gianm commented on issue #7288: AggregatorFactory.getMergingFactory() and getCombiningFactory() are too similar URL: https://github.com/apache/incubator-druid/issues/7288#issuecomment-474106325 I think in most (all?) cases, `foo.getCombiningFactory()` could be replaced by `foo.getMergingFactory(foo)`. > Even after reading Javadoc of both, I don't understand why both methods exist and what is the difference between them. It's not the prettiest design, but I can explain what it's for. The signature is `AggregatorFactory getMergingFactory(AggregatorFactory other)`. It's only used for one purpose: creating aggregatorFactories from segment metadata which can be used to query over the same column across multiple segments. Some example cases, - When a CompactionTask wants to compact two segments that both have a metric column `foo`, it uses `foo1.getMergingFactory(foo2)` to generate the aggregator factory that will be used to combine data from the two different segments. - When users issue a SegmentMetadata query with the "aggregators" analysis type and "merge" enabled (because they want to see what aggregators they should use at query time). It's different from `getCombiningFactory()` because it's a binary, not unary operation.
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
