refactor deployment scripts and updated README.md

scaffolds/app_engine/Makefile

@@ -23,6 +23,7 @@ tests:
 
 .PHONY: deploy
 deploy: clean
+	./scripts/create_app.sh
 	./scripts/deploy.sh
 
 .PHONY: run

scaffolds/app_engine/README.md

@@ -1,41 +1,51 @@
-# Developpment Workflow 
+# Development Workflow 
 
+## Local development
 
-## Local developpment
+The workflow described here works from CloudShell or any node with the [gcloud CLI](https://cloud.google.com/sdk/docs/install) has been properly installed and authenticated. 
 
-The workflow described here works from CloudShell or any node with the gcloud SDK authenticated
-,that is, you can do the developpement fully locally on your laptop, which is the preferred
-development environment.
+This means that you can develop your application fully locally on your laptop for example, as long as you have run `make auth` after installing the [gcloud CLI](https://cloud.google.com/sdk/docs/install) on it. 
+
+The first step is to add your `PROJECT` and `BUCKET` names in the following files:
+* `./scripts/config.sh`
+* `app.yaml`
+
+For local development, install then the gcloud CLI following [these instructions](https://cloud.google.com/sdk/docs/install).
+
+Make sure to accept upgrading Python to 3.10 if prompted, then authenticate for local development by running:
 
-For local development, first install the Google SDK following these instructions(https://cloud.google.com/sdk/docs/install#mac).
-Make sure to accept upgrading Python to 3.10 when prompted, then authenticate for local development by running:
 ```bash
 make auth
 ```
 
-Then create and populate the virtual environment:
+The second step is to create and populate the virtual environment with
+
 ```bash
 make venv
 ```
+After this step you should find a new folder called `venv` containing the virtual environment. 
 
-To run the tests:
+At this point you should already be able to run the tests by running
 ```bash
 make tests
 ```
 
-Run the answernaut locally with
+To run the app locally, simply run 
 ```bash
 make run
 ```
 
-Deploy on AppEngine
+At last to deploy the application on AppEngine run
 ```bash
 make deploy
 ```
 
-## Developpement workflow
+**Note:** `make clean` will remove all the built artifacts as long as the virtual environment created by `make venv`. This target is invoked by `make deploy` so that the built artifacts are not uploaded to AppEngine. The down-side is that the virtual environment will need to be recreated after each deployment.
+
+## Development workflow
 
 1. Edit the code
 1. Run the tests with `make tests`
 1. Test the app local with `make run`
 1. Deploy the app on AppEngine with `make deploy`
+

scaffolds/app_engine/app.yaml

@@ -1,10 +1,10 @@
 runtime: python310
-entrypoint: gunicorn -b :$PORT asl.server:app
+entrypoint: gunicorn -b :$PORT app.server:app
 
 runtime_config:
   operating_system: "ubuntu22"
 
 env_variables:
-  BUCKET: "dherin-dev"
+  BUCKET: "<YOUR_BUCKET>"
   LOCATION: "us-central1"
-  PROJECT: "dherin-dev"
+  PROJECT: "<YOUR_PROJECT>"

scaffolds/app_engine/scripts/config.sh

@@ -1,6 +1,6 @@
-PROJECT="dherin-dev"
+PROJECT="<YOUR_PROJECT>"
 LOCATION="us-central1"
-BUCKET="dherin-dev"
+BUCKET="<YOUR_BUCKET>"
 
 # Compute various paths from the ROOT_DIR
 ROOT_DIR="$(dirname $(cd $(dirname $BASH_SOURCE) && pwd))"

scaffolds/app_engine/scripts/create_app.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+
+. $(cd $(dirname $BASH_SOURCE) && pwd)/config.sh
+
+gcloud app describe --project=$PROJECT 2> /dev/null || {
+  echo "No App on AppEngine existing. Creating one."
+  gcloud app create --region=$LOCATION --project=$PROJECT
+}