Section for Week 2, Tuesday, 1/19/99

REGULAR EXPRESSIONS
-------------------

A Regular Expression (RE or regexp for short) is a pattern to be matched
against a string.  Much of the power and flexibility of Perl comes from REs,
so learning them is a key part of learning the language.

You may remember from the Unix warmup handout (B) the `grep' utility.  Using
it, you could find all the lines in a file containing a given string (eg,
you could use "grep oodle /usr/dict/words" to find all English words
containing "oodle": doodle, noodle, oodles, poodle, toodle.  This is a
simple example of pattern matching, which is the key use of REs.  In perl,
the same functionality would be achieved by:
while (<>) {		     # Iterate over all input lines
      if (/oodle/) {	     # Shorthand for if ($_ =~ /oodle/)
            print;	     # Shorthand for print $_;
      }
}

PATTERNS:
	Character Class:
		  Signified by square brackets: [AGH] will match any of A,
		  G, or H.  [A-Z] will match any uppercase letter, [0-9] any
		  digit.  [^0-9] will match any non-digit (^ means
		  negation).  
		  Predefined classes: \d == [0-9], \w == [a-zA-Z0-9_], 
		  \s == [ \r\n\f\t], caps for negation of these.
	Multipliers:
		* --> zero or more of the immediately preceding character 
		+ --> one or more of the immediately preceding character 
		? --> zero or one of the immediately preceding character 
	/x{5,10}/ --> between 5 and 10 x's (can omit second parameter, not
		      first) 
	Memory: 
		use parentheses to "remember" a pattern:
		/(.)\1/ will match any two consecutive identical non-newline
			characters. 
		(explain \1 \2 etc)
	Alternation:
		/Ami|ami/ will match either (read "or")
		/(song|Black)bird/
	Anchoring:
		\b --> word boundary (between \w and \W) (negation==\B)
		/\bred/ will match "hey redrobbin" but not "Fred"
		/^Bird/ will only match Bird at the beginning of the string.
		/bird$/ will only match Bird at the end of the string.
	Case:
		Append an 'i' after the closing slash to ignore case:
		/ami/i will match Ami, AMI, ami, aMi, etc.

OPERATORS:
	substitute: s/old/new/ will subst the first occurence of old with
		    new.  Add 'g' to end to make global.

	Transliteration:
		You should read about this (tr///) operator.

Other fun stuff about REs: see the perlre(1) man-page or ch2 of 
_Programming Perl_.