bb010g
bb010g
Short turnaround, but I changed the `nixpkgs.settings` submodule to use `shorthandOnlyDefinesConfig`. `nixpkgs.settings.config.allowUnfree = true;` really should just work, and shouldn't require `nixpkgs.settings.config.config.allowUnfree = true;`. Also, `settings.config.config` would probably be a...
> This is a big module. That's not necessarily bad, but not good for everyone. Flake-parts is meant to be an unopinionated core, so this module should, as suggested in...
@roberth Reading , it looks like both `outPath` and `sourceInfo` would have to be inputs to `outputs` that don't force `self` in order to consistently provide proper error messages when...
See also my investigation of this in #137. I should probably extract that commit set out into its own PR.
Would [OAuth 2.0 Rich Authorization Requests](https://datatracker.ietf.org/doc/draft-ietf-oauth-rar/) address the non-ticket use cases brought up here? For example, figure 1 from [draft 11](https://datatracker.ietf.org/doc/html/draft-ietf-oauth-rar-11) shows the common data field `locations` in use: ```json...
:+1: for getting this in soon.
As a side note, as long as we have `topgrade.d`, its semantics should be better defined. Configuration files in `topgrade.d` are read using [`std::fs::read_dir`]: The resulting parsed configurations are then...
Related to topgrade-rs/topgrade#345, an issue about UsoClient specifically failing on Windows 11. This issue is about topgrade more generally failing to invoke Windows Update on Windows 11.
Has there been any progress on this recently? I'm all for composibility and simplification, and the new code is a lot nicer to read.