In more or less any package code that we write for our pipelines that use duckdb, we end up implementing some version of:

def T(path): 
 _PATH= ... resolve absolute path or URL prefix
 return f"delta_scan('{_PATH / path}')

to be used in notebooks like:

duckdb.sql(f"""
 select * from {T('delta/table/path')}
""")

There is probably good reason to not add some utility function like this but sharing in case I'm wrong. 😊