This is probably enough to reproduce it: Cheers, Paul. On 15/12/2010 7:55 PM, Stefan Bodewig wrote: > On 2010-12-15, Paul King wrote: > >> After a little more testing, one thing I did notice was an >> apparent stricter treatment of property expansion. >> With 1.8.1, having an arg to a forked java task like this: > >> > >> was happily ignored if skipFetch wasn't defined. >> With 1.8.2, we now get a failed build with reason: > >> Property skipFetch was circularly defined. > >> The fix was to add a default property defn but it seems >> like such a behavior change in a minor update release >> should warrant an entry in the WHATSNEW at least? > > I don't think this a conscious change but rather a side effect of a > change somewhere else. But it must be more complex than that, I don't > get a warning like yours with > > > > ${x} > > > > Stefan > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org > For additional commands, e-mail: dev-help@ant.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org For additional commands, e-mail: dev-help@ant.apache.org