« BackTell HN: Five random IndieWeb blog links on your terminalundefinedSubmitted by susam 3 days ago
  • darrenf 2 days ago

    Not specifically trying to play golf, just to let you know that `curl -w` is your friend for extracting data from the headers:

        for _ in {1..5} ; do
         curl -sSw '%header{location}\n' https://indieblog.page/random | \
             sed -e s/.utm_.*$//
    done
    • atomic128 2 days ago

      Random recent browsing is also the best way to read Hacker News.

      It's not addictive, gives you a quick sample of what people are thinking about recently, prevents your own biases from narrowing your view, etc.

      See https://rnsaffn.com/zg2/

      Each refresh gives you a random post, plus its parent thread, plus some detail about that post's author's history.

      • zimpenfish 2 days ago

        To save the grep, you can do:

            sed -Ene 's/^location: (.*)?utm.*$/\1/p'
        If you've got a newer `curl`, you can just print the header:

            curl -I -s -o /dev/null -w '%header{location}' <url>
        But you'll obviously still need the sed if you want to trim off the UTM cruft. Unless you want to go the route of `bash` substitutions.

            l="$(curl -I -s -o /dev/null -w '%header{location}' <url>)"
    echo "${l%%?utm*}"
        • trollied 2 days ago

          This thread is a virtual UA meet

        • porridgeraisin a day ago

          I'll do one better

          eval `llm "bash script to get 5 random links from indieblogs and print one on each line"`

          You could say it's a self mutating program.

        • teotimepacreau a day ago

          Indieweb lacks discoverability, a way to get random indie blog posts is exactly what the indiweb needs, great idea !

          • rednafi 2 days ago

            Yo, this shit could be a tight JavaScript app with a reactive frontend. Y’all geezers still fuckin’ with them shell commands? /s

            • ramon156 2 days ago

              Give this man an electron app!