Hacker News Reader

Comment by moonchild | Hacker News Reader

See also: raku's grammars (https://docs.raku.org/language/regexes).

can't overstate how nice raku is, bit sharper learning curve but still

    raku:       ^ ( \d+ <?{ $/ <= 255 }> ) ** 4 % '.'  $
    slow raku:  ^ ( @(0..255) ) ** 4 % '.' $ 
    rulex:      <% ( :( range '0'-'255' ) '.' ){3} :( range '0'-'255' ) %>
    regex:      ^(?:(0|1[0-9]{0,2}|2(?:[0-4][0-9]?|5[0-5]?|[6-9])?|[3-9][0-9]?)\.)((0|1[0-9]{0,2}|2(?:[0-4][0-9]?|5[0-5]?|[6-9])?|[3-9][0-9]?))$

first one running code in the middle of the regex, to see if it matches a mathematical property, the second, running code to generate a [0|1|2|3|...|254|255] equivalent, which also matches as tightly as the rulex/regex code (e.g. no 00 allowed)

Seems like a case where I would just \d+\.\d+\.\d+\.\d+ and validate it in code

can still do that logic too, without leaving the parsing step, per say

    /^ (\d+) ** 4 % \. <?{ $/[0].all <= 255 }> $/

or if wanting it really explicit

    /^ (\d+) \. (\d+) \. (\d+) \. (\d+) <?{ $/.all <= 255 }> $/

None of those seem much easier to understand than the others. I suspect this is a classic example of "let's take something that is really messy in pcre and compare it other more brief equivalents" but I could easily turn the regex into something more readable and I can find more help for the standard regex syntax than I can for any alternative.

As other's have said, it is a minor arguably cosmetic improvement and something that is likely easier just to use a visual helper like regex101 to help you construct.

Reply on news.ycombinator.com