sql.js is a port of SQLite to JavaScript, by compiling the SQLite C code with Emscripten.

Demo

SQLite is public domain, sql.js is MIT licensed.

See demo.html for an example use, and see test.js and the comments inside for another example including the output you will receive.

The API is as follows:

• SQL.open() creates a new database, returning a database object

• SQL.open(data) creates a new database with given data, which should be a typed array of 8-bit values (typically generated from calling exportData, see below)

Database objects (created from SQL.open) have the following methods:

• .exec(command) runs a command in the database, returning JSON output

• .close() closes the database (this frees the memory it uses)

• .exportData() serializes the data to a typed array of 8-bit values, which you can save using any method you like (localStorage, indexedDB, send to a remote server, etc.), and later re-use by calling SQL.open with that data.

Note that the output rows from .exec(..) look like

    [{ "column": "a", "value": "1" }, { "column": "b", "value": "13153" }]

whereas in theory they could look like

    { a: "1", b: "13153" }

The reason for the more verbose format is that it preserves the order of columns. It also prevents problems with column names stepping on special JS property names. However, we should probably make it an option to get the other format, pull requests welcome.

Note that sql.js is not wrapped in a closure - it modifies the global scope. This is done because when wrapped in a closure it becomes slower in most JS engines. To be safe from global scope problems, you should either make sure your code doesn't use the global scope, or wrap sql.js in a closure, something like

var SQL = (function() { {{{ paste sql.js here }}} return SQL; })();

Another option is to run sql.js in a web worker.