Full-Text Search

The FTS plugin indexes string fields for fast search queries.

Setup

import { Database, ftsPlugin } from "ctrodb"

const db = new Database({
  schema: {
    version: 1,
    collections: {
      articles: {
        fields: {
          title: { type: "string" },
          body: { type: "string" },
        },
        searchable: ["title", "body"],
      },
    },
  },
  plugins: [ftsPlugin()],
})

How it works

The plugin stores a token-to-document mapping in a _ctrodb_fts collection.

Indexing:

• On create/update/delete, tokens are extracted from searchable fields
• Each token maps to the set of document IDs containing it
• Tokens are lowercased, deduplicated, and filtered for stop words

Basic search (substring matching)

Without the FTS index, .search() does case-insensitive substring matching:

const results = await articles
  .query()
  .search("body", "database schema")
  .fetch()

This checks if the body field contains the string "database schema" anywhere — it does not use the tokenized FTS index.

Indexed search (FTSIndexer)

To use the inverted index directly:

import { FTSIndexer } from "ctrodb"

const indexer = new FTSIndexer(adapter)
const ids = await indexer.search("articles", "database schema")
// ids contains IDs of documents matching ALL tokens (AND logic)

FTSIndexer.search() tokenizes the query and returns only document IDs that contain every token. This is true tokenized AND search.

Built-in stop words

Common English words are excluded: a, an, and, are, as, at, be, but, by, for, if, in, into, is, it, no, not, of, on, or, such, that, the, their, then, there, these, they, this, to, was, will, with.

Tokenizer

import { tokenize } from "ctrodb"

tokenize("Hello World!") // ["hello", "world"]

The tokenizer splits on non-alphanumeric characters and lowercases. Duplicates are removed. Stop words are filtered.