On 19/08/2011 16:45, Wolfgang Laun wrote: > I'll not create any JIRAs for problems that come and go with the > presence and absence of dialect "mvel". > > After one of those, Mark has urged me to create another JIRA for > MVEL, and this I just won't do. For me, as a /Drools /user, this > project /Drools /has adopted MVEL as a vehicle for primary expression > and statement evaluation; since 5.2.0 it seems that I don't have a > choice for some parts of the Drools language. For me, this means that > the full responsibility now rests with the Drools team - I cannot > simply work around by sticking with 'dialect "java"'.

> > Sorry if this sounds harsh and egocentric, but what I've heard about > MVEL maintainer's non-eagerness for fixing bugs hasn't inspired me to > do otherwise. (Since this is hearsay, I should add the politically > correct statement: "innocent until proven otherwise".) Mike just won't explore bugs, without unit tests. Once a unit test is in place he's responsive in fixing it. Which is fair enough. Mark > > Regards > Wolfgang > > > > _______________________________________________ > rules-dev mailing list > rules-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/rules-dev _______________________________________________ rules-dev mailing list rules-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-dev

On 19 August 2011 18:07, Mark Proctor <mproctor(a)codehaus.org <mailto:mproctor@codehaus.org>> wrote: Mike just won't explore bugs, without unit tests. Once a unit test is in place he's responsive in fixing it. Which is fair enough. The point is that for me (and presumably most other Drools users) the step from a DRL construct towards an MVEL unit test isn't a simple one. -W

_______________________________________________ rules-dev mailing list rules-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-dev

On Sat, Aug 20, 2011 at 12:36 AM, Mark Proctor <mproctor(a)codehaus.org&gt;wrote: > On 19/08/2011 20:44, Wolfgang Laun wrote: > > On 19 August 2011 18:07, Mark Proctor <mproctor(a)codehaus.org&gt; wrote: > >> >> Mike just won't explore bugs, without unit tests. Once a unit test is >> in place he's responsive in fixing it. Which is fair enough. >> > > The point is that for me (and presumably most other Drools users) the step > from a DRL construct towards an MVEL unit test isn't a simple one. > -W > > We can't write all the software ourselves, we have to re-use OSS libraries > and be a part of those communities, and the baggage that comes with them. > > We've hired Mario full time, who is now responsible for this area. So that > should atleast help in some aspect, as we can be more pro-active in > ownership of MVEL issues. > > Mark > The problem seems caused by MVEL not properly managing methods accepting a varargs. I created a failing test and sent a pull request including it: https://github.com/mvel/mvel/pull/11 As Mark wrote, in future I will try to help Mike in testing and possibly fixing those MVEL issues. Mario _______________________________________________ rules-dev mailing list rules-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-dev