1. 35
  1. 7

    There’s a few things not included (that they do explicitly mention aren’t included for convenience reasons) that are also good to shove into text-based APIs for testing purposes. In particular, I’m thinking of

    • Nulls (U+0000)
    • Broken Unicode characters
    • Very very long strings, e.g. 1M+

    It’s an interesting and probably-useful dataset, although I hate to wonder how much stuff at work would break if we started integrating it into our test suites.

    1. 2

      What is the “Human injection” string for?

      1. 10
        #   Human injection
        #   Strings which may cause human to reinterpret worldview
        If you're reading this, you've been in a coma for almost 20 years now. We're trying a new technique. We don't know where this message will end up in your dream, but we hope it works. Please wake up, we miss you.

        I see it as a joke—a string which, if a “human” were to happen upon, would throw them for a bit of a loop. For this one in particular, someone reading it might question if they were actually in a coma :)

        1. 2

          Yeah, seems to me like the premise for such a test string would be to insert it into situations that ostensibly break the fourth wall of a given program.

          That’s kind of a black hole of digression. You could sit around and dream up all kinds of philosophical meta-discussions to envelop a test scenario. I’d wager that’s a problem better suited for a social-engineering branch of security analysis.

          Like, hypothetically speaking, what if a person were to insert text that’s intended to (based on insider information, or information gleaned from pre-attack reconnaissance) unspool inside a debug log, and confuse the hell out of a developer tasked with writing up a post-mortem using logs for an outage incident?

          1. 1

            The string “This statement is false” …has been known to derail quality minds for years.



        2. 2

          I thought this was going to be a swear words dictionary.