On Sat, Aug 1, 2009 at 3:40 PM, Henrib<hbies...@gmail.com> wrote: > > > > Rahul Akolkar wrote: >> >> ... >> Correct, some patterns of usage make more sense than others. But to >> make any claims via the 223 API, we have to do rigorous code >> inspection and add some tests. Until someone does that, lets return >> null. >> ... >> > > Should I create a bug for later then - just as a reminder ? <snip/>
Sure, if you intend to look into it :-) > I can add some multi-threaded tests but I'm not sure they'll "prove" > multi-threading; they could only prove something if they were to fail. :-) > <snap/> Indeed (hence the code inspection I was talking about). > And about the ScriptContext (at least the engine & global bindings) , I > guess we'd need to alter those to be synchronized & thread safe. > > Aren't there any guidelines in the spec (or even tests via the TCK) or > something that could serve as a base ? > <snip/> Perhaps, I don't have the TCK (and ATM don't intend to ask for it either). -Rahul --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
