Introduction

This is an example Proof of Concept (POC) of using React/ES6 Modules and Algolia. I previously created a version of this application based on JQuery and Hogan templates. I removed these dependencies and rewrote the app teaching myself/using React, Algolia, ES6 Modules, modular SASS, and Webpack. The UX was simplified.

There are many things I could improve with more time, but I hope you enjoy the project as it is!

Instructions for Installation

Install Dependencies

node.js (version 7.1.0 was used for development)
ruby (version 2.3.0 was used for development)(only needed if you want to implement this POC in your own Algolia index)

Download

Clone the git repository (using e.g: git clone https://github.com/whartonn/algolia_with_react.git) to your local machine.
Run npm install to download the dependencies (from the project root)
Or run npm run watch

If you want to install from scratch in your own environment do the following:

Index Algolia

Open your terminal to the root of the project and execute ruby combine_and_export_as_json.rb.
The program will generate a new file, export_for_algolia.json in the root of the project.
import export_for_algolia.json to a new algolia index using Algolia's web interface.

Configure Algolia

Name the new index poc_restaurants (for the proof of concept restaurants listing)
Under the index tab, click on "ranking" and set "Searchable Attributes" to name, food_type, neighborhood, area, city.
Set "Custom Ranking Attributes" to name, and stars_count.
Under the index tab, click on "Display" and set "Attributes for Faceting" to foot_type and stars_count.

Configure Your HTML

edit the assets/js/CustomSettings.js file and set ALGOLIA_SETTINGS.APPLICATION_ID and ALGOLIA_SETTINGS.SEARCH_ONLY_API_KEY to Algolia's settings.

Start Node

Run npm install to download the dependencies
Then run npm run server to create a server, and automatically refresh the page every time on of the *.scss, *.less, *.html files get updated.
Or run npm run watch if you only want to run the server

Type!

When visiting http://localhost:3000 in your browser

Enter 'a' or anything in the search bar - and you'll start seeing results. If you have any questions, do not hesitate to contact me!

Original Specification (for reference) (Source: Algolia):

Our sales team has recently been contacted by a large restaurant reservation website, for who it as identified as strategic to build a demo. As an Engineer, you're asked to build a small prototype that, using the dataset and UI they have provided us, highlights the benefits of a great search experience.

Push the provided dataset to an Algolia index
Produce the HTML markup and CSS needed to reproduce the UI provided by the client. To do so, you can write with vanilla CSS or a processor of your choice. We've provided configs for Sass and LESS.
Using the Algolia JS Helper (documentation), implement an as-you-type search experience that enables users to easily find restaurants: both by passing a search query and/or filtering on the "type of cuisine".

Screenshot of the demo once completed

Important Notes

Graphical resources, including the Sketch mockup, are provided in the ./resources folder
The dataset given by the client is available in the ./resources/dataset folder. They have been able to extract 5000 restaurants from their database: restaurants_list.json. Unfortunately, because of some system complexity on their side, they haven't been able to provide everything in one file only. They sent us another file called restaurants_info.csv that contains additional information for all the extracted restaurants.
- You'll need to manipulate both data files in order to have access to the "type of cuisine".
The project is already all setup for you, if you're using Sass, LESS you shouldn't have to configure anything
- Run npm install to download the dependencies
- Then run npm run dev to create a server, and automatically refresh the page every time on of the *.scss, *.less, *.html files get updated.
- Or run npm run server if you only want to run the server
Feel free to use any front-end tooling with which you're the most confortable.
The blue highlight in the sidebar is an active / over state.

Evaluation criteria:

The goal of this test is to evaluate your ability to build a front-end implementation using Algolia.

Criterion

Overall look'n'feel of the demo: relevance, UI & UX
Quality of the code: legibility and re-usability
Attention given to details

Deliverable:

Create a Github public repository to host the code (including the import script if you have one)
Publish it using GitHub's gh-pages so we can interact with it

Note: The provided dataset has been created using https://github.com/sosedoff/opentable project.

whartonn/algolia_with_react

Algolia with React