RussellSpitzer commented on issue #1422: URL: https://github.com/apache/iceberg/issues/1422#issuecomment-700756335
So it's been a while and we've gotten a bit of feedback on the PR's - Thanks @kbendick , @aokolnychyi , and @rdblue Since I think at the moment it seems like the safer path is the MetadataTables based approach I'm going to start cleaning up and moving forward with that PR. For me the big benefits of the other approach were mainly around code re-use and the possibility of other systems to use the same method for parallelization. Since we haven't really heard from anyone using presto, flink, or whatnot I'm going to assume that the benefit is probably not that great. We can always use a set of common ideas with all these platforms as well each having their own planning algorithm if need be. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
