2

GitHub - f/honst: Fixes your dataset according to your rules.

 3 years ago
source link: https://github.com/f/honst
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client

README.md

chains honst

The name: I wanted it to be "honest", but it was occupied, so I removed a letter, and it became "honst".

Data Integrity fixer for an object matrix.

Use Cases

  • On editing a joined matrix data needs the data integrity to be kept.
  • Keeping your local state correct.

You can play with honst on CodeSandbox:

Overview

const data = [
  { username: "johndoe", name: "John", surname: "Doe", age: "22" },
  { username: "johndoe", name: "John", surname: "Doe", age: "25" },
  { username: "johndoe2", name: "John", surname: "Doe", age: "22" },
  { username: "johndoe2", name: "John", surname: "Doez", age: "22" },
]

According to username field, there are many integrity issues around the array. So let honst fix these:

[
  { username: "johndoe", name: "John", surname: "Doe", age: 22 },
-  { username: "johndoe", name: "John", surname: "Doe", age: 25 },
+  { username: "johndoe", name: "John", surname: "Doe", age: 22 },
  { username: "johnapple", name: "John", surname: "Apple", age: 29 },
- { username: "johnapple", name: "John", surname: "Orange", age: 22 },
+ { username: "johnapple", name: "John", surname: "Apple", age: 29 },
]
  1. Referencing to data[0].username, name, surname, and age should be John, Doe and 22.
  2. But data[1].age is 25 and it should be fixed.
  3. Referencing to data[3].username, name, surname, and age should be John, Apple and 29.
  4. But data[1].age is 25, data[1].surname is Orange and these should be fixed as well.

honst simply fixes these integirty issues.

Usage

Install using npm or yarn:

npm install honst
# or
yarn add honst

Now you can start:

import { honst } from 'honst';

const { data, delta } = honst({
  data: [
    { username: "johndoe", name: "John", surname: "Doe", age: "22" },
    { username: "johndoe", name: "John", surname: "Doe", age: "25" }
  ],
  pivot: 0,     // an index number, "scan" or "reverse-scan"
  delta: true,  // boolean
  rules: {
    // the ruleset of integrity
    // e.g. we want all names, surnames and ages to be same according to "username"
    name: ["username"],
    surname: ["username"],
    age: ["username"],
  }
})

This will generate following data:

const fixedData = {
  data: [
    { username: "johndoe", name: "John", surname: "Doe", age: "22" },
    { username: "johndoe", name: "John", surname: "Doe", age: "22" },
  ],
  delta: [
    {
      candidatePath: "age", // the path of the field
      falseValue: "25", // the wrong value
      candidateValue: "22", // the candidate value
      pivot: 0, // the pivot row index
      candidateIndex: 1, // the false row index
    }
  ]
}

Scanning and Reverse Scanning

Just pass pivot value scan or reverse-scan.

If you want to fix all the rows from top-down or bottom-up, this will scan the data and fix'em all.

const { data, delta } = honst({
  data: [
    { username: "johndoe", name: "John", surname: "Doe", age: "22" },
    { username: "johndoe", name: "John", surname: "Doe", age: "25" },
    { username: "johndoe2", name: "John", surname: "Doe", age: "22" },
    { username: "johndoe2", name: "John", surname: "Doez", age: "22" },
  ],
  pivot: "scan",
  delta: true,
  rules: {
    name: ["username"],
    surname: ["username"],
    age: ["username"],
  }
})

This will generate the following:

const fixedData = [
  { username: "johndoe", name: "John", surname: "Doe", age: "22" },
  { username: "johndoe", name: "John", surname: "Doe", age: "22" },
  { username: "johndoe2", name: "John", surname: "Appleseed", age: "22" },
  { username: "johndoe2", name: "John", surname: "Appleseed", age: "22" },
]

If you select scan the upper rows will be fixed first, and matching rows below will be updated accordingly. If you select reverse-scan it'll start fixing from bottom and scan to the top, so the bottom rows will be assumed correct.

Nested Objects?

Yes it supports nested objects as well.

const { data, delta } = honst({
  data: [
    // ... some nested data
  ],
  rules: {
    'user.name': ["account.username"],
    'user.surname': ["account.username"],
    'profile.age': ["account.username"],
  }
})

Contribution

  • Clone and edit the source as you wish.
  • Please do not forget to add tests.
  • Write a descriptive PR.

License

This project uses MIT license.


report

Recommend

About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK