Add a Metadata.detect_from_ddl function

[npatki]

Problem Description

In an upcoming version of SDV, we will be consolidating the Multi and Single table metadata objects into a streamlined Metadata object. Effectively, this object is the same as multi-table, so it includes functions that are suited for detecting multiple tables at once.

However, if you have a single table, it should still be usable without adding much friction. Currently, there is a SingleTableMetadata.detect_from_ddl function that has not been added to the consolidated Metadata object.

Expected behavior

Add a detect_from_ddl function to the Metadata object. This will work similarly to the existing function, but with slight modifications to work for both the single and multi-table cases.

To improve the usage of this function, we will make it a class function that ultimately returns an instance of the metadata object.

from sdv.metadata import Metadata

metadata = Metadata.detect_from_ddl(filepath=... db_type='ibm_db2', table_names=['...'])

Parameters:

• (required) filepath: The location of the DDL file that contains the schema
• (required) db_type: A string with the type of SQL variant. Currently only 'ibm_db2' is supported
• table_names: A list of strings representing the table names to extract. For a single table, supply a list of 1 element.
  • (default) Consider all tables in the database
• database_name: A string with the name of the database. In the DDL file, the database name is usually prepended to the table, such as users.transaction_log
  • (default) None: Don't consider any specific database

Functionality: Detect metadata for all the tables that the user has specified. If there is only 1 table specified, then only 1 table's information will be in the metadata. If there are multiple tables specified, then they will all be included in the metadata including any primary/foreign key relationships that we find between them.

Output: An instance of the metadata object with the detected metadata