About This Website

This site uses aws code pipeline to continuously deploy changes commited to the github repo.
Check out my buildspec.yml which handles deploying my static front end and microservice.
version: 0.2
phases:
install:
commands:
# Upgrade AWS CLI to the latest version
- pip install --upgrade awscli
# Download and install HUGO
- curl -Ls https://github.com/gohugoio/hugo/releases/download/v0.38/hugo_0.38_Linux-64bit.tar.gz -o /tmp/hugo.tar.gz
- tar xf /tmp/hugo.tar.gz -C /tmp
- ls /tmp/
- mv /tmp/hugo /usr/bin/hugo
- rm -rf /tmp/hugo*
pre_build:
commands:
# Discover and run unit tests in the 'tests' directory. For more information, see <https://docs.python.org/3/library/unittest.html#test-discovery>
- python -m unittest discover tests
build:
commands:
# Build Hugo Frontend
- cd frontend
- hugo
- cd ../
# Use AWS SAM to package the application by using AWS CloudFormation
- aws cloudformation package --template template.yml --s3-bucket $S3_BUCKET --output-template template-export.yml
post_build:
commands:
# Make sure your codestarworker for dddiaz.com (codebuild) has read,write,list permissions for the website s3 bucket
- aws s3 sync frontend/public s3://dddiaz.com
artifacts:
type: zip
files:
- template-export.yml