paul-rogers commented on issue #2030: Update docs for Metastore to point that all format plugins are supported URL: https://github.com/apache/drill/pull/2030#issuecomment-600926333 One additional question that came up in the context of reviewing another PR: we now have three ways to specify a schema: 1) table functions, 2) a provided schema file, and 3) metastore. Would be great if the docs can discuss how these are related. For example, suppose I have a provided schema and run `ANALYZE TABLE`. Does Drill uses my provided schema instead of trying to infer the schema? If so, this is a good way for users to resolve issues that might crop up for JSON or other formats where the metadata schema inference fails. If I have both a schema file, and the metastore, which schema is used to plan the query? Which is passed as the `schema` attribute to the `SubScan`? (From the user perspective, which is used at run time?) If I specify a table function schema for a table with either a metastore or provided schema, which is used? Presumably the table function? Does the use of the table function prevent the use of the metadata? Are the schemas combined? Do we check for schema conflicts?
---------------------------------------------------------------- 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
