denormalize-mongoose v0.0.4

denormalize-mongoose

denormalize-mongoose is a bidirectional denormalize plugin for Mongoose from version 4.4

Description

Plugin has to be used with a custom Type for Mongoose Schema called Denormalize, that have options to config how data have to be denormalized.\ It adds pre save handlers to your Schema and can add post remove handler (optional)

Installation

Plugin can be installed globally once from the library, but it needed to be used before mongoose initialization:

require('denormalize-mongoose')();
const mongoose = require('mongoose');

Or can be installed manually to every needed schema:

const { Denormalize, denormalizePlugin } = require('denormalize-mongoose');
const mongoose = require('mongoose');

const schema = new mongoose.Schema({
  name: String,
  post: {
    type: Denormalize,
    paths: ['createdAt', 'title', 'description'],
  },
});

If you use middlewares in Schema - manually initialization of plugin is recommended after all middlewares you want to add.

Options

paths - main option, it defines what fields will be taken from original Document during denormalize process.\ It can be an Array of strings or Object, also it can be combined.

const { Denormalize, denormalizePlugin } = require('denormalize-mongoose');
const { Schema, SchemaTypes } = require('mongoose');

const userSchema = new Schema({
  firstName: String,
  lastName: String,
  company: String,
  comments: [{
    type: Denormalize,
    of: SchemaTypes.ObjectId,
    paths: ['text'],
    ref: 'Comment',
  }],
  commentsData: [{
    text: String,
  }]
});

userSchema.virtual('fullName').get(function getFullName() {
  return `${this.firstName} ${this.lastName}`;
});

const commentSchema = new Schema({
  user: {
    type: Denormalize,
    of: SchemaTypes.ObjectId,
    paths: ['fullName', 'company'],
    ref: 'User',
  },
  userData: {
    fullName: String,
    company: String,
  },
  text: String,
});

userSchema.plugin(denormalizePlugin);
commentSchema.plugin(denormalizePlugin);

const User = model('User', userSchema);
const Comment = model('Comment', commentSchema);

As you can see from Schema, plugin can take even virtual fields from other Documents, it's because it populates all the Denormalize fields to Document, so all operations are running on Documents

suffix and to options doing similar things. They represent field of document where to denormalize data.\ to - is full path to Object or Array of Objects field to denormalize data, if it defines - suffix will be ignored\ suffix - string that will be added to name of field where Denormalize Type. That joined string will be used to denormalize data (as with to option), 'Data' by default

const userSchema = new Schema({
  comments: [{
    type: Denormalize,
    of: SchemaTypes.ObjectId,
    suffix: 'SomeEnding',
    paths: ['text'],
    ref: 'Comment',
  }],
  commentsSomeEnding: [{
    text: String,
  }]
});

is the same as:

const userSchema = new Schema({
  comments: [{
    type: Denormalize,
    of: SchemaTypes.ObjectId,
    to: 'commentsSomeEnding',
    paths: ['text'],
    ref: 'Comment',
  }],
  commentsSomeEnding: [{
    text: String,
  }]
});

of - option represents Type of data used for Document id, can be everything that Mongoose supports: ObjectId, String, Number, like it will be stored in DB. ObjectId by default

ref - option is using in the same way as in Mongoose docs, so it can be used for population data too.

Additional

Denormalize operation can also be used with nested fields, so paths can be an Object that represents paths inside Document:

const userSchema = new Schema({
  comments: [{
    type: Denormalize,
    of: SchemaTypes.ObjectId,
    to: 'commentsSomeEnding',
    paths: {
      text: {
        title: 1,
        desctiption: 'info' // Will rename field in denormalized data
      }
    },
    ref: 'Comment',
  }],
  commentsSomeEnding: [{
    text: String,
  }]
});