Skip to content

Latest commit

 

History

History
392 lines (283 loc) · 10.8 KB

README.md

File metadata and controls

392 lines (283 loc) · 10.8 KB

Modelist

Flexible & Customizable Modelstructure for awesome data management.

What is Modelist

Modelist is a client based collection model that makes handling data surprisingly convenient. It works great in reactive systems like Vue and React or flux stores like Vuex or Redux. Common tasks like recording, updating, deleting and selecting are made easy.

For Real-life examples checkout the Examples section.

Contributing

Contributions to this project are very welcome. Please read the Contributing.md document first.

Table of Contents:

  1. Installation
  2. Basic Usage
  3. The Entry Object
  4. Customize a Model
  5. Examples
  6. Meta

Installation

Modelist can be added to your project with ease.

npm install -D @codeship/modelist

# or

yarn add @codeship/modelist

Basic Usage

After adding modelist to your project, usage is quite simple.

import Model from "@codeship/modelist";

This gives you access to the Model contructor. That one can be used as is to have direct access to the everything Modelist brings to the table. Create a new instance, and start using it right away.

const users = new Model();

That's it. Nothing more that needs to be done to get access to a new collection and all the default methods that come with it.

Add entries to the collection

users.record({ id: 1, name: "Jane Doe" });

You can also record singe values like a Number or a String. Modelist will convert those into objects for you. -> Record standard Arrays

Get all entries of the collection

users.all();

Retrieve the first element of the collection

users.first();

Retrieve the last element of the collection

users.last();

Check if the collection has a record based on the primary key(default: id)

users.has(1);

Retrieve one specific element of the collection based on the primary key(default: id)

users.find(1);

Retrieve one specific element of the collection based on a custom key

users.findBy("name", "Jane Doe");

Update an entry based on the primary key Only the values that should be changed need to get passed in

users.update(1, { name: "Jane Goodall" });

Replace entries based on the primary key If you want to update multiple entries at once you can leverage replace. Notice though that this will really replace the element and not merge the existing values.

users.replace(1, { name: "Jane Goodall" });

Should replace be called on an element that can't be matched it will be recorded instead. If you want to record elements though, record is recommended as it's faster.

Remove an entry from the collection

users.destroy(1);

Get the number of elements in the collection

users.size;

$$reset This method will reset the Model and drop every data that's currently stored in the collection.

users.$$reset();

Internally all collection operations destroy, record and update use default Array methods, so reactive systems can track changes to the collection.

Single Entries

Usually there is something that needs to be done with an entry of the collection when it gets retrieved on it's own. Therfore Modelist will wrap that result in an `Entry1 object. Single entries aim to support functional composition. Entries come with default functionality that can be used to alter the value you received. Let's see this in action right away:

const fruits = new Model();
fruits.record({ id: 1, name: "Banana" });

// This will return the Entry object
const entry = fruits.find(1);

// Inspect a entry by using .toString() for easy debugging
entry.toString(); //= Entry({ id: 1, name: 'Banana' })

// By default an Entry comes with two functions
// map() to map over the current value and wrap the output in another Entry
// fold() to get the value out of the entry

entry.fold(); //= {id: 1, name: 'Banana'}

// Let's say you want only the name
entry.fold(e => e.name); //= 'Banana'

// What if you want to change the name before folding
entry.map(e => e.name.toUpperCase()).fold(); //= 'Banana'

// What happened here is, that we actually only took one value of the object
// and transformed it. We can still fold it afterwards as we receive a wrapped Entry.
// This gives you full control over the flow of what's happening with your entries

You can even add custom methods to your entries, so let's see how to configure the model.

Customize a Model

The Model comes with a bunch of preconfigured values that should make live easy and get you started quickly without making you jump through a bunch of hoops. But sometimes those values are not right, and that's why you can configure a model instance to your liking Let's stary easy by changing the primaryKey.

const products = new Model({
  primaryKey: "secret_id"
});

Now products will use secret_id as main identification for all the find, update and destroy methods. Here is a full list of available params:

{
  // Change the primaryKey
  primaryKey: 'custom',

  // Enforce a primary key to be set on each record by Modelist (uuid/v4)
  // default: false
  setPrimaryKey: true

  // Enforce that every entry is matched against the schema
  // default: false
  validate: false,

