liquidjs/docs/source/tutorials/caching.md

title

title
Caching

In a typical website project, we'll have a directory of view templates and they'll be rendered multiple times. In production environment the template files are not likely to be changed over time (other than re-deployments). Thus it makes sense to cache the file contents and the parsed templates (in a kind of AST) to improve performance.

LiquidJS provides multiple ways to cache the parsed templates to improve performance.

Programmatically

The .parse(), .parseFile(), .parseFileSync() APIs are used to parse templates from string or files. The result template can be then rendered multiple times with different context.

Parse from string:

var tpl = engine.parse('{{name | capitalize}}');

engine.renderSync(tpl, {name: 'alice'}) // 'Alice'
engine.renderSync(tpl, {name: 'bob'}) // 'Bob'

Parse from file:

var tpl = engine.parseFileSync('hello');    // contents of `hello.liquid`: {{name}}

engine.renderSync(tpl, {name: 'alice'}) // 'Alice'
engine.renderSync(tpl, {name: 'bob'}) // 'Bob'

The template string/file is parsed only once and rendered multiple times using different context. Templates for different files can be stored into a Map and can be retrieved directly for subsequent renders.

The cache Option

The cache option can be set to instruct liquidjs to use cached parsed templates each time you call renderFile or renderFileSync.

var { Liquid } = require('liquidjs');
var engine = new Liquid({
    cache: true
});

// liquidjs parses the hello.liquid, then renders it with {name: 'alice'}
engine.renderFileSync('hello', {name: 'alice'})

// liquidjs finds the cached template, then renders it with {name: 'bob'}
engine.renderFileSync('hello', {name: 'bob'})