feature request: add parquet file name to PARQUET_SCAN trace in the profiler

[hamilton]

When profiling a query, duckdb gives a bit of useful information about what table / entity is being scanned via SEQ_SCAN, but does not provide comparable information for PARQUET_SCAN, making it difficult to inspect the query sources.

The trace for SEQ_SCAN has this:

   {
      "name": "SEQ_SCAN",
      "timing": 0.000004,
      "cardinality": 1,
      "extra_info": "tmp\n[INFOSEPARATOR]\na\nb",
      "timings": [],
      "children": []
    }

I can easily parse the extra_info field on [INFOSEPARATOR] to get the table name.

The PARQUET_SCAN entry does not, however, include the parquet file name that is being scanned:

{
          "name": "PARQUET_SCAN",
          "timing": 0.05395,
          "cardinality": 2048,
          "extra_info": "unique_key\ncreated_date\nclosed_date",
          "timings": [],
          "children": []
 }

Ideally we could add the file name to the extra_info entry like this:

        {
          "name": "PARQUET_SCAN",
          "timing": 0.05395,
          "cardinality": 2048,
          "extra_info": "'./data/download.parquet'\n[INFOSEPARATOR]\nunique_key\ncreated_date\nclosed_date",
          "timings": [],
          "children": []
        }

[github-actions]

This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 30 days.

[github-actions]

This issue was closed because it has been stale for 30 days with no activity.