I was thinking of adding some KV usage to this, however I realised I would run into a problem for anyone wanting to have a successful one click deployment because KV binding is done using a namespace name + id right? e.g
kv_namespaces = [{ binding = "TODO", id = "06779da6940b431db6e566b4846d64db" }]
kv_namespaces = [{ binding = "TODO", id = "06779da6940b431db6e566b4846d64db" }]
.
So, if I had KV in wrangler.toml, then the one click deployment done by the button wouldn't work since that KV id doesn't exist on their CF account. I was thinking about this, and I was wondering if it was a good idea to allow KV binding using just the namespace name instead of the id. That would allow me to have the github action create the namespace if it didn't exist. A great example of this is R2, which the project currently uses. An R2 binding is just using the name of a bucket, so it can be transparently created in CI without having to modify wrangler.toml.