Canvas Data 2 Object Processing

[BradDudra]

I've been tasked with extracting various tables from Canvas to insert the data into another relational database.  Per specifications, the output type for the files needs to be parquet.  When I retrieve the output documents from the Canvas Data 2 API it appears that the first object in S3 is a schema definition for the table.  Is it safe to make the assumption that the first object will always be a schema definition?  The reason I ask is that I'm trying to configuration my ETL jobs to handle these files in a repeatable way and I need a way to identify which files need to be omitted from processing.