Re: How should the current Tree and NodeState apis behave in the case of a null child name?

2013-02-19 Thread Alex Parvulescu
> Null doesn't make any sense there, so I'd consider it invalid Perfect, see attached patch on OAK-635. I used a combo of @Nonnull and checkNotNull to enforce this constraint. alex On Tue, Feb 19, 2013 at 10:22 AM, Michael Dürig wrote: > +1 > Michael > On Feb 19, 2013 9:20 AM, "Jukka Zitting"

Re: How should the current Tree and NodeState apis behave in the case of a null child name?

2013-02-19 Thread Michael Dürig
+1 Michael On Feb 19, 2013 9:20 AM, "Jukka Zitting" wrote: > Hi, > > On Tue, Feb 19, 2013 at 11:14 AM, Alex Parvulescu > wrote: > > I'm looking at the Tree and the NodeState interfaces, I see a lot of api > > docs about the returned value but no constraints about the input child > name. > > Is n

Re: How should the current Tree and NodeState apis behave in the case of a null child name?

2013-02-19 Thread Jukka Zitting
Hi, On Tue, Feb 19, 2013 at 11:14 AM, Alex Parvulescu wrote: > I'm looking at the Tree and the NodeState interfaces, I see a lot of api > docs about the returned value but no constraints about the input child name. > Is null considered a valid input? Null doesn't make any sense there, so I'd con

How should the current Tree and NodeState apis behave in the case of a null child name?

2013-02-19 Thread Alex Parvulescu
hi, apologies if this has been discussed already and I missed it. I ran into this via some tets in the AccessControlManagerImplTest that trigger a NPE in the Segment MK impl [0]. I'm looking at the Tree and the NodeState interfaces, I see a lot of api docs about the returned value but no constra