moltar512 has asked for the wisdom of the Perl Monks concerning the following question:

i was given the code: 
$A = -4;
$w = 500;
then

$re =
'/^\s*(\Q$A\E\s*\*?\s*\Qcos\E\s+\Q$w\E\s*\Qt\E\s+\QA\E||\Q$A\E\s*\*?\s
+*\Qcos\E\s*\(\s*\Q$w\E\s*\Qt\E\s*\)\s+\QA\E)\s*$/';

will accept answers in the following form:
-4cos 500t A
-4 cos 500t A
-4*cos 500t A
-4*cos(500t) A   -- and any number of spaces before and after *, t, ( 
+and )

[download]
It works great.. but that regular expression.. Jesus Christ.. Is there a way to dumb that down for me so that i can apply it to other situations, or would that take a complete O'Reilly book to figure out?

Replies are listed 'Best First'.
Re: is there an easy way to dumb down this regular expression for me?
by japhy (Canon) on Oct 19, 2005 at 20:26 UTC
    It also accepts an empty string, since you've got a doubled '|' in there. One | separates alternatives, not two.

    As for dumbing it down, I'd rewrite it like so: 

    $A_rx = qr{ \Q $A }x;
$w_rx = qr{ \Q $w }x;

$rx = qr{
  $A_rx \s*
  \*? \s*
  cos \s*
  (?:
    \( \s* $w_rx \s* \*? \s* t \s* \)
    | $w_rx \s* \*? \s* t \s
  ) \s*
  A
}x;

    [download]
    Then you can use it in the following manner: $string =~ /^\s*($rx)\s*$/.
    Jeff japhy Pinyan, P.L., P.M., P.O.D, X.S.: Perl, regex, and perl hacker
    How can we ever be the sold short or the cheated, we who for every service have long ago been overpaid? ~~ Meister Eckhart
[reply]
[d/l]
[select]

      A nit - you'll notice that $re is assigned '/.../', not a qr expression. I prefer your assignment of the qr// but its worth noting.

[reply]
[d/l]
      thanks! but i have to ask a few more questions yours is infintely more easy to read.. but now i'm trying to figure out what you did using http://www.perlmonks.net/?node=perlop what does the x behind all of your qr{ ... }x mean?
[reply]
        Look at perlre as well for the /x modifier:
        x Extend your pattern's legibility by permitting whitespace and comments.
        It was used so that everything could be spaced out for clarity -- that's also why al the \s* are in there because when using /x you have to explicitly say when you actually want to match whitespace.
[reply]
[d/l]
[select]
Re: is there an easy way to dumb down this regular expression for me?
by diotalevi (Canon) on Oct 19, 2005 at 20:41 UTC

    You had a bug. The alternation || is only written with *one* pipe character. The zero length pattern between each of the |s is also available so you could match anything that /^\s*$/ would match. The \Q...\E means quotemeta(). I've removed the \Q$A\E by doing quotemeta outside teh regexp. Since quotemeta has no effect on stuff like t and A, I removed it from those as well. I factored out the common parts of the expression so they aren't repeated. I also used the /x flag so I could add in as much extra whitespace as a I want - makes it easier to read. You'll notice how now everything except one branch of the alternation is followed by optional whitespace. That's to prevent "tA" from matching.

    $A = quotemeta -4;
$w = quotemeta 500;

  $re =
  '/^  \s*
   $A \s*
   -? \s*
   cos \s*
   (?: \( \d+ t \)
     | \d+ t \s ) \s*
   A \s*
   $/x;

    [download]
[reply]
[d/l]
Re: is there an easy way to dumb down this regular expression for me?
by GrandFather (Saint) on Oct 20, 2005 at 00:28 UTC

