valkey-search

This package provides support for the valkey-search module, which adds indexing and querying support for data stored in Valkey Hashes or as JSON documents with the valkey-json module. It extends the Node Valkey client to include functions for the FT.* commands.

To use these extra commands, your Valkey server must have the valkey-search module loaded. To index and query JSON documents, you'll also need the valkey-json module. The easiest way to get both is the valkey/valkey-bundle docker image (Valkey 8.1 and later; valkey-search 1.1+ for FT.AGGREGATE and TEXT fields).

Valkey compatibility

This package was forked from @redis/search, which targeted RediSearch. valkey-search implements a subset of that command set and query language:

| Commands | Status on Valkey | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------| | FT.CREATE, FT.DROPINDEX, FT.INFO, FT._LIST, FT.SEARCH, FT.AGGREGATE | Supported by valkey-search | | FT.ALIAS*, FT.ALTER, FT.CONFIG, FT.CURSOR*, FT.DICT*, FT.EXPLAIN*, FT.PROFILE, FT.SPELLCHECK, FT.SUG*, FT.SYN*, FT.TAGVALS, aggregate cursors | Not implemented by valkey-search. Kept in the API for compatibility but fail with ERR unknown command |

Differences to keep in mind when using the supported commands against valkey-search:

• Field types: TEXT, TAG, NUMERIC and VECTOR are supported; GEO is not. SORTABLE, NOSTEM, SEPARATOR and CASESENSITIVE are accepted; WEIGHT other than 1, PHONETIC and NOINDEX are rejected.
• FT.CREATE options FILTER, TEMPORARY, MAXTEXTFIELDS, NOHL, NOFIELDS and NOFREQS are rejected.
• The match-all query * is not accepted on its own (it is only valid as the pre-filter of a vector query such as *=>[KNN 10 @vec $BLOB]). Use a filter such as @field:[-inf +inf] or @tag:{a|b} instead.
• FT.AGGREGATE: fields have to be LOADed before they can be used in GROUPBY/REDUCE/APPLY, and GROUPBY needs at least one property (GROUPBY 0 is rejected). For reducers over all documents, APPLY '1' AS all and group by @all.
• client.ft.info() is parsed by field name and returns the fields reported by valkey-search (indexName, indexDefinition, attributes, numDocs, numRecords, numTerms, state, backfillInProgress, ...).

Usage

For complete examples, see search-hashes.js and search-json.js in the Node Valkey examples folder.

Indexing and Querying Data in Valkey Hashes

Creating an Index

Before we can perform any searches, we need to tell valkey-search how to index our data, and which Valkey keys to find that data in. The FT.CREATE command creates a valkey-search index. Here's how to use it to create an index we'll call idx:animals where we want to index hashes containing name, species and age fields, and whose key names in Valkey begin with the prefix nodevalkey:animals:

await client.ft.create('idx:animals', {
  name: {
    type: SchemaFieldTypes.TEXT,
    SORTABLE: true
  },
  species: SchemaFieldTypes.TAG,
  age: SchemaFieldTypes.NUMERIC
}, {
  ON: 'HASH',
  PREFIX: 'nodevalkey:animals'
});

See the FT.CREATE documentation for information about the different field types and additional options.

Querying the Index

Once we've created an index, and added some data to Valkey hashes whose keys begin with the prefix nodevalkey:animals, we can start writing some search queries. valkey-search supports a rich query syntax for full-text search, faceted search, aggregation and more. Check out the FT.SEARCH documentation and the query syntax reference for more information.

Let's write a query to find all the animals where the species field has the value dog:

const results = await client.ft.search('idx:animals', '@species:{dog}');

results looks like this:

{
  total: 2,
  documents: [
    {
      id: 'nodevalkey:animals:4',
      value: {
        name: 'Fido',
        species: 'dog',
        age: '7'
      }
    },
    {
      id: 'nodevalkey:animals:3',
      value: {
        name: 'Rover',
        species: 'dog',
        age: '9'
      }
    }
  ]
}

Indexing and Querying Data with valkey-json

valkey-search can also index and query JSON documents stored in Valkey using the valkey-json module. The approach is similar to that for indexing and searching data in hashes, but we can now use JSON Path like syntax and the data no longer has to be flat name/value pairs - it can contain nested objects and arrays.

Creating an Index

As before, we create an index with the FT.CREATE command, this time specifying we want to index JSON documents that look like this:

{
  name: 'Alice',
  age: 32,
  coins: 100
}

Each document represents a user in some system, and users have name, age and coins properties.

One way we might choose to index these documents is as follows:

await client.ft.create('idx:users', {
  '$.name': {
    type: SchemaFieldTypes.TEXT,
    SORTABLE: 'UNF'
  },
  '$.age': {
    type: SchemaFieldTypes.NUMERIC,
    AS: 'age'
  },
  '$.coins': {
    type: SchemaFieldTypes.NUMERIC,
    AS: 'coins'
  }
}, {
  ON: 'JSON',
  PREFIX: 'nodevalkey:users'
});

Note that we're using JSON Path to specify where the fields to index are in our JSON documents, and the AS clause to define a name/alias for each field. We'll use these when writing queries.

Querying the Index

Now we have an index and some data stored as JSON documents in Valkey (see the JSON package documentation for examples of how to store JSON), we can write some queries...

We'll use the valkey-search query language and FT.SEARCH command. Here's a query to find users under the age of 30:

await client.ft.search('idx:users', '@age:[0 30]');