HN Explorer

Ask HN: What do you think a good GraphQL boilerplate needs?

2 points by sbacic ↗ HN
On my last project that involved GraphQL, I spent an inordinate amount of time fixing version/library conflicts, reinventing the wheel for things that should honestly be available out of the box (user groups, access control, auth, etc) and generally wrestling with the whole ecosystem in order to get it working the way I want to.

That's why I've decided to go and write an opinionated full-stack boilerplate generator for GraphQL that comes with batteries included and tries to automate as much of the repetitive work as possible. I'm pretty set on JavaScript for the whole stack, with Postgres as the database and React for the frontend. I can't decide between Prisma and Sequelize for the ORM.

Which do you think is more appropriate?

My chief goal is to have the user only define the most elementary elements of their API (such as the data model) and have the boilerplate generate everything else - all the way from the database to the frontend CRUD UI, with sane defaults that could be modified by the user to suit the project's particular needs.

What do you think a good GraphQL boilerplate generator would need in order to satisfy your needs?

2 comments

[ 3.1 ms ] story [ 3.0 ms ] thread
verdverm
Check out https://github.com/sysgears

They have a boilerplate A.U.S.K, CLI tool for adding new models, spinjs for managing the build system, nd uapm is in the works for larger modules with diff3 capability

sbacic
It looks pretty interesting, the only thing that's turning me off is that it's the first time hearing about some of those technologies, so I'm not sure how safe of a choice they are compared to more established solutions like Sequelize.

Also, knex.js is not really an ORM layer, just a query builder?