More smooting, or how do we make a title long enough for Discourse to not care?

I just now made a commit where I ran go fmt on the entire project. That’s particularly interesting because it means I’ve gotten enough “sorted” that I want to take this time now.

I also just ran golangci-lint and will be working on issues it identified.

I added code to handle uploads better so that it doesn’t create duplicate revisions for a given snap upload if it already exists. This was needed because it created a bug that you could deal with but it required manual intervention. This is now fixed.

A previous issue where a snap would not download it’s base (say: core18) and would require that it be manually installed first is also fixed.

There is nothing to stop you from going from a system with no snapd to snapd fully installed and installing snaps in a matter of minutes.

This is basically the map. That’s increasingly how I am looking at it. I hope it evolves into the “real thing” but maybe, slowly we start to pull at pieces and create “all new things” to serve in the place of the old stuff but at least we have the map now.

It feels pretty good.

This is also (not exclusively) in the context of a single user. I’ve done some multi-user work and testing and have certainly laid the ground work with all the previous work on OIDC, OAuth2 and Macaroons but it’s one area that will need even more mapping out.

Well I got to the beginning of one of the things on my bulleted list from yesterday for today but actually made quite a few significant improvements today so I’ll take that.

Onward, and upward? :slight_smile: