Skip to content

Latest commit

 

History

History
152 lines (97 loc) · 4.33 KB

README.md

File metadata and controls

152 lines (97 loc) · 4.33 KB

jsonscript-js

JavaScript interpreter for JSONScript

Build Status npm version Code Climate Coverage Status

Install

npm install jsonscript-js

Getting started

Sequential execution of script commands:

var JSONScript = require('jsonscript-js');
var js = new JSONScript;

js.addExecutor('router', getRouter());

var script = [
  { '$$router': { path: '/resource/1' } },
  { '$$router.put': { path: '/resource/1', body: { test: 'test' } } }
];

js.evaluate(script).then(function (res) {
  console.log(res);
  /**
   *  [
   *    { response: 'loaded /resource/1' },
   *    { response: 'updated /resource/1 with {"test":"test"}' }
   *  ]
   */
});


function getRouter() {
  function router(args) {
    return router.get(args);
  }

  router.get = function (args) {
    var response = 'loaded ' + args.path;
    return Promise.resolve({ response: response });
  };

  router.put = function (args) {
    var body = JSON.stringify(args.body);
    var response = 'updated ' + args.path + ' with ' + body;
    return Promise.resolve({ response: response });
  };

  return router;
}

In the example above the second request is sent only after the first result is received, so you can both get the current resource value and and update it in one script call.

Parallel execution:

var script = {
  res1: { '$$router.get': { path: '/resource/1' } },
  res2: { '$$router.get': { path: '/resource/2' } }
};

js.evaluate(script).then(function (res) {
  console.log(res);
  /**
   *  {
   *    res1: { response: 'loaded /resource/1' },
   *    res2: { response: 'loaded /resource/2' }
   *  }
   */
});

In the example above the second request is sent in parallel, without waiting for the response from the first request.

API

new JSONScript(Object options) -> Object

Create JSONScript interpreter instance.

.validate(Any script) -> Boolean

Validate script. This method is called automatically before the script is evaluated.

.evaluate(Any script, Any data) -> Promise

Evaluate script. Returns Promise that resolves to the script evaluation result or rejects in case of validation or executor error.

.addExecutor(String name, Function|Object executor)

Add executor to the interpreter. Can be an object or a function with methods.

This method throws exception if you use the name that is already used (including predefined executors array, calc and str).

.addInstruction(Object definition, Function func)

Define JSONScript instruction. Core instructions are added using this method too.

definition should be valid according to the instruction schema.

func is the function used to evaluate instruction, it can return:

  • Promise that resolves to the evaluation result
  • instance of Script that can contain:
    • a script that should be evaluated
    • a Promise that resolves to a script (for delayed evaluation).

Class Script is available as the property of both the class and the instance of JSONScript interpreter.

.addMacro(Object definition)

Define macro. Core macros are added using this method too.

definition should be valid according to the macro schema.

Options

Defaults:

{
  strict: false,
  executors: {}
}
  • strict: use strict JSONScript schema (see schemas). Strict schema validates instruction keyword values if they are defined as constant so it would fail faster if the script is invalid and no instructions will be executed, but the validation itself takes longer.
  • executors: an object with executors (keys are names).

Language

See JSONScript language documentation for more information.