V1.0 development (just a draft PR to view the diff)

[PThorpe92]

this is so everyone can fork this branch, and make their own edits, add whatever features, etc. so instead of a massive PR that is impossible to review, we can each add things incrementally

[PThorpe92]

I see the git history is diverged enough where it might have to be rebased off main to one commit :/

[MartinFillon]

Gonna merge and update it

[MartinFillon]

Okay so how are we doing on that v1.0 release ?

[daviessm]

I really think that clap should be merged prior to v1, there are quite a few things queued up waiting for it and we should aim to get all those merged and tested before declaring we're stable.

[PThorpe92]

I'm all for it. Yeah this is really more about clap and anything that would come along with it that would have breaking changes than it would be actually 1.0

more specifically I just wanted a branch people can fork and push their changes to, instead of having 1 PR and requesting a bunch of changes for one person. I think the sooner we get it merged into main the better. you're absolutely right, from a stability perspective this needs to be in well before 1.0 I was caught up on the breaking change perspective

[MartinFillon]

Well guess this can be made ready for review ?

[MartinFillon]

Hey @tertsdiepraam you created a breaking change on time-style argument, could you have a look at it
For an exemple please see the test file named long_time_style_custom_non_recent_and_recent_nix.toml and tell me if its intended.

[tertsdiepraam]

@MartinFillon I don't think it's due to my changes. With git bisect I've narrowed it down to this commit where the time-style does not accept "+NON_RECENT\n RECENT" anymore: 47b9e45

[MartinFillon]

Well my bad sorry 😞 I should have bisect more in depth

Gonna fix it

[MartinFillon]

This is ready @cafkafk

[cafkafk]

This is ready @cafkafk

Okay let's get going on review then!

[cafkafk]

First order of business is to resolve clippy lints so CI pass it seems

[MartinFillon]

I can't understand nix flake check outputs. It seems to always fail

[tertsdiepraam]

Seems to be a formatting issue. I think nix fmt can fix it, but cargo fmt doesn't seem to fix it, which seems strange, because nix fmt should essentially be running cargo fmt under the hood. Maybe the rust-toolchain.toml file might be ignored by nix and it will therefore run a newer version of rustfmt?

Edit: yeah that seems to be it. I temporarily deleted rust-toolchain.toml and then cargo fmt gives the correct diff.

[MartinFillon]

well thats strange Should I fix it ?

[MartinFillon]

Up @cafkafk