On 10 Dec 2011, at 10:04, Pete Muir wrote: > https://issues.jboss.org/browse/ISPN-1474 > > Currently we do a mix of: > > enableXXX() > disableXXX() > enabled(boolean b) > XXX(boolean b) > > where XXX is something to enable or disable (e.g. purgeOnStartup). We also the scatter the word use around in a very inconsistent fashion. this is a bit ugly indeed

> > I would like to rationalise this, and would propose that every boolean has: > > enableXXX > disableXXX > xxxEnabled(boolean b) > > The former 2 are nice for hardcoded config, the latter is nice when you are adapting one format to another (e.g. Paul). I think this would be much more readable. My only comment is that we allow two ways of doing the same thing, which might be slightly confusing for the users. Also the easiness of migration from old to new configuration is important, but IMO not as important to make the new config simple.

> > I would deprecate the usage of XXX(boolean b) and useXXX(boolean b) and remove in 6. > > Thoughts? > _______________________________________________ > infinispan-dev mailing list > infinispan-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/infinispan-dev _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev

On Tue, Dec 13, 2011 at 10:44 AM, Galder Zamarreño <galder(a)redhat.com&gt; wrote: > > On Dec 12, 2011, at 3:45 AM, Mircea Markus wrote: > >> >> On 10 Dec 2011, at 10:04, Pete Muir wrote: >> >>> https://issues.jboss.org/browse/ISPN-1474 >>> >>> Currently we do a mix of: >>> >>> enableXXX() >>> disableXXX() >>> enabled(boolean b) >>> XXX(boolean b) >>> >>> where XXX is something to enable or disable (e.g. purgeOnStartup). We also the scatter the word use around in a very inconsistent fashion. >> this is a bit ugly indeed > > +1 > >>> >>> I would like to rationalise this, and would propose that every boolean has: >>> >>> enableXXX >>> disableXXX >>> xxxEnabled(boolean b) >>> >>> The former 2 are nice for hardcoded config, the latter is nice when you are adapting one format to another (e.g. Paul). >> I think this would be much more readable. My only comment is that we allow two ways of doing the same thing, which might be slightly confusing for the users. Also the easiness of migration from old to new configuration is important, but IMO not as important to make the new config simple. > > I agree with Mircea that having two ways of doing the same thing might confuse users. If I had to choose I'd probably go for the latter option. >

How about: xxxEnabled() xxxEnabled(boolean b) However, I don't think enabled is going to work in every situation (loader.sharedEnabled?), so we might want to keep the xxx()/xxx(boolean b) form in some places. >>> >>> I would deprecate the usage of XXX(boolean b) and useXXX(boolean b) and remove in 6. >>> >>> Thoughts? >>> _______________________________________________ >>> infinispan-dev mailing list >>> infinispan-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/infinispan-dev >> >> >> _______________________________________________ >> infinispan-dev mailing list >> infinispan-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/infinispan-dev > > -- > Galder Zamarreño > Sr. Software Engineer > Infinispan, JBoss Cache > > > _______________________________________________ > infinispan-dev mailing list > infinispan-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/infinispan-dev _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev