Update existing rows via merge

To change existing rows, use merge_rows with _operation: "UPDATE", keyed on the primary key. Send only the PK and the fields you're changing - omitted columns keep their current values.

Update

{ "table": "eddytor.cfg_xxx.<uuid>_products",
  "comment": "Q1 repricing",
  "rows": [
    { "_operation": "UPDATE", "product_id": "P001", "price": 34.99 },
    { "_operation": "UPDATE", "product_id": "P002", "price": 12.50, "status": "active" }
  ]}

eddytor merge table eddytor sales orders --file ./price-changes.csv

Update by condition

merge_rows keys on the PK, so to update "all rows where …":

1. query_rows with the filter to get the matching PKs.
2. Build UPDATE rows for those PKs.
3. merge_rows with a comment.

Atomic & validated

The batch is atomic and fully validated - if any updated row breaks a constraint or domain, the whole batch is rejected. Then validate and check the history metrics.

Next

• Delete rows · Bulk corrections safely.