wuchong commented on issue #9203: [FLINK-13375][table-api] Improve config names in ExecutionConfigOptions and OptimizerConfigOptions URL: https://github.com/apache/flink/pull/9203#issuecomment-514511313 Hi @twalthr , I agree we should keep config names short. However, IMO, 4 levels is too strict, because in fact, we uses only have 2 levels to describe a config. There are some configs are difficult to squash to 4 levels. For example: `table.optimizer.distinct-agg.split.bucket-num` and `table.optimizer.distinct-agg.split.enabled`. I think maybe we can relax the restrictions to 5 levels? Regarding to the aligned by operator vs functionality, I would like to align by functionality. Because some options are not relative to a specific operator, e.g `table.optimizer.reuse.sub-plan.enabled`. And some functionality words are well-known, e.g. `hash-agg`, `join-reorder`. Regarding to the `.enabled`, my concern is 1. it would be better to align all `enabled` options in the same style 2. using `.enabled` is more flexible for introducing new options under the same parent. > we don't need a hierarchy because they just consist of a single child. I think it's fine to have a hierarchy even if it consist of a single child. Because we can't enumerate all the configs at one blow. We should leave some space for the future. And the [Flink convention](https://docs.google.com/document/d/1JGRVK6udBFa8xbyv7ipPQiLEhxebA64f8KKcaAH0UPY/edit#) doesn't restrict this, right? What do you think? @twalthr @godfreyhe
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services
