A MySQL plugin that filters rows by meaning (built on TypeSafe Jev)

A native MySQL plugin for filtering rows and comparing text columns with natural-language conditions, powered by TypeSafe Jev.

Install and check compatibility · Try the Docker demo

Preview release. Requires a TypeSafe API key. Note: evaluated values and prompts are sent to TypeSafe.

Find columns based on NL meaning:

SELECT film_id, title, description
FROM sakila.film
WHERE film_id BETWEEN 1 AND 8
  AND description AILIKE 'The story takes place somewhere in Asia';

AILIKE returns 1 for a match and 0 otherwise. Any NULL argument returns NULL. Arguments are string. use CAST(... AS CHAR) for other types.

Narrow candidates with ordinary SQL conditions: each uncached evaluation makes an API request.

The demo dataset contains three supplier products and three catalog products. With AILIKE installed, open mysql -u root -p from the repository root and load it into a fresh database:

CREATE DATABASE ailike_join_demo;
USE ailike_join_demo;
SOURCE sql/join-demo.sql;

Pass both columns and describe the relationship:

SELECT a.id AS supplier_id, a.description AS supplier,
       b.id AS catalog_id, b.description AS catalog
FROM supplier_products AS a
JOIN catalog_products AS b
  ON a.category_id = b.category_id
 AND ailike(
   a.description,
   b.description,
   'Left and right describe the same kind of product, '
   'with matching material and key features. Wording may differ.'
 )
ORDER BY a.id, b.id;

Expected matches: (1, 1) for the steel bottle and vacuum flask, and (3, 3) for the headphones. The glass carafe and plastic bottle have no matching pair.

Format-independent date/period match #

SELECT rental_id, rental_date
FROM sakila.rental
WHERE customer_id = 1
  AND ailike(CAST(rental_date AS CHAR),
             'In July 2005')
ORDER BY rental_id;

AILIKE returns a model judgment. Use SQL date functions for exact comparisons.

Contributing · Report a vulnerability · AI coding agent skill · GPL-2.0-only