TODO

• Consider setting up a google Cloud Run pipeline to automatically downsize large images.
  • https://cloud.google.com/run/docs/tutorials/image-processing

Structure

• Images are stored in GCP Storage bucket
• App is hosted on GCP App Engine
• Uses an Express server to create routes for getting image metadata and serve the React client app
• React client app presents the images in a gallery.

Storage

Set the default sharing policy to public: gsutil defacl set public-read gs://[YOUR_BUCKET_NAME]

Local Dev

When running locally, you can use the Google Cloud SDK to provide authentication to use Google Cloud APIs:

`gcloud init`

`gcloud projects list`
`gcloud config set project craigschoppe`

Start the server using npm run-script start:local. This sets the appropriate gcloud environment variables.

And start the react app in dev mode using npm run-script start:react. This includes a proxy to call the correct local server, and still has hot reloading.

Deploy

Then build locally before deploying:

`npm run-script build`
`gcloud app deploy`

Images

Images are stored on Google Cloud storage in a bucket. Multiple buckets are used for performant images.