Updating Tables

updateTable modifies an existing table without reloading the source data. This is useful when you have already rendered a map layer and want to update its attributes (for example, after a computation or a user-driven filter).

Replace Strategy

Drops and recreates the table with new data. Use this when the entire dataset changes:

const updatedGeojson = /* new FeatureCollection */;

await db.updateTable({
  tableName: 'buildings',
  data: updatedGeojson,
  strategy: 'replace',
});

For non-layer tables (CSV, JSON), pass an array of plain objects instead of a FeatureCollection:

await db.updateTable({
  tableName: 'incidents',
  data: [{ id: 1, severity: 3 }, { id: 2, severity: 5 }],
  strategy: 'replace',
});

Update by ID Strategy

Updates only the rows that match on a given ID column, leaving unmatched rows untouched:

await db.updateTable({
  tableName: 'buildings',
  data: updatedGeojson,
  strategy: 'update',
  idColumn: 'properties.building_id', // dot notation for nested properties
});

idColumn supports:

• 'id' — matches on the top-level id field of each GeoJSON feature
• 'properties.some_attribute' — matches on a nested property

Combine with autk-compute

A common pattern is to compute new values with autk-compute, then call updateTable to write the results back before re-rendering the map layer with updated thematic data.

Return Value

updateTable returns the updated Table object with refreshed column metadata, which you can inspect to confirm the schema:

const table = await db.updateTable({ ... });
console.log(table.columns);