Search Index

Search Index is the MongoDB component for managing the search indexes of an Atlas collection: those for text search and those for search by meaning with MongoDB Vector Store.

When to use it

• Creating the vector index a collection needs before searching it by meaning.
• Checking which indexes a collection has.
• Deleting an index no longer in use.

Configuration

The Operation field decides what the component does: List Search Index (default), Create Search Index, Update Search Index or Drop Search Index.

You need a MongoDB Connection credential with your MongoDB cluster details. See Credentials.

Field	What it does
Credential (required)	The MongoDB Connection credential.
Database Name (required)	The database that holds the collection.
Collection Name (required)	The collection whose indexes are managed.
Operation (required)	What the component does.
Index Type (required)	vectorSearch (default), for search by meaning, or search, for text search.
Index Name (required)	A unique name for the index, for example products_index.
Index Definition (required)	The index structure in MongoDB format. For a vector index, give the field that holds the vectors, their size and how similarity is measured; the field’s help includes an example.

Output

The Data output holds the list of indexes or the result of the operation.

Example

Before searching its catalogue by meaning, a shop creates the vector index.

1. Add Search Index with the MongoDB Connection credential, the shop database and the products collection.
2. Choose Create Search Index, the vectorSearch type and the name products_index.
3. In Index Definition, describe the vector field following the help’s example, and run it.

Related components

• Document: works with a collection’s documents.
• MongoDB Vector Store: uses the vector index to search.