gravitino [#4089] fix(hive catalog): the problem of slow acquisition of hive table list

What changes were proposed in this pull request?

the problem of slow acquisition of hive table list. Using listTableNamesByFilter replace the getTableObjectsByName method.

Why are the changes needed?

I found that list-table will takes 300s when a schema has 5000 tables .

Fix: #4089

Does this PR introduce any user-facing change?

no

How was this patch tested?

Manual testing

Aug 09 '24 12:08 mygrsun

Hi, @mygrsun could you plz resolve the comment and CI issue? Thanks

Aug 12 '24 07:08 mchades

@FANNG1 @yuqi1129 Can you help review it?

Aug 19 '24 03:08 mchades

Now,the ci issue is a troublesome problem. https://github.com/apache/gravitino/actions/runs/10452698362/job/28942732918?pr=4469 This ci issue is due to a bug in hive, and the iceberg project has the same problem. https://github.com/apache/iceberg/pull/2722#issuecomment-867363019

This hive bug is only happen in Derby. In our environment , the metastore storage is mysql,so we don't encounter this problem.

Aug 21 '24 05:08 mygrsun

@mygrsun Is there any progress? May I take on this?

Sep 26 '24 08:09 mchades

@mygrsun Is there any progress? May I take on this?

thanks,you can do it

Sep 26 '24 08:09 mygrsun

I have verified locally that through this PR, the time consumption of listing 1000 tables can be reduced from 2043ms to 14ms.

It's ready for review now. @jerryshao

Nov 01 '24 11:11 mchades