TODO:
- Use a frontend framework like Vue or React to component-ize things
- Figure out which fields are actually needed in the
features.csv
andtemplates.csv
indexes - Add vendor-specific code to fetch stars/downloads of OCI Repository sources
npx serve
Or any static HTTP server
Users can create issues from the add-thing.yml
Issue Form which will then
trigger the add-thing.yml
GitHub Actions workflow. This will then add the
thing (feature, feature collection, template, or template collection) to the
main _data/things.csv
file. This is the index of all the "entry points" (for
lack of a better term) that we then crawl to maintain a concrete list of
features and templates. Why can't we do this at runtime? Because to make the
search responsive we need to precompute all the features and templates otherwise
we'd be making 700+ API requests to fetch all the metadata for all the leaves of
the _data/things.csv
list!
Why a DB-like CSV file?
Because this isn't near the scale of update rate or quantity where a full-on DB/API
makes sense. For the volume of updates (a few per week) and the quantity of items
(<5000) it makes sense to use a simpler solution.