setup VS oil

https://github.com/kbd/binrun

I just wrote it today and still need to package it. I was tired of alt+tab + up arrow + enter constantly to execute builds etc. in my terminal. It calls out to a wrapper script I wrote[1] that does things like queries kitty for its running windows so that when I launch from vscode it can find the right kitty window for the vscode workspace and execute there...

Point is, Kovid Goyal is awesome and the extensibility he wrote into kitty makes all that possible. I had no idea he was also the author of Calibre until I'd been using kitty for a while.

[1] https://github.com/kbd/setup/blob/master/HOME/bin/kw

If anyone cares, here's my config: https://github.com/kbd/setup/blob/master/HOME/.hammerspoon/i...

It shows off a tiny bit of what you can do with Hammerspoon:

- window and app management

I'll put it this way: Nu shell seems perfectly supportive of my philosophy that a shell is basically a REPL for a computer, and they're taking the ergonomics of an interactive REPL along with the programming language that powers that REPL seriously.

The thing is, there's currently NOTHING GOOD for "shell scripting". Shell sucks (yes it does), so for anything more than very short things I'd rather write Python. But Python sucks for shell-like things, parallelization, it has slow startup, and you also can't do things like put environment variables into your session or change the working directory, so you often wind up writing shims (eg. Broot's br alias - https://dystroy.org/broot/install-br/).

Yes I've looked at Xonsh but maybe the additional syntax is offputting to me. Like, I wouldn't use it as a shell over Zsh (how's Xonsh's fzf support? I don't know, but I know everything's going to support Zsh), and I dunno if I want to use its syntax extensions over just Python. Though It's always on my list of things to re-explore, and maybe it'll click one day. But it being based in Python makes it feel slow (I wrote my prompt in Zig to get it to be fast...)

This is relevant to mention: I wrote a small Python library (https://github.com/kbd/aush) that's basically a DSL for subprocesses, so it tries to make it more convenient to do shell-like things. I find it preferable to shell or Python alone most of the time. Here's an example of its use in my script that creates a new Python project: https://github.com/kbd/setup/blob/master/HOME/bin/create-pyt...

I haven't figured out a convenient way to implement shell piping well with Python's pipe operator, or pass through interactive output directly (so things that "update" the display, like poetry and npm don't behave the same as they do interactively) so it's still .9 status, but it works really well for what it is, and you can always write "regular Python" along with it.

Anyway, Nu seems to be an attempt to put a "real" programming language REPL in my shell, from people who have serious language experience, so I'm hopeful it'll be great.

Just sharing my take on that interactive jq (or anything else) repl:

https://github.com/kbd/setup/blob/master/HOME/bin/fzr

It's just an fzf wrapper that sets up temporary files and so on. It works really well; it's amazing all the things one can use fzf for.

Since we're sharing, my dotfiles setup has pretty much reached its final form. I use my symgr[1] to symlink my dotfiles repo into my home dir. Pretty much everything I think about this topic is in its readme, as well as a link to my setup[2] repo with my dotfiles showing how I use symgr.

[1] https://github.com/kbd/symgr

[2] https://github.com/kbd/setup

It's not exactly a tiling window manager, but if you can program some simple Lua then Hammerspoon is a godsend. You can program anything any of the other window managers for Mac (like Rectangle, Spectacle, etc.) can do and have complete freedom to set up your own keyboard shortcuts for anything.

I have some predefined layouts[1] for my most common usage. So, one keyboard shortcut arranges the screen how I want, and I have other keyboard shortcuts[2] (along with using Karabiner Elements for a 'hyper' key) to open or switch to common apps.

[1] https://github.com/kbd/setup/blob/1a05e5df545db0133cf7b6f1bc...

[2] https://github.com/kbd/setup/blob/1a05e5df545db0133cf7b6f1bc...

Figured I'd link my git aliases here, that make heavy use of fzf. The goal is generally to never have to type a filename (eg. for git add) or a commit hash (eg. for cherry-pick).

Here's a link to my 'cp' alias that lets me choose a branch, then a commit to cherry pick into my current branch:

https://github.com/kbd/setup/blob/e23b3e8e2363284c3c766c0be2...

Some related references (on a somewhat messy wiki page) - https://github.com/oilshell/oil/wiki/Compact-AST-Representat...

Feel free to add others

will prevent almost all of the "silent footguns".

YSH has strict:all and then a bunch of NEW features.

There's been good feedback recently, which has led to many concrete changes. So your experience can definitely influence the language! https://github.com/oilshell/oil/wiki/Where-To-Send-Feedback

Regarding writing tools/tests/benchmarks in bash+Python, vs. writing tools in your main language:

I think we might eventually concede that something Debian-like is the “standard development environment” (at least for server side stuff, i.e. not iOS apps)

In this case, bash+Python is a non-issue. It works extremely reliably. That’s actually why I use it! Everything else seems to break, or it’s really slow (node.js is a very common alternative).

- Microsoft conceded this back in ~2017, by building Linux into their kernel with WSL, and providing Ubuntu on top

Yes bash + Python is a disaster on Windows (I have scars from it), but Microsoft agrees that the right place to solve that is in Windows :-)

- Every CI system runs Debian/Ubuntu

- Every hosting provider runs Debian/Ubuntu

- Every online dev env like gitpod.io provides Debian/Ubuntu

This is somewhat related to remote dev envs: https://lobste.rs/s/ucirlx/lapdev_self_hosted_remote_dev

One vision for https://www.oilshell.org/ is that the CI environment is the dev environment is the hosting environment.

