Automate all the things with Swift Subprocess

Automate all the things with Swift Subprocess(blog.jacobstechtavern.com)

40 points by jakey_bakey 217 days ago | 17 comments

w10-1 209 days ago |

OP objections to scripting in Swift are (1) requires a build step; and (2) subprocess API requires wrapping for each tool -- both are true.

The Swift team did just release Subprocess, but it doesn't break hugely new ground. Swift has had API's for running system processes, and the best wrapper has been the Shwift library[1] which supports async operations and bash-like syntax for captures.

Wrapping tools with static functions that make convenient choices is helpful, but the real benefit comes with using type-safe API's - e.g., records for git-log output, enumerations for configuration, etc.

For the update-build-and-run dance, there are tools like clutch [2]. It helps most when you have a bunch of scripts that benefit from common (wrapping) libraries - as with builds, processing pipelines, etc. - because the common code doesn't need to be re-built, only the script.

- [1] shwift: https://github.com/GeorgeLyon/Shwift

- [2] clutch: https://github.com/swift-nest/clutch

jbverschoor 209 days ago |

Related to using Swift as scripting: https://github.com/jrz/tools (transparent compilation vs interpreting)

jakey_bakey 209 days ago | |

Sick!

jbverschoor 208 days ago | | |

Thanks. Was scratching my own itch. Got frustrated with startup times for reasonably simple scripts in ruby and applescript. Rewrote some things in swift, but then got annoyed by the startup times of interpreted swift. I'm dogfooding my own tools, so I want both development and usage to be quick.

eviks 209 days ago |

> The Swift code is compiled and executed, assuming all the corresponding overhead.

Unless it's cashed, of course, then you get the extra overhead on first run, but lower overhead on at the subsequent runs? This is even mentioned in the article, so what's the issue in this specific peculiarity?

John23832 209 days ago |

I feel like Swift could be such a great language if it was only given the proper care and feeding of open source. Instead it's largely locked in the apple walled garden with tokens given to the outside.

w10-1 209 days ago | |

> it's largely locked in the apple walled garden with tokens given to the outside

So, the compiler, stdlib and runtime, core libraries, build system ... not enough? What else would you want?

I feel the problem is not what's in open source, but that the open-source community cannot really form, since no outsider can significantly change what the Apple contributors decide. Some of the peripheral projects have relatively free rein, but they can't compete e.g., with server libraries elsewhere.

Also, the Apple people have to track what Apple needs, so they'll put out stuff per schedule that works for them but falls apart on untested code paths. And they don't really deprecate stuff, so you end up with multiple ways to do the same thing. And there seems to be no budget for documentation, which is quite far behind esp. for concurrent programming. And so it goes.

We'll see where they get with ownership and inter-op with C/C++/Java. Concurrency + ownership + legacy + FFI inter-op => combinatoric complexity...

John23832 209 days ago | | |

> So, the compiler, stdlib and runtime, core libraries, build system ... not enough? What else would you want?

First class IDE support (Xcode is not that). Better documentation. The build system only half works. Better packaging. I can keep going.

The swift ecosystem often feels like just enough was done to lock in iOS devs (but not enough to actually provide a good developer experience) and then they stopped because Apple has no incentive to do more than that.

> I feel the problem is not what's in open source, but that the open-source community cannot really form, since no outsider can significantly change what the Apple contributors decide. Some of the peripheral projects have relatively free rein, but they can't compete e.g., with server libraries elsewhere.

So you agree. This is exactly the point I was making.

groundzeros2015 209 days ago | | |

Concurrency has dropped a complexity nuke on this language that it will never recover from.

randomNumber7 209 days ago | |

I think the problem is that C/C++ interop in practice sucks because they made pointers annoying to use (for ideological reasons. Pointer == Evil).

It's sad because technically they have amazing C/C++ interop, but using s.th. like SDL2 to write some toy game would be way less pain in C++.

You need to link against C libraries as a compiled system language and you just need a lot of pointers to do anything meaningfull in C.

jakey_bakey 209 days ago |

Thanks for sharing! :)

hooch 209 days ago |

tldr;

Before:

    let process = Process()
    process.executableURL = URL(fileURLWithPath: "/bin/ls")

    let pipe = Pipe()
    process.standardOutput = pipe

    try! process.run()
    process.waitUntilExit()
    let data = pipe.fileHandleForReading.readDataToEndOfFile()
    if let output = String(data: data, encoding: .utf8) {
        print(output)
    }

After:

    let result = try await run(
        .name("ls"),
        arguments: ["-1"],
        output: .string(limit: 1 << 20)
    )
    print(result.standardOutput ?? "")

fainpul 209 days ago | |

Why bother with invoking shell programs instead of using built-in ways of the language to do the same task?

  let result = try! FileManager.default.contentsOfDirectory(atPath: "/etc")
  print(result)

Various other FileManager methods are available:

https://developer.apple.com/documentation/foundation/fileman...

let process = Process() process.executableURL = URL(fileURLWithPath: "/bin/ls") let pipe = Pipe() process.standardOutput = pipe try! process.run() process.waitUntilExit() let data = pipe.fileHandleForReading.readDataToEndOfFile() if let output = String(data: data, encoding: .utf8) { print(output) }