  // Schema can be defined where each entry will be matched against
  schema: {
    custom: 'String'
  },

  // Pass in data by default
  data: [...],

  // Add custom filters to the collection that will be made available
  // as getters on the root of the model.
  filters: {
    // i.e.: return all completed tasks of a task list
    // This will also give you access to the subset of the collection as a fresh Model instance
    completed: (collection) => collection.filter( item => item.done === true)
  },

  // Add methods to entries
  methods: {
    customMethod: (e) => {
      e.amount--
      return e
    },
  }
}

Extending the model with custom filters

Sometimes you want specific subsets of all the stored data of your Model. That's a perfect opportunity to add filters to your instance. Filters are custom methods that allow you to retrieve a subset of the passed in collection. The result is than passed down for you as a fresh Model instance.

Some caveat: The new instance gives you access to all core functions but without the filters. So it's not possible to call a filter method and chain another custom filter.

const tasks = new Model({
  data: [
    { task: "Go for a walk", done: false },
    { task: "Buy Milk", done: true },
    { taks: "Feed the cat", done: true }
  ],
  filters: {
    completed: c => c.filter(t => t.done === true)
  }
});

console.log(tasks.completed.size); //= 2
console.log(tasks.completed.all()); //= [{task: 'Buy Milk', done: true}, {taks: 'Feed the cat', done: true}]

Using custom methods for entries

If you passed in methods to the Model, they're made available for you on the Entry object. By default those methods will be applied as folds so they just return whatever the function returns.

const fruits = new Model({
  primaryKey: "name",

  data: [{ name: "Apple" }],

  methods: {
    getSaladName: e => `${e.name} Salad`
  }
});

fruits.find("Apple").getSaladName(); //= Apple Salad

But you can actually also map over the value and get another Entry should you want to do something else or chain various methods. In this case the method name needs to be passed as a String.

fruits
  .find("Apple")
  .map("getSaladName")
  .toString(); //= Entry('Apple Salad')

Record standard Arrays

In some instances the data is not prepared as an array of objects, but simple values. Modelist allows you to record simple values alongside of objects. If you pass in an simple value, it will be transformed into an object with an unique ID based on the primaryKey and the value stored alongside it.

model.record(...[1, 'fred'])

model.all() // [{ id: uuid(), value: 1 }, { id: uuid(), value: 'fred' }]

Schemas

Schemas allow defining how the root data structure of each data sent into the collection should look like. It support primiteves like Number, String, Boolean, Object, Array or event custom types.

To leverage the schema validation, just add the validation key to your Model. For example you can successfully match this data:

const data = {
  id: 1,
  name: "Jane",
  favs: ["a", "b"],
  settings: {},
  custom: new Custom()
};

against this schema

const schema = {
  id: Number,
  name: String,
  favs: Array,
  settings: Object,
  custom: Custom
};

If something is not fitting that schema it will print out a warning, but still try to process the data.

Examples

Use Modelist within in Vuex

Let's see a simple example how to use Modelist within in Vuex. This makes mutations, and actions way more convenient, but can also make your components a little more structured.

import Model from "@codeship/modelist";
import uuid from "uuid/v1";

export default {
  state: {
    tasks: new Model()
  },

  mutations: {
    recordTask(state, entry) {
      state.tasks.record(entry);
    },

    updateTask(state, id, data) {
      state.tasks.update(id, data);
    }
  },

  actions: {
    addTask({ commit }, task) {
      const newTask = { id: uuid(), body: task, done: false };
      commit("recordTask", newTask);
    },

    completeTask({ commit }, task) {
      commit("updateTask", task.id, { done: true });
    },

    changeTask({ commit }, task, body) {
      commit("updateTask", task.id, { body });
    }
  },

  getters: {
    tasks: state => state.tasks
  }
};

Eventually you can structure the store in a very convenient way and, see that the updateTask mutation can actually be used for two different actions. For the getter, it's very easy to just return the whole collection and give the components access to it's internal functions if you like. On the client this could look like this:

<ul>
  <task v-for="task in tasks.all()" :task="task">
</ul>

The important piece here is, that by using the whole task collection, task will be the plain object for your convenience.

Meta

This Project is managed by Roman Kuba rkuba@cloudbees.com from the Codeship Team. You can best reach me on Twitter @Codebryo.