• @[email protected]
    link
    fedilink
    29 months ago

    You can do most things by combining simple cmdline tools. E.g. filter out some specific lines from all files in a directory, get the value after the second :, write those to another file and then sort, deduplicate and count them.

    This may sound complicated, but it’s pretty easy and fast if your are familiar with a shell. To be that efficient with your shell you want it to actually be powerful and not just a plain text input. Also writing cmdline tools is rather easy compared to a usable GUI tool.

    • @[email protected]
      link
      fedilink
      English
      19 months ago

      Genuine question: why not use grep, awk, sed, or any of the other gnu tools that can already do that?

      • @[email protected]
        link
        fedilink
        19 months ago

        You could, but maybe a good shell makes it easier than the external tool. Or maybe you use the shell to effectively combine the inputs and outputs of the other tools.

        • @[email protected]
          link
          fedilink
          English
          19 months ago

          I guess that’s convenient if you’re only ever on one machine, I prefer commands that work (almost) everywhere!

          • @[email protected]
            link
            fedilink
            29 months ago

            Me too. I could never get into nushell or fish because they’re not posix and I don’t need to learn two ways to do something.

      • @[email protected]
        link
        fedilink
        19 months ago

        That’s what I meant, using your shell to run command line tools to solve your issue at hand. And having a powerful shell with e.g. context dependend autocomplete (and a lot more) helps to speed up that task.