Help On Spot Lambdas

This repo contains all aws lambda functions, plus a commons library.

Structure

• /common includes common libraries that include code that is shared between lambdas

• /common/help-on-spot-models - REST models + DB access

• /lambdas includes the lambda function code. General idea: one lambda function for every API endpoint

Start locally

• npm run test - in any lambda folder
• If you want to lint/format your code, run the root-level lint/format npm commands (see /package.json)

Local Postgis DB

• If you need postgis support for your local database use:

docker run --name "postgis" -p 5432:5432 -d -t kartoza/postgis:10.0-2.4

• You can then login with user "docker", password "docker" and databasename "gis"

Local environment variables

• Create a local .env file in the root directory of each lambda

Available .env variables:

DATABASE_HOST=
DATABASE_USERNAME=
DATABASE_PASSWORD=
DATABASE_NAME=
AWS_ACCESS_KEY=
AWS_SECRET_KEY=
AWS_REGION=
GEOSERVICE_LAMBDA_NAME=

Deploy lambdas to AWS

• install aws cli
• setup your AWS credentials wth aws configure
• create a files named .lambda-envs-dev.txt and .lambda-envs-prod.txt in project root. Places envs needed for lambdas there (comma separated i.e VAR1=value,VAR2=value2,...)
• run ./deploy-lambda.sh in project root
• select the stage you want to deploy to (dev,prod, etc) If you want to add a new stage you also need to create it in API gateway + the needed stage variables
• select the lambda you want to deploy or all to deploy every lambda

Note that the deployment of common layer is included inside deploy-lambda.sh, so no need to execute deploy-layers.sh seperately

Steps to create a new lambda function

• create a new folder in /lambdas, name it according to the served endpoint

• setup a typescript project, make sure the tsconfig and jest-config have a reference to the commons project you want to include

• add folder name to deploy-lambda.sh script's lambda array (Line 5)

• point API gateway's endpoint 'Integration Request' to new lambda (make sure to use stage variable in lambda name so that different stages can point to different lambdas)

• assign correct input/output models or create them if they don't exist yet

• if changes have been made to the API, export OpenAPI docs and commit them to Helfer-Portal/helponspot-api-gateway repo

RDS Postgres DB

• The application is using an AWS RDS instance with postgres version 10.11
• If you need to enable the PostGIS extension on the db you need to execute create extension postgis; for enabling the extension
  • You need the PostGIS extension for saving spatial data like "point" of address