Moved

** FluentDB has been renamed to fluent-data. **

If you are here, search for fluent-data in npmjs or github.

Introduction

** I am close to release of Version 1! Go to The Future and check out 'Before Version 1 Release'. **

Manipulate datasets by chaining methods. Includes capacity to map, filter, sort, group, reduce, and merge data.

FluentDB works like many of the methods on Array.prototype. However, FluentDB makes it much easier to work with arrays when their elements are objects. It also includes methods simply not available on Array.prototype.

FluentDB syntax is similar to LINQ in c#. C# developers frustrated with the lack of a LINQ functionality in javascript may be encouraged by FluentDB. Some of the syntax can even be friendlier and more powerful in comparison.

Getting Started

To install:

npm install FluentDB 

To import:

// client
import $$ from './node_modules/FluentDB/dist/FluentDB.client.js';

// server
let $$ = require('FluentDB');

// but the examples in this documentation will use
let $$ = require('./dist/FluentDB.server.js');

Example:

Consider these datasets:

let customers = [
    { id: 1, name: 'Alice' },
    { id: 2, name: 'Benny' } 
];

let purchases = [
    { customer: 2, speed: 15, rating: 50, storeId: 1 },
    { customer: 1, speed: 5, rating: 90, storeId: 1 },
    { customer: 1, speed: 7, rating: 55, storeId: 1 },
    { customer: 2, speed: 6, rating: 88, storeId: 1 },
    { customer: 1, speed: 25, rating: 35, storeId: 1 },
    { customer: 1, speed: 40, rating: 2, storeId: 3, closed: true },
    { customer: 2, speed: 4, rating: 88, storeId: 1 },
    { customer: 1, speed: 1, rating: 96, storeId: 2 },
    { customer: 1, speed: 2, rating: 94, storeId: 2 },
    { customer: 1, speed: 1, rating: 94, storeId: 2 }
];

The following exmaple uses many of the methods available to analyze the two datasets.

let $$ = require('./dist/FluentDB.server.js');

let result = 
    $$(purchases)
    .filter(p => !p.closed)
    .merge(customers, (p,c) => p.customer == c.id, 'both null') // inner join
    .group(p => [p.customer, p.storeId]) 
    .reduce(p => ({
        customer: $$.first(p.name),
        store: $$.first(p.storeId),
        orders: $$.count(p.id), 
        speed: $$.avg(p.speed),
        rating: $$.avg(p.rating),
        correlation: $$.cor(p.speed, p.rating)
    }))
    .sort(p => [p.customer, -p.rating])
    .get(p => ({
        ...p, 
        speed: $$.round(p.speed, 2),
        rating: $$.round(p.rating, 2),
        orders: undefined // won't show in final results
    }));

console.log(result);

This results in three rows for analysis:

[
  {
    customer: 'Alice',
    store: 2,
    speed: 1.33,
    rating: 94.67,
    correlation: -0.5
  },
  {
    customer: 'Alice',
    store: 1,
    speed: 12.33,
    rating: 60,
    correlation: -0.8315708645692353
  },
  {
    customer: 'Benny',
    store: 1,
    speed: 8.33,
    rating: 75.33,
    correlation: -0.9853292781642932
  }
]

Operations and Features

The following operations are available on FluentDB:

• get: Returns the dataset as an array.
• map: Replaces each row in a dataset with the result of a function called on each row.
• filter: Chooses particular rows from a dataset.
• sort: Sorts a dataset.
• distinct: Eliminates duplicates in a dataset.
• merge: Brings in values from another set of data. Can be done horizontally (such as with a join) or vertically (such as with an insert).
• group: Group rows of a dataset into nested datasets. Or reverse this with ungroup
• reduce: Aggregate a dataset. Create custom aggregators with reducer.
• with: Work with a dataset without breaking the fluency/chaining syntax.

Click on the links to go to the wiki and learn more about them.