rust-lang/WG-CLI

https://rust-lang-nursery.github.io/cli-wg/ – tracking repo: https://github.com/rust-lang-nursery/cli-wg/issues

Russell Cohen

@rcoh

@epage I'm seeing a weird behavior with cross on travis when combined with assert-cli. All my regular tests run, but then the first two of my assert-cli tests take ~5-10 minutes (as if maybe it's fully compiling from scratch again?)

Pascal Hertleif

@killercup

@rcoh yeah, assert-cli calls cargo internally in a specific way that sometimes makes it recompile everything :( this especially happens when you specify --target… so using cross will trigger this

see assert-rs/assert_cmd#57

Daniel Sockwell

@codesections

Hi all! I'm working on a simple CLI app and have gotten to writing the man(1) page for the app, so I'm looking for the current best way to do so. I found https://github.com/rust-clique/man ; is that the best option? I'm asking because it looks like that repo isn't that active—no commits in a couple of months, and the README example doesn't work (e.g., it uses the .description method instead of the .help method in the code.

I also found https://rust-lang-nursery.github.io/cli-wg/in-depth/docs.html, which discusses rendering man pages from Clap, but that doesn't seem to be implemented (yet?)

I'd be happy to help out with the man crate—it seems really cool and helpful (I'm a big fan of man pages for CLI apps!). But I don't want to spend time opening issues/PRs if the community or WG has moved on to a different solution

Pascal Hertleif

@killercup

@codesections hi! the current support for man pages in clap is quite experimental, and requires the alpha version of clap 3. the man crate(s) are also quite simple right now --if you want to help out there it would be greatly appreciated!

i'm at work right now, but we can talk later if you want, just ping me!

Daniel Sockwell

@codesections

@killercup thanks for the reply. I'd be happy to help with the man crate! Just to be clear, though, it's not going to be depreciated when clap 3 has support for generating man pages, right? No sense on working on something that is about to be replaced/eliminated, after all

Pascal Hertleif

@killercup

@codesections AFAIK clap will use https://github.com/rust-clique/man

the internals of that might change drastically depending on what we want to do

(the roff crate for example is super simplistic right now)

Daniel Sockwell

@codesections

@killercup Ok, cool. I've opened two issues and submitted one (minor) PR for man (which, if merged, would close one of the issues). I'll probably wait a bit before getting started on the PR for the second issue to see if you/anyone else have other ideas for a good way to proceed.

Pascal Hertleif

@killercup

@codesections hey, thanks for all the contributions to the man crate! the roff crate is pretty much just a POC so far, but @aoikonomopoulos_gitlab made an alternative impl some time ago: https://gitlab.com/aoikonomopoulos/rroff

if you're interested in working on a good roff crate and bring the design together, i'd love to add you as a contributor (and maybe move it to the clique github orga) :)

Daniel Sockwell

@codesections

Yeah, I'd definitely be interested in that/helping out with more CLIque stuff in general.

What do you think the next steps are for creating a good roff crate?

Dylan DPC

@Dylan-DPC

can help with that s well :)

Pascal Hertleif

@killercup

@codesections moved to https://github.com/rust-clique/roff-rs and added you to maintainers group

Dylan DPC

@Dylan-DPC

:+1:

Pascal Hertleif

@killercup

good question about next steps. i don't know. the "grand vision" was to have clap generate man pages, and make prototypes that allowed that but i haven't planned beyond that

Daniel Sockwell

@codesections

Thanks! What does it mean to be part of the maintainers group? I thought that would be the group that has write access to the repo, but it doesn't look like I do (unless I'm missing something)

Pascal Hertleif

@killercup

@codesections you should have write access? let me double check

@codesections ah you need to explicitly accept the invitation it seems

Daniel Sockwell

@codesections

Aha, that resolved it, thanks. (Odd UX for GitHub to send an email about that but not have a notification in the web UI)

What's the etiquette within CLIque for merging ones own PRs? I know some groups dislike that/want more code review but others like it to keep the pace of development up/avoid blockers

Pascal Hertleif

@killercup

@codesections don't think we have a specific etiquette established here, so let me come up with something: it's fine if the PRs implement what the core maintainers decided to do

Daniel Sockwell

@codesections

Makes sense :+1:

Pascal Hertleif

@killercup

which leads us to: who are the core maintainer of that repo? i'd propose you, @yoshuawuyts, and @Dylan-DPC who volunteered earlier :)

yes this is super informal and boils down to: best just have a quick chat about it and then get coding :)

Daniel Sockwell

@codesections

:) Sounds about right to me! I think though, that at least in the short term, there's no a whole lot that roff-rs needs to support the core work in man. So, unless anyone has particular features they'd like to see in roff-rs, I expect the majority of my time will be spent on man

It feels like man is very close to being usable for most CLI apps (even if better integration with clap would be great in a post-clap-3.0.0 world)

Pascal Hertleif

@killercup

sounds great!

(that reminds me to add the clique maintainers as owners of the crate on crates.io too. done)

Yoshua Wuyts

@yoshuawuyts

killercup: hi, yes — totally on board with having codesections on the project!

Input has been super helpful; I love that it's active again; never saw it through to the end, so happy it's getting there ^_^

Pascal Hertleif

@killercup

@yoshuawuyts great!

killercup @killercup is revising the CLIque 2019 roadmap to include "stick it to the man"

Ricky

@deg4uss3r

Any other rust cli nerds at LCA?

Pascal Hertleif

@killercup

i don't even know what that is :sweat_smile:

Daniel Sockwell

@codesections

^^^

Ricky

@deg4uss3r

Linux Conference Australia

https://linux.conf.au/

Pascal Hertleif

@killercup

ah, okay, not the Life Cycle Initiative :D

Daniel Sockwell

@codesections

That sounds more fun than the Louisiana Correctional Association

Ricky

@deg4uss3r

lol probably a bit more @codesections

Pascal Hertleif

@killercup

maybe it's really just a bunch of people in front of a whiteboard finding the Lowest Common Ancestor in a graph 🤷‍♂️

Ricky

@deg4uss3r

plz2no

Daniel Sockwell

@codesections

Could we do a patch version bump on man sometime? I've been working on a CLI app that uses the example method we added to man, and I can't publish it on crates.io with man as a dependency from github

Pascal Hertleif

@killercup

@codesections I've only seen the amount of PRs you made, not read their contents but I'd be amazed if there was no breaking change in there :) also: you should have the rights to publish the crate

Daniel Sockwell

@codesections

Haha, fair enough. I don't have write access to the man repo, though, just to roff-rs

Dylan DPC

@Dylan-DPC

will publish a new verson in a few days

Where communities thrive

People

Repo info

Activity