Indicative Parser

Converts indicative rules and messages schema to a tree

Indicative parser pre-compiles the Indicative schema to a recursive tree of nodes. Each node is given one of the following types.

• object: Node with one or more nested children.
• array: Node with one or more index or wildcard based nested children.
• literal: The leaf nodes.

Do note, that the literal type is not equal to literal values in Javascript. For parser, the literal nodes are nodes with no leaf.

Table of contents

• Why Indicative needs a parser?
• Usage

Why Indicative needs a parser?

If you look at the Indicative schema, it is very concise and developer friendly. However, the same schema needs to be parsed to execute the validation rules.

{
  username: 'required',
  'account.type': 'required|in:email,social'
}

One way is to loop over the schema object keys, split them by . and then inline execute the validations for each field. This process is very straight forward, but will have performance issues.

Instead, we parse the schema into a tree. The tree is later converted to an array of top level functions that are highly optimized for performance.

Usage

Install the package from npm registry as follows:

npm i indicative-parser

# yarn
yarn add indicative-parser

and then use it as follows:

import { rulesParser } from 'indicative-parser'

rulesParser({
  username: 'required',
  'account.type': 'required|in:email,social'
})

Above code outputs the following tree.

{
  "username": {
    "type": "literal",
    "rules": [
      {
        "name": "required",
        "args": []
      }
    ]
  },
  "account": {
    "rules": [],
    "type": "object",
    "children": {
      "type": {
        "type": "literal",
        "rules": [
          {
            "name": "required",
            "args": []
          },
          {
            "name": "in",
            "args": [
              "email",
              "social"
            ]
          }
        ]
      }
    }
  }
}