Hacker News Reader

Comment by vollmond | Hacker News Reader

I haven't seen the "words of code" metric before, and some cursory web searching doesn't reveal much. Is there a standard? Just split on the given language's punctuation/delimiters?

View on HN

I wondered that too, then figured you'd want to exclude non /[a-z0-9]/I characters.

They answer what it means and provide the command used to figure it out in the first FAQ: https://tinyssh.org/faq.html

Just use wc, it's close enough.

    $ wc
    void main(int argc, char **argv) { printf("Hi, I'm %s\n", argv[0]); }
          1      11      70
      lines   words   chars

    $ tr '()[]{}*+.>,";'"'"- ' ' | wc
    void main(int argc, char **argv) { printf("Hi, I'm %s\n", argv[0]); }
          1      13      70

It's also interesting to see 'Postquantum crypto'.

I remember first seeing it somewhere on cr.yp.to, where it was counted with some regex that did something like splitting it on some delimiter-like characters.

Reply on news.ycombinator.com