http://www.perlmonks.org?node_id=1010060

ABCXYZ has asked for the wisdom of the Perl Monks concerning the following question:

I've just downloaded Rakudo, and changing my Perl5 code into Perl6, but I've encountered some difficulties:

1. print '218' =~ s/^18/3/r

2. $_ = '2316'; substr($_, -1, 1) = 8; print

3. map {$s+=int $_*($i=abs int 999/$_)*++$i/2 },(3,5,-15);print $s

4.

LOOP: say q/Please input a number (>1):/; while(<>){ chomp; say $_; goto LOOP; }

How to make the above four scripts work in Perl6?

Replies are listed 'Best First'.
Re: How to change these small one-liners into Perl6 code?
by moritz (Cardinal) on Dec 23, 2012 at 08:22 UTC

    1:

    print '218'.subst(rx/^18/, 3);

    2: this is supposed to work with substr-rw, but doesn't with current rakudo.

    3: Sorry, I don't even understand what this code does in Perl 5. Modifying and using a variable in the same statement is very detrimal for readability.

    4: labels and goto aren't implemented in Rakudo. You can use prompt, or for lines() { } (since lines() is a lazy list, it doesn't read all the lines upfront; this was broken in some earlier Rakudo versions, but should be fine in the 2012-12 and possibly the 2012-11 releases).

    I can recommend hanging out in the #perl6 IRC channel, we try to be helpful and friendly (and we succeed most of the time :-).

      First, thank you very much.

      >1: print '218'.subst(rx/^18/, 3);

      Great, this works for me.
      It was a real effort for me to figure out how "s///r" works in Perl6... It turned out to have a completely different syntax, but I do miss the Perl5's version: it's so concise, and since it is like a mother language, it can't be obfuscated anyway. I hope Perl6 can still allow the Perl5's sed syntax, conforming to TMTOWTDI. :-)

      >2: this is supposed to work with substr-rw, but doesn't with current rakudo.

      I see.

      >3: Sorry, I don't even understand what this code does in Perl 5. Modifying and using a variable in the same statement is very detrimal for readability.

      I will explain these code in the next post.

      >4: labels and goto aren't implemented in Rakudo. You can use prompt, or for lines() { } (since lines() is a lazy list, it doesn't read all the lines upfront; this was broken in some earlier Rakudo versions, but should be fine in the 2012-12 and possibly the 2012-11 releases).

      I am using version 2012-11, and it is slurping. but I wonder is the parentheses with "lines()" mandatory? It doesn't seem like Perl's style though?

      >I can recommend hanging out in the #perl6 IRC channel, we try to be helpful and friendly (and we succeed most of the time :-).

      I will try that.
      I was simply translating my Perl5 one-liners on ProjectEuler into Perl6's code.

      These are all on its first problem:
      If we list all the natural numbers below 10 that are multiples of 3 or + 5, we get 3, 5, 6 and 9. The sum of these multiples is 23. Find the sum of all the multiples of 3 or 5 below 1000. (I generalized this to: below 10**$n for code #1 and #2, or below $N for code #3)
      This is my fastest solution in Perl5: (not in the code #1..4 above)
      perl5 -e '$n=3;print 2,3x--$n,1 .6x$n+2'

      The following is the origin of code #1, which works beyond the limit of 64 bit int, since it uses string op:
      $n=3;print 2,(3x--$n.1 .6x--$n.8)=~s/^18/3/r

      This is the origin of code #2:
      $n=3;$_=2 .3x--$n.1 .6x$n;substr($_,-1)+=2;print

      So code #3 is a more general solution for $N which is typically not an integer exponentiation of 10:
      map{$s+=int$_*($i=abs int 999/$_)*++$i/2}(3,5,-15);print$s

      They all give the same result: 233168 for $n=3, but work for other $n as well.
        https://github.com/perl6/perl6-examples/tree/master/euler should interest you.

        Five solutions have been added to this repo over the last 4 years. The latest one was added 18 days ago:

        say [+] grep * %% (3|5), ^1000;

        This means "say the sum of numbers that are divisible by 3 or 5 in the range zero up to (but not including) 1000".

Re: How to change these small one-liners into Perl6 code?
by ABCXYZ (Novice) on Dec 23, 2012 at 01:44 UTC
    I've made #4 work finally:

    while prompt qq/Please input a number (>1):\n/ -> $_ { say $_; }

    A lot of unexpected things:

    1. For some reason, Rakudo does not recognize the "LOOP:" construct, I just can't use "goto LOOP" to jump to a label.

    2. The simple "while(<>){}" idiom doesn't work anymore. Though I've heard that "for =<> {}" is one alternative, but it seems that Rakudo doesn't support it yet.

    3. The compiler suggest me to use "lines()" instead of "while(<>)", but "lines()" slurps all the input, so I can't input line by line, luckily there is a "prompt" function instead.

    4. Since "while(<>){}" does not work, the implicit variable is not assigned, so I have to explicitly assign it by: "-> $_".

    5. For some reason, bare function "chomp" and "say" does not work any more, I have to pass the parameter "$_" explicitly.

    6. But the chomp function is not necessary anymore, the NL is automatically chomped. This sounds like the "perl5 -l" option, I guess we will need a "perl6 -L" option to do the inverse in the future?

    I am surprised that Perl6 is so much different from Perl5.
    But the code does become much shorter and clearer, at least from this example.
      > 1. For some reason, Rakudo does not recognize the "LOOP:" construct, I just can't use "goto LOOP" to jump to a label.

      loop is a control keyword in perl6 replacing c-style For Loops, IIRC to avoid the confusion with Foreach Loops.

      You can't use a keyword as a label.

      Cheers Rolf

        >> 1. For some reason, Rakudo does not recognize the "LOOP:" construct, I just can't use "goto LOOP" to jump to a label.

        >loop is a control keyword in perl6 replacing c-style For Loops, IIRC to avoid the confusion with Foreach Loops.
        >You can't use a keyword as a label.

        But I was using "LOOP:" instead of "loop", the interpreter should have distinguished them.