I saw this in someone's post: 


@monthPost = (
"-", "st", "nd", "rd", "th", "th", "th", "th", "th", 
"th", "th", "th", "th", "th", "th", "th", "th", "th",
"th", "th", "th", "st", "nd", "rd", "th", "th", "th",
"th", "th", "th", "th", "st"
); 

$textInput = "$mday$monthPost[$mday]" # etc.

[download]
and it reminded me that I once set off a little golf on that topic. The winner was this (sub) from Jonas Nilsson: 

$_[0]=~/1\d$/?0:qw(0 st nd rd)[$_[0]%10]or'th';

[download]
-- 
($_='jjjuuusssttt annootthheer
     pppeeerrrlll haaaccckkeer')=~y/a-z//s;print;

[download]

Replies are listed 'Best First'.
Re: English Date-Suffixes
by cLive ;-) (Prior) on Nov 04, 2002 at 07:54 UTC
    You can get it down another 6 chars to 41 by a little rewriting, and the $ in the regex and final semi-colon aren't needed: 
    shift;/1\d/?0:qw(0 st nd rd)[$_%10]or'th'

    [download]

    .02

    cLive ;-)

[reply]
[d/l]

      If you can live with the risk of the sub stomping on the callers copy of $_ by not localising it, then you could trim a couple more keystrokes by using pop instead of shift, and you also can trim the regex by another 1 

      pop;/1./?0:qw(0 st nd rd)[$_%10]or'th'

      [download]
      Nah! Your thinking of Simon Templar, originally played by Roger Moore and later by Ian Ogilvy
[reply]
[d/l]
      Is this what you want? (Neither pop nor shift load $_ by defalult -- they just default to using @_ in a sub or @ARGV in the main prog.)
      $_=pop;qw(0 st nd rd)[!/1./&&$_%10]or'th'

      [download]


        p
[reply]
[d/l]
Re: English Date-Suffixes
by BrowserUk (Patriarch) on Nov 04, 2002 at 05:19 UTC

    There are still a couple of loose ones in there. -2 or -3 if you count the superfluous ; 

    ($£=pop)=~/1./?0:qw(0 st nd rd)[$£%10]or'th'

    [download]

    My best alternate ain't even close, substr is too long. 

    sub nthit{
substr'thstndrd',($£=pop)=~/1./||($£%=10)>3?0:2*$£,2
}

    [download]
    Nah! Your thinking of Simon Templar, originally played by Roger Moore and later by Ian Ogilvy
[reply]
[d/l]
[select]
      What's with the accent marks and sterling pounds? I read that as a variable name, but why use two chars for an arbitrary name? Must be something special, or something else that got displayed wrong.
[reply]

        In essence, any single char scalar name would do.

        I started using $£ as a variable name in golf-type solutions as its the only punctuation global that isn't used for something somewhere by Perl itself, therefore it doesn't cause warnings with strict, and I don't have to localise it.

        Quite why one instance of the £ character got converted during the process of posting the code, and none of the others did I am not quite sure. I even tried to correct it by cut&pasting one of the others that shows correctly (in my browser at least) over the top of the one that comes out wrong, but the results where always the same. I have no idea how to correct it.

        Nah! Your thinking of Simon Templar, originally played by Roger Moore and later by Ian Ogilvy
[reply]
[d/l]
[select]
Re: English Date-Suffixes
by petral (Curate) on Nov 06, 2002 at 16:34 UTC
    Actually,
    pop=~/1?./;qw(0 st nd rd)[$&]or'th'

    [download]


      p
[reply]
[d/l]
Re: English Date-Suffixes
by petral (Curate) on Nov 07, 2002 at 15:43 UTC
    And qw(0 st nd rd)[pop=~/(1?.)$/]or"th" will do any integer.


    Note that for pre-5.6 perl you have to do (qw(...))[index]because qw was translated into  split' ','...' during compilation.

      p
[reply]
[d/l]
[select]

      small improvement:

      (0,st,nd,rd)[pop=~/1?./g]||th
        Wow, didn't know about dropping the qw!
        Why does the ||th  lead to only returning the last value from the /1?./g -indexed list?

        Update:  I see, it becomes a comma list which returns the last value ("the result of the last expression").
        see, A list returns its last, an array returns its weight

          p
[reply]
[d/l]
[select]

Back to Meditations