Python: Upsert data

• Primary keys must be included in the values dict to use upsert.
• Chain .select() after upsert() to return specific columns from the upserted row(s).

Parameters

• json

  (Required)

  The values to insert. Pass an dict to insert a single row or an list to insert multiple rows.

• count

  (Optional)

  The property to use to get the count of rows returned.

• returning

  (Optional)

  Either 'minimal' or 'representation'. Defaults to 'representation'.

• ignore_duplicates

  (Optional)

  Whether duplicate rows should be ignored.

• on_conflict

  (Optional)

  Specified columns to be made to work with UNIQUE constraint.

• default_to_null

  (Optional)

  Make missing fields default to `null`. Otherwise, use the default value for the column. Only applies for bulk inserts.

Examples

Upsert your data

response = (
    supabase.table("instruments")
    .upsert({"id": 1, "name": "piano"})
    .execute()
)

Bulk Upsert your data

response = (
    supabase.table("instruments")
    .upsert([{"id": 1, "name": "piano"}, {"id": 2, "name": "guitar"}])
    .execute()
)

Upserting into tables with constraints

response = (
    supabase.table("users")
    .upsert(
        {"id": 42, "handle": "saoirse", "display_name": "Saoirse"},
        on_conflict="handle",
    )
    .execute()
)

Upsert and return selected columns

response = (
    supabase.table("instruments")
    .upsert({"id": 1, "name": "piano"})
    .select("id, name")
    .execute()
)