Implement information_schema.schemata

[alamb]

The idea is to implement https://www.postgresql.org/docs/current/infoschema-schemata.html in DataFusion to make it easier to work with standard tools like Apache Superset

[Tangruilin]

I am interested on this problem.
Maybe you can assign it to me

[alamb]

Thanks @Tangruilin !

cc @JanKaul who made the original request in Discord

[JanKaul]

Thanks @Tangruilin !

Let me know if I can assist you somehow.

[Tangruilin]

Thanks @Tangruilin !

Let me know if I can assist you somehow.

I have read about some of the code.
But I am a little confused how to start. Could you tell me where the metadata like information_schema.table add in the code, and I will look at that part

[JanKaul]

I think the following steps could work:

1. Create a InformationSchemaSchemataBuilder (similar to the TablesBuilder) that has all the columns of the information_schema.schemata table.
2. Create a make_schemata function (similar to make_tables) that obtains the schemas from the catalog list. This can be used to populate the previously created builder.
3. Create a InformationSchemaSchemata (similar to InformationSchemaTables) that uses the make_schemata method to implement the PartitionStream trait.

Maybe it helps