curl --request POST \
--url https://ndi-api.nace.ai/v1/workspaces/{workspace_id}/tools/catalog \
--header 'Content-Type: application/json' \
--header 'X-API-Key: <api-key>' \
--data '
{
"sql": "<string>"
}
'{
"columns": [
"<string>"
],
"empty_tables": [
{
"columns": [
"<string>"
],
"document_type": "<string>",
"table": "<string>"
}
],
"row_count": 123,
"rows": [
{}
],
"tables": [
{
"columns": [
{
"description": "<string>",
"kind": "<string>",
"name": "<string>"
}
],
"description": "<string>",
"in_scope_components": 123,
"table": "<string>"
}
],
"truncated": true
}{
"error": {
"code": "invalid_request",
"detail": null,
"message": "Request body has extra fields that this operation does not accept.",
"request_id": "req-01j9k2n3p4q5r6s7t8v9",
"retryable": false
}
}{
"error": {
"code": "invalid_request",
"detail": null,
"message": "Request body has extra fields that this operation does not accept.",
"request_id": "req-01j9k2n3p4q5r6s7t8v9",
"retryable": false
}
}catalog
The metadata catalog ingestion extracted from every document: its tables, or one SELECT’s rows.
Cost class: fast. Answers inline: no job, no credits. Without sql: one table per document
type the workspace holds (invoices, contracts, statements…) with its typed columns, plus
catalog_document over every record. With sql: one read-only SELECT over those tables, its rows
capped and truncated set when cut. A refused or failing statement answers invalid_request with
the reason; so does a workspace whose catalog is too large to query whole.
A record is one document section, spreadsheet tab, recording or JSON section, never one printed line:
the entries a listing, register or ledger prints (each credit note in a credit-note report) are not
records of their own. The listing is one record; read it with read-file, or query a tab’s rows
with run-sql. Every record carries source_path, the path the other file tools take.
curl --request POST \
--url https://ndi-api.nace.ai/v1/workspaces/{workspace_id}/tools/catalog \
--header 'Content-Type: application/json' \
--header 'X-API-Key: <api-key>' \
--data '
{
"sql": "<string>"
}
'{
"columns": [
"<string>"
],
"empty_tables": [
{
"columns": [
"<string>"
],
"document_type": "<string>",
"table": "<string>"
}
],
"row_count": 123,
"rows": [
{}
],
"tables": [
{
"columns": [
{
"description": "<string>",
"kind": "<string>",
"name": "<string>"
}
],
"description": "<string>",
"in_scope_components": 123,
"table": "<string>"
}
],
"truncated": true
}{
"error": {
"code": "invalid_request",
"detail": null,
"message": "Request body has extra fields that this operation does not accept.",
"request_id": "req-01j9k2n3p4q5r6s7t8v9",
"retryable": false
}
}{
"error": {
"code": "invalid_request",
"detail": null,
"message": "Request body has extra fields that this operation does not accept.",
"request_id": "req-01j9k2n3p4q5r6s7t8v9",
"retryable": false
}
}Authorizations
Path Parameters
Workspace identifier.
Body
catalog: no sql describes the catalog's tables; one sql runs a read-only SELECT over them.
One read-only SELECT (or WITH ... SELECT) over the catalog tables; omit to list the tables.
1Response
Successful Response
A describe fills tables/empty_tables; a query fills columns/rows/row_count/truncated.
With sql: the result's column names, in order.
Document types with no records here: queryable, every query over them is empty.
Show child attributes
Show child attributes
With sql: how many rows rows holds.
With sql: the result rows, each keyed by column name.
Show child attributes
Show child attributes
Without sql: catalog_document and every document-type table with records.
Show child attributes
Show child attributes
True when the result had more rows than were returned.