dear all, We've made good progress on the 0.7.0 release. There's still a little over 40 issues left in the release milestone
https://issues.apache.org/jira/projects/ARROW/versions/12341141 Any help (especially on investigating bugs) would be appreciated. There are many patches in code review -- any assistance with reviewing (especially from Java developers / users) would also be great. I'll be working on burning down as many of these as I can and we can discuss where we stand in the sync call on Thursday at 16:00 UTC (will send out Hangout link to the mailing list). Thanks, Wes On Fri, Sep 1, 2017 at 9:54 AM, Uwe L. Korn <uw...@xhochy.com> wrote: > Hello all, > > I'm currently working on adding support for inserting Arrow records into > Databases with turbodbc via OBDC. I've already discovered a small bug > (PR is already sent) in Arrow. But as I touch a bit of code that was not > used much outside of Arrow/Parquet, I'd prefer to have the > implementation in turbodbc in Review before we start the 0.7.0 release > process. This will either happen today or latest on Tuesday. Just a > small-heads up so that we have all it needs for this in the next Arrow > release. > > Uwe > > On Thu, Aug 31, 2017, at 10:40 PM, Wes McKinney wrote: >> hi folks, >> >> There are a number of Java-related patches in flight that are going to >> need some TLC from folks with deeper experience using the affected >> code in downstream projects: >> >> Union changes + integration tests: >> https://github.com/apache/arrow/pull/987 >> Decimal type improvements + integration test support: >> https://github.com/apache/arrow/pull/994 >> Fixed size binary + integration tests: >> https://github.com/apache/arrow/pull/1012 >> >> It would be really super to get 3 more logical types fully integration >> tested against C++ in 0.7.0. This additionally would allow Spark to >> send Decimal data to Python via C++. >> >> Getting these done would get us closer to complete integration >> coverage. Still some things left to do (intervals, fixed size lists). >> >> thanks, >> Wes >> >> On Mon, Aug 28, 2017 at 10:03 AM, Wes McKinney <wesmck...@gmail.com> >> wrote: >> > hello again, >> > >> > There is still a lot of work to do for 0.7.0. There are another ~60 >> > JIRAs marked for the 0.7.0 release -- I don't think these will all get >> > done, but we should stretch to do as many as possible. I think we may >> > be able to complete Java/C++ integration tests for (sparse) unions and >> > decimals, which would be nice progress toward format stability. >> > >> > Please contribute whatever you can -- the target time frame (assuming >> > serious blockers do not arise) for a release candidate is the end of >> > next week. >> > >> > - Wes >> > >> > On Wed, Aug 23, 2017 at 10:00 AM, Wes McKinney <wesmck...@gmail.com> wrote: >> >> hi all, >> >> >> >> I would like to push for a 0.7.0 release candidate roughly 2 weeks for >> >> now, so shortly after the US Labor Day holiday. >> >> >> >> The release page on JIRA is here: >> >> https://issues.apache.org/jira/projects/ARROW/versions/12341141 >> >> >> >> I have not very carefully curated these JIRAs, so I will make a pass >> >> through when I can and remove anything in my purview (C++ and Python) >> >> that does not seem like a "must do" or "stretch goal" for the release. >> >> If there is any work you would like to see go into the next release, >> >> can you update JIRA accordingly (adding a Fix Version or creating new >> >> issues)? >> >> >> >> Thanks! >> >> Wes
