Regex::PreSuf - create regular expressions from word lists

	use Regex::PreSuf;

	my $re = presuf(qw(foobar fooxar foozap));

	# $re should be now 'foo(?:zap|[bx]ar)'

The presuf() subroutine builds regular expressions out of 'word lists', lists of strings. The regular expression matches the same words as the word list. These regular expressions normally run faster than a simple-minded '|'-concatenation of the words.

Examples:

• 	'foobar fooxar' => 'foo[bx]ar'
  
  

• 	'foobar foozap' => 'foo(?:bar|zap)'
  
  

• 	'foobar fooar'  => 'foob?ar'
  
  

The downsides:

• The original order of the words is not necessarily respected, for example because the character class matches are collected together, separate from the '|' alternations.
• The module blithely ignores any specialness of any regular expression metacharacters such as the .*?+{}[]^$, they are just plain ordinary boring characters.

For the second downside there is an exception. The module has some rudimentary grasp of how to use the 'any character' metacharacter. If you call presuf() like this:

	my $re = presuf({ anychar=>1 }, qw(foobar foo.ar fooxar));

	# $re should be now 'foo.ar'

The module finds out the common prefixes and suffixes of the words and then recursively looks at the remaining differences. However, by default only common prefixes are used because for many languages (natural or artificial) this seems to produce the fastest matchers. To allow also for suffixes use

	my $re = presuf({ suffixes=>1 }, ...);

To use only suffixes use

	my $re = presuf({ prefixes=>0 }, ...);

(this implicitly enables suffixes)

In case you want to flood your session without debug messages you can turn on debugging by saying

	Regex::PreSuf::debug(1);

How to turn them off again is left as an exercise for the kind reader.

Jarkko Hietaniemi

This code is distributed under the same copyright terms as Perl itself.