Don't we already shade it as a part of our shaded client artifact?
On Mon, Jul 11, 2016 at 10:38 AM, Andrew Purtell <[email protected]> wrote: > Even if they do - especially if they do - we should consider doing the same > for our downstreamers. > > >> On Jul 11, 2016, at 8:12 AM, Andrew Purtell <[email protected]> wrote: >> >> I would think so yes >> >>> On Jul 11, 2016, at 7:31 AM, Sean Busbey <[email protected]> wrote: >>> >>> If Hadoop shades their protobuf that should keep any altering they do from >>> impacting us, right? >>> >>> -- >>> Sean Busbey >>>> On Jul 9, 2016 3:02 PM, "Andrew Purtell" <[email protected]> wrote: >>>> >>>> PB3 sounds like a plan for 2.0 but what about all shipping versions. >>>> >>>> >>>>> On Sat, Jul 9, 2016 at 12:49 PM, Stack <[email protected]> wrote: >>>>> >>>>> The plan in HBASE-15638 is to shade our protobuf so we are independent of >>>>> anyone else's protobuf and so we can move on to one of our choosing or >>>> even >>>>> check in our own protobuf if we have to (protobuf is lacking in support >>>> for >>>>> offheap). Anoop and Ram are thinking we should go to pb3. I'll let them >>>>> talk of the testing they have done so far. >>>>> >>>>> St.Ack >>>>> >>>>> On Sat, Jul 9, 2016 at 10:44 AM, Andrew Purtell <[email protected]> >>>>> wrote: >>>>> >>>>>> We should look at HBASE-15638 again in light of HADOOP-13363. >>>>>> >>>>>> >>>>>> -- >>>>>> Best regards, >>>>>> >>>>>> - Andy >>>>>> >>>>>> Problems worthy of attack prove their worth by hitting back. - Piet >>>> Hein >>>>>> (via Tom White) >>>> >>>> >>>> >>>> -- >>>> Best regards, >>>> >>>> - Andy >>>> >>>> Problems worthy of attack prove their worth by hitting back. - Piet Hein >>>> (via Tom White) >>>>
