https://github.com/garnetred/loc-collective-backend
This is the back-end for an application that allows users to search more easily for locticians by querying the Yelp API.
https://github.com/garnetred/loc-collective-backend
natural-hair yelp yelp-api
Last synced: about 2 months ago
JSON representation
This is the back-end for an application that allows users to search more easily for locticians by querying the Yelp API.
- Host: GitHub
- URL: https://github.com/garnetred/loc-collective-backend
- Owner: garnetred
- Created: 2022-07-29T19:20:16.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2023-11-06T06:09:52.000Z (almost 2 years ago)
- Last Synced: 2025-01-05T04:33:07.460Z (9 months ago)
- Topics: natural-hair, yelp, yelp-api
- Language: JavaScript
- Homepage:
- Size: 269 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# loc-collective-backend
# About
This is the API to the front end application [Loc Collective](https://github.com/garnetred/loc-collective).
# Background
[Loc Collective](loccollective.com) is a directory of locticians that allows users to search for locticians based on hair style and location. This works by consuming the Yelp API.
I decided to create this project as a bootcamp student due to the difficulty of finding a loctician as a person with locs.
I created this API specifically because the API key was exposed on the front-end. This codebase takes requests from the front-end and forwards them to the Yelp API while tacking on the API key in order to make it more secure.
# Setup
You can clone down the repo by running `git clone` and the SSH url for this repo in your terminal. From there, you can run `npm install` to install any dependencies and dev dependencies.
To run the server locally, you can use the command `npm run dev`.
# Testing
To run tests, you can use the commands `npm test` or `jest`. (If you use `jest`, you will need to manually exit out of the tests or use the `--forceExit` flag.)
# Tech Stack
The following technologies are used for this repository:
- NodeJS
- Express
- Supertest
- Nock
- Jest
- Render
# The Why's Behind the Tech Stack
## Why Node?
I chose Node primarily because it uses JavaScript. Since all I needed was a small API to hide an API key, it didn't make sense for me to use something more robust such as Ruby on Rails.
## Why Supertest and Nock?
I tried testing with jest-mock and found that it just seemed overly complicated for what I was trying to accomplish.
## Why Jest?
I like the matcher functions that Jest has and the extended matcher functions using Jest-DOM. Since it's so popular it's easy to troubleshoot. I've used other test suites in the past but personally prefer Jest.
## Why Render?
I chose Render primarily because Heroku ended their free tier, but I still wanted to be able to use my site. I attempted to migrate it to Vercel but since Vercel uses serverless functions it would've taken a bit more work than was logical for a side project. I was able to get set up with Render within ten minutes and it has served my purposes well, despite the throttling on the free tier that slows down all network calls.
# API Endpoints
There are currently three API endpoints, and they are as follows:
- `/api/search`
- This endpoint is sent a `term` (such as "sisterlocks" or "interlocks") and a `location` in the US (ex. "Boston" or "90210") and then passes that information on to Yelp.
- `/api/stylist/:id`
- This endpoint takes in a stylist id, usually when a user navigates to a stylist's page from the list of search results.
- `/api/reviews/:id`
- Similar to the endpoint above, it takes in a stylist id in order to shows reviews for a particular stylist.
# Deployment
This codebase has continuous deployment via Render, so once pull requests are merged to main, they are deployed automatically.
# Next Steps
I might migrate the front-end of this application from React to NextJS. There are known issues with single-page applications and SEO that I'm currently experiencing. The site shows up in Google searches, but not under the search terms I would like. Using server-side rendering seems like a great way to fix this problem.
At some point I hope to add mobile responsiveness to the front-end as well. I created additional mobile and tablet designs via Figma but need to build them.
I would also like to improve error handling on the front-end side.
I might add Github Actions to this repo to keep track of what tests are failing and when.