Ginkgo Classroom Emailer

The Ginkgo classroom at Lee Montessori has three weekly services provided by student families. This project parses the sign-up sheet in tsv form and sends reminder emails to the families who have signed up a few days beforehand.

Configuration

To run the function, you need a .env file with

Your gmail username and an app-specific password
The unique key for your Google spreadsheet (check your spreadsheet's URL: it's the hash after /d/)
A comma-separated list of the room parents' email addresses. Not the family email addresses, but the parents who are managing classroom volunteer stuff for the year.

You can use the sample.env file, included in the repo, as reference.

cp sample.env .env

Sending these emails to actual people requires a src/familyEmails.js file, which is not included in this repo for privacy reasons. Consult src/sample.familyEmails.js for reference on how to format it.

Scheduling the script to actually run is done through the AWS console using CloudWatch, not through any configuration present in this project.

Usage

The function is an AWS lambda, which gets uploaded and independently scheduled via CloudWatch to run daily at a certain time (8 AM, currently). It is controlled via a few different flags in the event object it receives:

key	type	Description	Default
debug	boolean	Displays more information via console.log when run, and doesn't send any actual emails.	true
date	string	Override the automatic date with a different date of your choosing.	undefined
sendToRealPeople	boolean	Setting this to true will send notifications to actual people: if false, sends all email to me.	false

To start, you'll need node. Then run

npm install

You can invoke the function locally using the event object in debug-send-event.js:

# runs the current src/index.js
npm run invoke

# runs the current built file in build/index.js
npm run invoke-built

The project uses webpack to bundle and minify the JavaScript into a single file, then zips it and the other files it needs into a single file, ready to upload.

To actually upload these changes to AWS, make sure you have the aws-cli set up and properly credentialed, then run