This was my attempt at a Vite plugin that would be able to load Markdown files individually, but also in bulk. The bulk loading offered to reduce/summarize all notes into single data structure.
My motivation behind this plugin was to support visualizations of notes, but I decided for a simple 2-step approach: pre-processing the markdown files first to jsons and then simply importing them.
- unopinionated approach -- any framework, just vite
- conceptually simple -- just say what you want to import, we'll process it for you
- no added build step
- a lot of complexity in one place -- there is a lot to manage, all while being restricted by Vite
- some processing cannot be done separately for each file -- difficult to ensure consistency between the imported files
- ultimatelly was only a step toward my ultimate goal
- there are some problems still to be solved
- watching for changes
- ensuring consistency (a note A references B, but B gets deleted)
Overall I don't think this is necessarily a bad idea. There is just few reasons to overcome the complexity of this approach and prefer it to the simpler 2-step approach.