pushup VS bob

I think this is adding unnecessary complexity. One of the reasons developers gravitate towards a framework like Alpine or HTMX is to write less JS and go back to enjoying HTML. Of course there are a lot of use cases that require custom JS scripting. But bootstrapping a project with another Go web framework and adding Alpine is also trivial. But keep going and follow your vision. I love these types of projects. Check this one out:

https://pushup.adhoc.dev

It's got some unique ideas.

This strikes me as incredibly clever.

The compiler and code-generation implementation seems equally straightforward and complex, my mind is racing trying to build a mental model of the whole thing.

The code generation comes after a feature-packed parser and "compilation" step. Emitting go source code is nice because subsequent compilation tells you if its valid or not. I'm wondering if there is a specific reason the "framework" source code is emitted via a series of printf calls[0]. A library of go template "fragments" might be easier to manage (and debug).

[0] https://github.com/adhocteam/pushup/blob/0519a782c1c9fc79877...

Use something that does codegen, like sqlc or Bob.

Also hout out to https://github.com/stephenafamo/bob

If you want to read the schema from the DB, use Bob. https://github.com/stephenafamo/bob

Generation for enums is done if you use Bob to generate your ORM.

Bob (https://github.com/stephenafamo/bob) generates models for a database schema, and for ease of use, I think it makes sense to be able to work with whatever format the schema is defined with.

Bob ORM for DB access

For an ORM recommendation, I'm heavily biased towards Bob (https://github.com/stephenafamo/bob) since I created it.

Website

Bob's GitHub Repository

I would definitely appreciate contributions to Bob: https://github.com/stephenafamo/bob