    Fixing the || bug, refactoring the expression and commenting gives this:

    use strict;
use warnings;

my $A = -4;
my $w = 500;

my $re = qr
  /
  ^\s*        #Start of line followed by any amount of white space
    \Q$A\E    #Litteral contents of $A
    \s*       #Any amount of white space
    \*?       #Optional *
    \s*cos    #Any amount of white space followed by cos
      \s*       #Any amount of white space
    (       #Start a capture group
      \s        #At least one white space character
      \Q$w\E    #Litteral contents of $w
      \s*t      #Any amount of white space followed by t
    |       #Match either expression (i.e. - with or without brackets)
      \(\s*     #( followed by any amount of white space
      \Q$w\E    #Litteral contents of $w
      \s*t      #Any amount of white space followed by t
      \s*\)     #Any amount of white space followed by )
    )       #End of capture group
    \s+A      #At least one white space character followed by A
  \s*$
  /x;

while (<DATA>)
  {
  print "Match: $_" if $_ =~ $re;
  }
  
__DATA__
-4cos 500t A
-4 cos 500t A
-4*cos 500t A
-4*cos(500t) A

    [download]
    Perl is Huffman encoded by design.
[reply]
[d/l]
[select]
Re: is there an easy way to dumb down this regular expression for me?
by ioannis (Abbot) on Oct 20, 2005 at 03:15 UTC
    Sometimes, it is more prudent to declare a pattern with our .

    Thanks to GrandFather who styled and commented the whole mess. The pattern now takes a huge space on my screen and the Perl code is mixed with typical ugly regexes. So, how do I reuse the pattern in another module; simple you say, declare it with our so the pattern is accessable from everywhere. Could we do better and avoid the our declaration? Yes, you say, declare a function that returns the regex, and all we have to do later is check for match using $_ =~ cos_pattern() . Lets declare the function:

     sub cos_pattern { $re }

    You see the problem? We have inadvertently created a closure.

    And how do we solve this problem: back declaring with our again: 

     our $re = qr / 
....big...
...multiline...
...pattern... 
/x;

    [download]
[reply]
[d/l]
[select]
Re: is there an easy way to dumb down this regular expression for me?
by Anonymous Monk on Oct 21, 2005 at 14:56 UTC
    YAPE::Regex::Explain might help 
    The regular expression:

(?-imsx:^\s*(\Q$A\E\s*\*?\s*\Qcos\E\s+\Q$w\E\s*\Qt\E\s+\QA\E||\Q$A\E\s
+*\*?\s*\Qcos\E\s*\(\s*\Q$w\E\s*\Qt\E\s*\)\s+\QA\E)\s*$)

matches as follows:

NODE                     EXPLANATION
----------------------------------------------------------------------
(?-imsx:                 group, but do not capture (case-sensitive)
                         (with ^ and $ matching normally) (with . not
                         matching \n) (matching whitespace and #
                         normally):
----------------------------------------------------------------------
  ^                        the beginning of the string
----------------------------------------------------------------------
  \s*                      whitespace (\n, \r, \t, \f, and " ") (0 or
                           more times (matching the most amount
                           possible))
----------------------------------------------------------------------
  (                        group and capture to \1:
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    $                        before an optional \n, and the end of
                             the string
----------------------------------------------------------------------
    A                        'A'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \*?                      '*' (optional (matching the most amount
                             possible))
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    cos                      'cos'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s+                      whitespace (\n, \r, \t, \f, and " ") (1
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    $                        before an optional \n, and the end of
                             the string
----------------------------------------------------------------------
    w                        'w'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    t                        't'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s+                      whitespace (\n, \r, \t, \f, and " ") (1
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    A                        'A'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
   |                        OR
----------------------------------------------------------------------
   |                        OR
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    $                        before an optional \n, and the end of
                             the string
----------------------------------------------------------------------
    A                        'A'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \*?                      '*' (optional (matching the most amount
                             possible))
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    cos                      'cos'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \(                       '('
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    $                        before an optional \n, and the end of
                             the string
----------------------------------------------------------------------
    w                        'w'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    t                        't'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
    \s*                      whitespace (\n, \r, \t, \f, and " ") (0
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \)                       ')'
----------------------------------------------------------------------
    \s+                      whitespace (\n, \r, \t, \f, and " ") (1
                             or more times (matching the most amount
                             possible))
----------------------------------------------------------------------
    \Q                       'Q'
----------------------------------------------------------------------
    A                        'A'
----------------------------------------------------------------------
    \E                       'E'
----------------------------------------------------------------------
  )                        end of \1
----------------------------------------------------------------------
  \s*                      whitespace (\n, \r, \t, \f, and " ") (0 or
                           more times (matching the most amount
                           possible))
----------------------------------------------------------------------
  $                        before an optional \n, and the end of the
                           string
----------------------------------------------------------------------
)                        end of grouping
----------------------------------------------------------------------

    [download]
[reply]
[d/l]

Back to Seekers of Perl Wisdom