Lance
GitHub Get started
Docs /Integrations /Apache Spark /Operations /DDL /DROP INDEX

DROP INDEX

Remove an index from a Lance table.

Spark Extension Required

This feature requires the Lance Spark SQL extension to be enabled. See Spark SQL Extensions for configuration details.

Usage

ALTER TABLE lance.db.users DROP INDEX idx_id;

Output

The DROP INDEX command returns the following information:

Column Type Description
index_name String The name of the dropped index.
status String The result status (dropped).

How It Works

The DROP INDEX command removes the named index from the dataset manifest. Physical index files are not deleted immediately; they are cleaned up by the VACUUM command during garbage collection. This allows older dataset versions to continue referencing the index for time travel.

Notes and Limitations

  • Dropping a non-existent index results in an error.
  • Index names are case-insensitive (stored as lowercase).
  • To recreate an index after dropping, use CREATE INDEX.
  • To list existing indexes before dropping, use SHOW INDEXES.
← Previous
CREATE INDEX