.rdf.core.impl.SimpleLiteralFactory.createTypedLiteral(SimpleLiteralFactory.java:229) >>> >>> I'm seeing that exception only when running the tests on the whole >>> projects, not when running the individual test files. >>> >>> the exception is >>> >>> //added this to make sue value is not null >>> if (value == null) { >>> throw new IllegalArgumentException("value may not be null"); >>> } >>> for (Map.Entry<Class<?>, TypeConverter<?>> converterEntry : >>> typeConverterMap.entrySet()) { >>> if >>> (converterEntry.getKey().isAssignableFrom(value.getClass())) { >>> // <-ON THIS LINE >>> >>> maybe I'm just to tired, would be glad if somebody had a look at it. >>> >>> reto >>> >>> On Sat, Feb 13, 2010 at 6:08 PM, Apache Hudson Server >>> <[email protected]> wrote: >>> >>>> >>>> See >>>> <http://hudson.zones.apache.org/hudson/job/clerezza-trunk-1.6/org.apache.clerezza$org.apache.clerezza.rdf.core/106/changes> >>>> >>>> >>>> >>>> >> > > -- > Hasan Hasan > trialox AG > Binzmühlestrasse 14, CH-8050 Zürich > Tel: +41 44 635 7577 > Fax: +41 44 635 7574 > URL: http://trialox.org > >
