drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Parth Chandra <pchan...@maprtech.com>
Subject Re: My last patch moved us to protobuf 2.6
Date Thu, 19 Mar 2015 22:00:57 GMT
I think the C++ client will be fine since we don't generate the code every
time. There may be some protobuf changes coming soon that will require
updating the code. I'll update the cmake required version at that time.

Parth

On Thu, Mar 19, 2015 at 2:43 PM, Xiao Meng <xiaom@simba.com> wrote:

> Hi,
>
> I think this may cause issues for C++ client, which relies on protobuf 2.5.
>
> It would be good to keep protobuf 2.5.
>
> On OS X,  you can use homebrew-versions to get an older version of protoc:
> https://github.com/Homebrew/homebrew-versions
>
> Thanks,
>
> Xiao
>
> ________________________________________
> From: chriswestin42@gmail.com <chriswestin42@gmail.com> on behalf of
> Chris Westin <cwestin@yahoo.com>
> Sent: March 19, 2015 2:20 PM
> To: dev@drill.apache.org
> Subject: My last patch moved us to protobuf 2.6
>
> Jacques just committed my patch for DRILL-2245. Among other things, this
> moved us to protobuf 2.6 (we were on 2.5, release notes are here
> https://code.google.com/p/protobuf/source/browse/trunk/CHANGES.txt).
>
> I needed to add some new enum symbols to QueryState, which required
> regenerating some of the .proto files. I didn't have protoc from before.
> When I did a search, the top result was for how to install protoc using
> brew. When I did that, I got 2.6.
>
> When I regenerated the .java files for the .protos, about half a dozen of
> them changed, but the changes were all comments, so I didn't think anything
> of it. I don't think it should be a problem, but just wanted to check to
> see if anyone knows of a reason we shouldn't do this. If so, then I can
> submit another patch just to downgrade protoc back to 2.5 (that's a better
> bet than trying to undo this rather large patch).
>
> Chris
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message