DB.SCHEMA

=DB.SCHEMA lists the columns of a source you named with =DB.SOURCE. Use it to discover what you can query before writing the SQL for =DB.QUERY.

Signature

=DB.SCHEMA(source)

It takes exactly one source cell. To inspect several sources, call =DB.SCHEMA once per source — there is no multi-source form.

Parameters

Parameter	Type	Description
`source`	cell	A single cell holding the result of `=DB.SOURCE`. A bare URL, a filename, or an array of sources is rejected — name the source with `=DB.SOURCE` first.

Return shape

DB.SCHEMA spills a 2-column (name, type) grid: one row per column, with no header row. The first column is the column name; the second is its DuckDB type.

For the samples sales source the spill looks like:


`order_id`	`INTEGER`
`order_date`	`DATE`
`region_id`	`INTEGER`
`product_id`	`INTEGER`
`quantity`	`INTEGER`
`revenue`	`DOUBLE`

(The grid above shows the two spilled columns; DB.SCHEMA adds no header of its own.)

Limits

60-second timeout — describing an unreachable or very slow source is cancelled after 60 seconds with #N/A.
Recalculation — if the source cell changes while DB.SCHEMA is resolving, the in-flight describe is cancelled and re-run.

Errors

Condition	Cell error	Message
The source host can’t be reached	`#N/A`	Source not reachable: alias
The describe ran longer than 60 seconds	`#N/A`	Query timed out (60 s) — try a smaller filter
The source can’t be parsed as its declared format	`#VALUE!`	Cannot parse source: …
DuckDB rejects the describe	`#VALUE!`	SQL: …

Example

Name the samples sales file in A1, then list its columns in B1:

A1: =DB.SOURCE("https://dev-fdb.sumpalabs.com/samples/sales.parquet", "sales")
B1: =DB.SCHEMA(A1)

B1 spills the six (name, type) rows shown above.