Is it possible to query many Parquet files stored on S3 with a single query?

[collimarco]

Let's say that you have thousands of Parquet files already stored on S3.

The schema is similar, but it is not identical for all the files. For example:

• some may have some additional columns or less columns
• rarely a column may be of different type (like a status column may be an integer but sometimes a string).

Is it possible to use Trino to query all the files in a directory on S3?
Like SELECT count(*) FROM /s3path/custom/2023/06/*.parquet WHERE status = 300

Or it possible to give Trino a long list of files on S3 to query dynamically?
Like SELECT count(*) FROM s3_file_1, s3_file_2, ... s3_file1000 WHERE status = 300

Ideally each query uses a different set of files (they are grouped in partitions), so it would be better to be able to execute the queries directly on a list of files, without having to perform too many intermediate steps.

Is this possible with Trino?