• PunchingWood@lemmy.world
      link
      fedilink
      arrow-up
      19
      ·
      11 days ago

      I frequently had to test other developer’s code, the fucking amount of times I heard that shit.

      “Your code ain’t working”

      “It’s working for me”

      “Well it isn’t working for me, did you actually test it”

      “Yes”

      “On anything else than your own PC?”

      “It was working fine for me”

      “It fucking isn’t, look!”

      “…huh”

      For fuck sake just take like 1 minute to press on a frikking button to see if it actually does work.

      Glad I’m not doing that shit anymore.

      • ugo@feddit.it
        link
        fedilink
        arrow-up
        8
        ·
        edit-2
        11 days ago

        Im my defense, I did test it. It was working for me.

        But then someone else touched it without understanding it or flipped the truth value returned by one of the functions used by my code without properly fixing the call sites.

        Edit: or they had a merge conflict and they don’t know how to correctly resolve those.

    • 𝚜𝚑𝚊𝚍𝚎𝚊𝚛𝚐@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      ·
      11 days ago

      Meh. Pedant time.

      The first step of every testcase is supposed to fail to ensure a testing framework can fail in the scope of the testcase. In practice nobody acknowledges this, much less incorporates it.

      Without, one is forced to concede that the test environment is not sane—and, therefore, no case under test is certifiable—because it cannot be known, for certain, if failure was a possibility.

      Nobody cares.