codevideo_generator_backend

The backend engine of codevideo.io used to generate code videos.

Run Locally

First, ensure you have a .env file in the root. You can use the .env.example file as a template:

cp .env.example .env

Make sure to define all values there. To save you (and us) headaches, you should also copy the .env file to the deployments directory:

cp .env deployments/.env

(We're working on sourcing everything from the root .env file.)

With your env setup, install dependencies:

npm install

Run the server:

npm start

The server should be accessible at http://localhost:<<PORT>>, where <<PORT>> is the value of the PORT environment variable you've selected.

On Premise / Local Docker Deployment

codevideo_generator_backend also ships as a completely standalone docker image. To build and run it, run the following command:

source .env &&
docker build -t codevideo_generator_backend --build-arg port=$PORT -f deployments/Dockerfile.local . &&
docker run -d -p $PORT:$PORT codevideo_generator_backend 

The container codevideo_generator_backend should then be up and running, accessible at http://localhost:<<PORT>>, where <<PORT>> is the value of the PORT environment variable you've set in .env. It's up to you to configure further network stuff such as a reverse proxy, SSL, etc. Note the difference here from AWS deployment in that the serverless handler is not the entrypoint of the container, but rather just the npm start command.

Local AWS Deployment

If you want to deploy the AWS docker image locally, you can do so by running the following command:

source .env &&
docker build -t codevideo_generator_backend --build-arg port=$PORT -f deployments/Dockerfile.aws . &&
docker run -d -p $PORT:$PORT codevideo_generator_backend

AWS Deployment

AWS tooling is more than... elegant. We are hoping to get an AWS lambda deploy working. Until then, see the complete nginx stack deploy section below

AWS deployment is managed by pulumi within the deployments directory. It handles truly everything, from user creation, roles, and so on. To get started, first ensure pulumi is installed:

brew install pulumi/tap/pulumi

Then ensure you have a .env file in the deployments directory. You can use the .env.example file as a template:

cp deployments/.env.example deployments/.env

Then, make sure to define all values there.

You should then be able to run the following commands to deploy the app:

cd deployments &&
source .env && 
pulumi up --yes

If all goes well, you should see pulumi begin to deploy your stack.

Complete Nginx Stack Deploy

If you want to deploy the entire stack, including nginx, you can do so by running the following command:

source .env && docker-compose -f deployments/docker-compose.yml up -d

Certs for your stack

Note - the certification process assumes you have two domains, one of them having a staging. subdomain prefix. If you don't want this pattern, you'll need to modify nginx/conf.d/api.conf and nginx/conf.d/staging.api.conf to match your domain names.

Move into deployment directory:

cd deployments

Download the helper script:

curl -L https://raw.githubusercontent.com/wmnnd/nginx-certbot/master/init-letsencrypt.sh > init-letsencrypt.sh

Open the script and edit the following lines:

domains=(YOUR_DOMAIN_ONE YOUR_DOMAIN_TWO)
data_path="./volumes/certbot"
email="<<YOUR_EMAIL>>" # Adding a valid address is strongly recommended
staging=0 # Set to 1 if you're testing your setup to avoid hitting request limits

Allow execution of the script:

chmod +x init-letsencrypt.sh

Run the script:

./init-letsencrypt.sh