search

bchao1 / bullet

🚅 Interactive prompts made simple. Build a prompt like stacking blocks.
https://pypi.org/project/bullet/
MIT License
3.55k stars 113 forks source link

Is this project maintained? #87

Open jeffwright13 opened 2 years ago

jeffwright13 commented 2 years ago

Haven't seen any activity in a while. There are several PRs out that would improve it, but I don't seem them being addressed, either.

a-luna commented 2 years ago

I tried emailing the maintainer a long time ago (July 2021) asking if I could offer my services as a maintainer to help triage issues and deal with PRs, etc. No response whatsoever unfortunately.

jeffwright13 commented 2 years ago

Might have to fork it. Or, consider another library. Lots of areas for improvement.

davedittrich commented 2 years ago

I started using it in several projects years ago. Switching to another library (assuming one can be found that is similar in function and ease of use) would be a bit of work, but using an abandoned library is also unappealing. Any ideas of maintained alternatives? (Or can @a-luna let us know if you decide to fork?)

jeffwright13 commented 2 years ago

Well there is SimpleTerm https://pypi.org/project/simple-term-menu/ - I have not used it but at least there was a release a couple of months ago.

Maybe we should collectively take over this repo, or fork it and start making it better. Anyone want in with me? Or should I try the simple-term route to see if that's a reasonable alternative?

a-luna commented 1 year ago

Sorry for the delayed response, I would definitely be willing to fork and participate as a maintainer.

I do not have the bandwidth to take over the project solely by myself, but would love to help create/maintain the roadmap, merge/reject the current PR backload, and triage issues.

I would only be willing to move forward if the two of you (@jeffwright13 and @davedittrich) can commit to a similar workload. Any lurkers out there reading this please chime in if you would also be willing to help out.

However, it might be wise to create a single new release of bullet that merges critical PRs and fixes blocking issues and cease new development.

Why? Because @willmcgugan (creator of the excellent rich library (over ⭐️39K)) is working on a terminal UI framework called textual (over ⭐️13K) which honestly makes bullet look sad and unnecessary in comparison.

I can see there being an argument for bullet as a different category than textual, with a much, much smaller feature set that therefore makes it easier to learn, thus requiring less time to apply to a project. And if you really only need the menu system that bullet provides, trying to figure out how to create the same experience in textual would require a steep learning curve for no benefit since you won't ever use the other 95% of textual's features.

Sorry for this somewhat contradictory ramble of ideas, I'd love to hear both of your thoughts (and anybody else out there reading this, please reply as well!)

jeffwright13 commented 1 year ago

Hi @a-luna,

Yes, I would be willing to commit to such a workload. TBH this is a fairly small codebase, and I don't think it would be too much effort to get it into a state where it's meeting everyone's needs (i.e. PRs merged, critical bugs addressed, etc.). I would look to you for guidance on the roadmap but I can provide help with PR reviews, fixing issues, etc. I'd also like to see some tests put in place and possibly a Github Actions build/release train.

BUT - what I like about Bullet as it stands now is (1) the code is in place and it mostly works; (2) it's really easy to get off the ground and running without a lot of setup. :-)

As far as Textual goes, I'm not sure a full blown TUI is necessary for such a simple application as this, but maybe I am missing something. I can certainly see how Rich (the 'other' Will McGugan library that Textual depends on) would put a fresh face on Bullet though. I am pretty familiar with both Rich and Textual, having used Textual to build up a TUI for the pytest-tui plugin, so it would be fun to use those to make over Bullet.

pzelnip commented 1 year ago

While I appreciate that textual is quite sophisticated & in that sense supersedes bullet, the thing I like about bullet is that it's so easy to just quickly get some simple prompts working. Textual's docs are a bit daunting as it does so much, so for small projects where you just want a Yes/No or Password input, Textual feels like overkill.

h4rldev commented 1 year ago

https://github.com/h4rldev/rebullet 👍

hlin-neo4j commented 1 year ago

https://github.com/h4rldev/rebullet 👍

Hi! Would you mind providing a brief context on the pull requests and/or fixes you've applied in your fork? Thanks!

h4rldev commented 1 year ago

https://github.com/h4rldev/rebullet 👍

Hi! Would you mind providing a brief context on the pull requests and/or fixes you've applied in your fork? Thanks!

There might be more, I haven't worked on rebullet in a while, but I can say it works on 3.10, I don't really know if it works below 3.10, if it doesn't. Make a pull request that reverses these 3.10 only features 👍

Although I might not be able to fix any new issues right now, I haven't worked in python in a while, neither do I have my environment set-up, waiting on a new m.2 ssd.