api

API for Vigilante Web Heist

Installing postgres

This is geared towards Mac users, sorry in advance for any non-Mac environments. Anyways, this documentation is based on these amazing docs for installing and using Postgres, and the link covers Mac, Linux, and Window installations.

1. Install Postgres locally by following these instructions. (If using a Mac you will download the Postgres app)

2. Start up Postgres on port 5432 (if you have an error because there is already a process running on this port, you likely already have a previous version of postgres running. The following command should fix it: sudo pkill -u postgres).

3. Run psql to start a command-line client for interacting with Postgres (if you are having a command not found error and have a Mac, run the following command to link the command line tools with Postgres: sudo mkdir -p /etc/paths.d && echo /Applications/Postgres.app/Contents/Versions/latest/bin | sudo tee /etc/paths.d/postgresapp).

4. Now that you are in the psql client, run the following commands to create a gorm database along with a gorm user (based on this documentation).

   1. CREATE USER gorm WITH PASSWORD 'gorm';
   2. CREATE DATABASE gorm;
   3. GRANT ALL PRIVILEGES ON DATABASE gorm to gorm;

5. Now you should be able to build and run the API without any issues. At this point, I would suggest installing a Postgres GUI Client for interacting with the gorm database, such as Postico for Mac. Here is the official list of clients

How to run

Prerequisites:

• Golang is installed
• Docker is installed
• Postgres is installed (should come with psql too)
• Postgres client is installed (optional but recommended)

Steps to run:

With Golang

1. Run go build at the root directory to compile the executable from the file "main.go"
2. Run the executable named "api" (e.g. if using mac run ./api)

With Docker

Both a Dockerfile and a Dockerfile.prod are present for this service. The Dockerfile file is used to run the service locally, and the Dockerfile.prod is for a production environment, which needs a proxy alongside the api. To run just this image locally, use the following commands:

 docker build --tag api .      # Build the docker image
 docker run -p 8080:8080 api   # Run image as container on port 8080

Alternatively, You can set up the entire system using Docker and Docker-Compose. See more information and instructions in the csci4950tgt/utils repository.

How to submit to Google Cloud Run:

While the owners of this repository figure out how they want to integrate CI/CD with this service, the only current way to deploy the api as a Cloud Run Service is through the following commands. Note that this will only work if you have the correct permissions with Google Cloud, and the required secret.json file.

 docker build . -f Dockerfile.prod --tag gcr.io/vigilante-web-heist/api  # Build image with Google's Container Registry
 gcloud run deploy --image gcr.io/vigilante-web-heist/api --platform managed  # Deploy as Cloud Run Service

Steps to test:

1. Setup a testing user and a testing database using psql
   1. CREATE USER testing WITH PASSWORD 'testing';
   2. CREATE DATABASE testing;
   3. GRANT ALL PRIVILEGES ON DATABASE testing to testing;
2. Run go test ./... to test recursively (or go test ./... -v if you want verbose information)