Everything is just an equal node in a distributed system. BUT it’s more git like, in that you explicitly sync and work “locally”, wherever that is. You don’t have the network chatter and flakiness of “the cloud”.

Oils has a very large set of monotonically increasing properties too - https://www.oilshell.org/release/0.21.0/quality.html

All that is bash+Python that is run on every commit, and it’s extremely good at catching bugs and perf regressions.

I’m skeptical that any project has that level of quality automation written in pure Rust or Zig. More likely it’s a bunch of cloud services with YAML.

Also a bunch of “hard-coded” toolchains that you can’t script with bespoke code. Like some shell commands in your package.json, which is just a worse way of writing a shell script.

Our quality process is all self-hosted, in the repo, and runs on both Github Actions and sourcehut - https://www.oilshell.org/release/0.21.0/pub/metrics.wwz/line...

bash and Python runs perfectly on Github Actions and sourcehut, with zero change. Containers also do.

(Although we need to unify the CI and release, because the release runs on 2 different real hardware machines, while CI is cloud only.)

Also, a main point Oils is that bash now has another highly compatible, spec-driven implementation – OSH. Having 2 independent implementations is something newer languages don’t have.

(copy of lobste.rs comment)

in your bashrc to enable it. I've used it for probably ~18 years now.

It also works with https://www.oilshell.org/ since we use GNU readline. Just 'set -o vi' in ~/.config/oils/oshrc

No other shell does that.

But I didn't know it was called MULTIOS until now. (I guess that's read "mult I/O's"? I have a hard time not reading it was multi-OS :) )

It seems a bit niche to be honest, but it's possible to support in Oils.

---

Oils also uses Unix domain sockets already for the headless shell protocol

https://github.com/oilshell/oil/wiki/Headless-Mode

We could do something like dgsh, but so far I haven't seen a lot of uptake / demand. Every time it's mentioned, somebody kinda wants it, and then it kinda peters out again ... still possible though.

I think flat files work fine for a lot of use cases, and once you add streaming, you also want monitoring, more control over backpressure/queue sizes, etc.

which works well. You don't have to clean when rebuilding variants. IMO this is 100% essential for writing C++ these days. You need a bunch of test binaries, and all tests should be run with ASAN and UBSAN.

---

I wrote a mini-bazel on top of Ninja with these features:

https://www.oilshell.org/blog/2022/10/garbage-collector.html...

So it's ~1700 lines, but for that you get the build macros like asdl_library() generating C++ and Python (the same as proto_library(), a schema language that generates code)

And it also correctly finds dependencies of code generators. So if you change a .py file that is imported by another .py file that is used to generated a C++ header, everything will work. That was one of the trickier bits, with Ninja implicit dependencies.

I also use the Bazel-target syntax like //core/process

This build file example mixes low level Ninja n.rule() and n.build() with high level r.cc_library() and so forth. I find this layering really does make it scale better for bigger projects

https://github.com/oilshell/oil/blob/master/asdl/NINJA_subgr...

Some more description - https://lobste.rs/s/qnb7xt/ninja_is_enough_build_system#c_tu...

This is sort of a category error...

re2c is a lexer generator, and YAML and Python are recursive/nested formats.

You can definitely use re2c to lex them, but it's not the whole solution.

I use it for everything possible in https://www.oilshell.org, and it's amazing. It really reduces the amount of fiddly C code you need to parse languages, and it drops in anywhere.

SEEKING VOLUNTEERS - https://www.oilshell.org/ - https://github.com/oilshell/oil/

I'm looking for people to help fill out the "standard library" for Oils/YSH. We're implementing a shell for Python and JavaScript programmers who avoid shell!

On the surface, this is writing some very simple functions in typed Python. But I've realized that the hardest parts are specifying, TESTING, and documenting what the functions do.

---

The most recent release announcement also asks for help - https://www.oilshell.org/blog/2024/01/release-0.19.0.html (long)

If you find all those details interesting (if maybe overwhelming), you might have a mind for language design, and could be a good person to help.

Surveying what Python and JavaScript do is very helpful, e.g. for the recent Str.replace() function, which is nontrivial (takes a regex or string, replacement template or string)

But there are also very simple methods to get started, like Dict.values() and List.indexOf(). Other people have already contributed code. Examples:

https://github.com/oilshell/oil/commit/58d847008427dba2e60fe...

https://github.com/oilshell/oil/commit/8f38ee36d01162593e935...

This can also be useful to tell if you'll have fun working on the project - https://github.com/oilshell/oil/wiki/Where-Contributors-Have...

More on #help-wanted on Zulip (requires login) - https://oilshell.zulipchat.com/#narrow/stream/417617-help-wa...

Please send a message on Github or Zulip! Or e-mail me andy at oilshell dot org.

This is true, but then the corrolary is that new PRs need to come with this higher and rigorous level of test coverage.

And then that becomes a bit of a barrier to contribution -- that's a harness

I often write entirely new test harnesses for features, e.g. for https://www.oilshell.org, many of them linked here . All of these run in the CI - https://www.oilshell.org/release/latest/quality.html

The good thing is that it definitely helps me accept PRs faster. Current contributors are good at this kind of exhaustive testing, but many PRs aren't