stevedylan.dev
Since .sequoia.state.json helps perf when run across sessions, should that be committed? It's in .gitignore, though I don't recall if I did that manually or if sequoia init did. This could probably also help with CI-driven publication.
A CLI for publishing standard.site documents to ATProto
sequoia.pub
standard
site
lexicon
cli
publishing
Should .sequoia-state.json be committed? #14
open
opened by
heaths.dev
heaths.dev
heaths.dev
(author)
...though, thinking more about it, I'm not sure this solves the problem either since - typically -you wouldn't want your CI to push files back to the repo, which it would need to. So build artifacts might be the only reasonable way, apart from just regenerating the file...which works until you have a lot of posts.
heaths.dev
(author)
Though, maybe I’m conflating two different workflows here. It wouldn’t practically work for a CI workflow, but it would help for a manual workflow, at least for single- or few-publisher sites.
I thought about this the other day and came to the same conclusion; will need to update the init script to not alter the gitignore