Fish rewrite-it-in Rust progress: 100%

Here's one issue they hope to solve with this rewrite: https://github.com/fish-shell/fish-shell/issues/238

One big, long-standing issue is that fish can't run builtins, blocks or functions in the background or at the same time.

That means a pipeline like

 fish

    
seq 1 5 | while read -l line
    echo line; sleep 0.1; 
end | while read -l line
    echo line; sleep 0.1
end

  

will have to wait for the first while loop to complete, which takes 0.5s, and then run the second.

So it takes 0.5s until you get the first output and a full second until you get all of it.

Making this concurrent means you get the first line immediately and all of it in 0.5s.

While this is an egregious example, it makes all builtin | builtin pipelines slower.

Other shells solve this via subshells - they fork off a process for the middle part of the pipeline at least. That has some downsides in that it's annoyingly leaky - you can't set variables or create a background job in those sections and then wait for them outside, because it's a new process and so the outer shell never sees them.

Interesting that you mention it. I mainly use fish but always do some stuff in nu to check out its progress. They are in my opinion the two most interesting interactive shells at the moment that I know of, the third shell I keep an eye on is oil but rather as a replacement for bash when used in scripts rather than interactive. The project also has ysh which also doesn't look too bad and seems to go in a similar direction as fish.

Large parts of the rewrite came from contributors who had never worked on fish before.

That's pretty useful alone.

And there's this:

Thread Safety

Allowing background functions and concurrent functions has been a goal for many years. I have been nursing a long-lived branch which allows full threaded execution. But though the changes are small, I have been reluctant to propose them, because they will make reasoning about the shell internals too complex: it is difficult in C++ to check and enforce what crosses thread boundaries.

This is Rust's bread and butter: we will encode thread requirements into our types, making it explicit and compiler-checked, via Send and Sync. Rust will allow turning on concurrent mode in a safe way, with a manageable increase in complexity, finally enabling this feature.

They did it "for the vibes"

Vibes are just as important to free/open source software as proprietary software and although there were solid technical reasons for the port, the PR outcomes are added benefits.

It was needed to safely further support for concurrent features? If they follow through on adding that support, there will likely be adoption.

The problem is in most cases the implementers stop at "same thing but in rust" without taking advantage of that.

I can't fully blame them since just duplicating an existing thing is a huge undertaking.