Setup

1. Setup your database locally with two options:
   • PostGRES
     1. Install postgres locally and use the username/password in DATABASE_URL provider in .env, these should be used for connection:

     host: localhost
     port: 5432
     username: local
     password: password
     

     2. Launch the local PostGRES and use the provided .env
   • SQLite
     1. Swap out your datasource in schema.prisma:

       datasource db {
         provider = "sqlite"
         url      = env("DATABASE_URL")
       }

     2. Update your .env to instead point to:

     DATABASE_URL="file:./dev.db"

     3. Delete ./prisma/migrations
     4. Run prisma migrate dev --name init to initialize the local DB - this should create a ./prisma/dev.db if one doesn't exist
     5. You can see the database by running yarn prisma seed
2. Run locally: yarn dev

Retrospective

What could be improved?

• ran into issues with types and generation artifacts from schema to zod validation resolvers
• the graph has a simple interface that could be improved with a more readable x-axis labeling
• the notes page could include more styling clean up to improve the negative space
• the Layout component added more white space than originally intended. Didn't bother to dig into this for now.
• configuring the Heroku Postgres instance was an extra step that required additional testing. Next time, I'd start with setting that up.
• improve the loading states - they're pretty basic
• ended up skipping tests - could configure them to run when deployment
• add husky git hooks to prevent deployment of TypeErrors
• add import aliases to clean up the import paths in files
• graph could look better on mobile (currently flattens)
• improve the UX of the timestamps (date/time formatting) - originally had issues with the chart library where it shifted the values if I modified the labels

What went well?

• Enjoyed trying out some new technologies and organizing the code
• Made use of generics
• Converted the test data for graph into an API as well
• Setup a local Postgres instance for deployment