A restaurant app for fueled (test)

During lunchtime, the members of Fueled are faced with the problem of having too many restaurant venues to choose from. Make a Django app that helps us decide where the team should dine. Additionally the app should have the following

To deploy this to azure use this link.

• Keep track of visited restaurants.
• Ability to read/write reviews for a visited restaurant. Reviews are displayed on the restaurant page. A special * symbol is displayed if it is your review.
• Other users can comment on reviews (no need for nested comments).
• Ability to thumbs-down a restaurant such that it is never considered as an option for dining.
• Extend the user profile and add some relevant functionality.

The system is basically a Back-end application built on Django. There's also a Front-end app for testing and demostration purposes.

The back end is Django Site with a "restaurant" app that serves the different endpoints. It also provides a landing page that exposes the differents parts and links of the system.

The client is a one page application built on Angular + Coffescript.

• Django and Python are used in the backend.
• Django Rest Framework is used to build the API.
• Django Swagger was installed to provide a better documentation for the API.
• Django-vote is used to manage the thumbs down functionality.
• Factory-boy is used to simplify the testing of the models, and it can be reused in the future for performance tests.
• Factual app is used as an adapter to communicate with Factual.com API and it retrieves the restaurants (see section "Import Restaurants into the system" for more information about why using Factual).
• Yeoman was used to start the frontend app.
• Angular and coffee script are used to code the app.
• Grunt was used during the development of the frontend app, and also for the distribution of it.

See demo video here: http://youtu.be/Mn78aR2Hxho

The system has Django Back-end, a landing page served by Django and a static app that use the endpoints provided by the backend. To start using the system you can visit the landing page served by Django at:

http://localhost:8000

There you will find links to navigate to the other pages and apps

Standard Django Admin to mange models and create users

Search Page of restaurants

Restaurants detail page

In order to use the test site you need to be logged in (if not you won't be able to modify any data). To do that you can use the links provided on the site or use this link: http://localhost:8000/api-auth/login/?next=/

In the test pages users can:

• Login / Logout
• See a list of restaurants (paginated by 10)
• Search by type of cuisine and name of the restaurant
• Order restaurant list by ID, name, Thumbs down, and rating. To do this a user needs to click on the table head, if a user clicks again the ordering is removed. Thumbs down column is order incrementally, so if a user click on it he won't see Restaurants with Thumbs down at the top.
• See the details of a restaurant. If a user click on the name of a restaurant, the page will scroll down and the details of the restaurant will be shown.
• In the details of a restaurant a user can:
• Thumbs Down/Up the restaurant (the Up only works as an undo of the thumbs down)
• Mark the restaurant as vistied
• Leave a comment (the comment will be displayed at the top of the comment list)

For this installation procedure we will assume that you have the following packages already installed and configured on your system: python, pip, virtualenvwrapper

Create a virtual env with virtualenvwrapper

$ mkvirtualenv resto

Activate the environment:

$ workon resto

Clone/Download this repo:

$ git clone git@github.com:rafen/resto.git

Enter to the just clonned repo:

$ cd resto

Install requirements by entering to the repo folder and run

$ pip install -r requirements.txt

Initialize your local database:

$ ./manage.py migrate

Create a superuser:

$ ./manage.py createsuperuser

Enter username, email and password, as needed.

Import Restaurants (This command will only import 500 restaurants to the system, after that it will raise an error, see section "Import Restaurants into the system" for more info):

$ ./manage.py import_restaurants

Run a local test server:

$ ./manage.py runserver

Visit the backend test landing page by opening your browser at

http://localhost:8000

To run the backend test just type:

$ ./manage.py test

All API endpoints are created with django_rest_framework. It provides an easy solid way to implement API. It also has as self documented doc to instrospect all endpoints. Endpoints were designed to minimize the amount of requests made to the backend. The main endpoint is /restaurants/restaurant/ that will return a page (of 10 elements) with almost all the information needed to display a list of restaurants, with visitor comments, etc The API also has other endpoint to create comments, votes, visits, etc. Almost all of them are based on the Django Model which reduce the amount of code and logic on the system.

Factual.com is used to get the restaurant information. I considered the problem of getting the initial data, processing and sorting beyond the scope of this test. Using a third party API was the best solution in order to focus on the problem of commenting, vote, thumbs down for restaurant which was explicitly asked by the stake holders.

For future references: The list of restaurant retreive from Factual.com is sorted by "placerank": http://developer.factual.com/search-placerank-and-boost/#placerank

In resto-ui folder you will find the frontend app to test the application. The distribution folder of the front-end app is already included and linked in the static folder of the repo, but if you need to work on it, you can do this: To install it and run it you need to:

Enter to the folder:

$ cd resto-ui

Install dependencies:

$ npm install
$ bower install

You might need to upgrade your compass gem. In that case run:

$ sudo gem install compass --pre

Then run the app with grunt:

$ grunt serve

Note: you need to have the Django App running on a different terminal to see the Front-end app working.

• Add celery task to the list of restaurants updated using the factual API
• Import georefernces of restaurants and expose those fields in the API
• Include GeoDjango to the project to make GeoSpatials queries and found the closest restaruants
• Build a good Front-end application, with:
• with better look & feel
• Unit and functional tests
• Add Login/Logout
• Add functionality to edit/delete comments, visits, etc.
• Add Google Map to show restaurants location in the details.
• Add GeoSpatials search.
• Deployment
• Etc