A summary of the changes in Apocalypse 5:

Unchanged features

• Capturing: (...)
• Repetition quantifiers: *, +, and ?
• Alternatives: |
• Backslash escape: \
• Minimal matching suffix: ??, *?, +?

Modifiers

• The extended syntax (/x) is no longer required...it's the default.


• There are no /s or /m modifiers (changes to the meta-characters replace them - see below).


• There is no /e evaluation modifier on substitutions; use s/pattern/$( code() )/ instead.


• The /g modifier has been renamed to e (for each).


• Modifiers are now placed as adverbs at the start of a match/substitution:

      @matches = m:ei/\s* (\w*) \s* ,?/;



• The single-character modifiers also have longer versions:

          :i        :ignorecase
          :e        :each



• The :c (or :cont) modifier causes the match to continue from the string's current .pos:

      m:c/ pattern /        # start at end of
                            # previous match on $_



• The new :o (:once) modifier replaces the Perl 5 ?...? syntax:

      m:once/ pattern /    # only matches first time



• The new :w (:word) modifier causes whitespace sequences to be replaced by \s* or \s+ subpattern:

      m:w/ next cmd =   <condition>/



• Same as:

      m/ \s* next \s+ cmd \s* = \s* <condition>/



• The new :uN modifier specifies Unicode level:

      m:u0/ .<2> /        # match two bytes
      m:u1/ .<2> /        # match two codepoints
      m:u2/ .<2> /        # match two graphemes
      m:u3/ .<2> /        # match language dependently



• The new :p5 modifier allows Perl 5 regex syntax to be used instead:

      m:p5/(?mi)^[a-z]{1,2}(?=\s)/



• Any integer modifier specifies a count. What kind of count is determined by the character that follows.


• If followed by an x, it means repetition:

      s:4x{ (<ident>) = (\N+) $$}{$1 => $2};

      # same as:

      s{ (<ident>) = (\N+) $$}{$1 => $2} for 1..4;



• If followed by an st, nd, rd, or th, it means find the Nth occurance:

      s:3rd/(\d+)/@data[$1]/;

      # same as:

      m/(\d+)/ && m:c/(\d+)/ && s:c/(\d+)/@data[$1]/;



• With the new :any modifier, the regex will match every possible way (including overlapping) and return all matches.

      $str = "abracadabra";

      @substrings = $str =~ m:any/ a (.*) a /;

      # br brac bracad bracadabr c cad cadabr d dabr br



• The :i, :w, :c, :uN, and :p5 modifiers can be placed inside the regex (and are lexically scoped):

      m/:c alignment = [:i left|right|cent[er|re]] /



• User-defined modifiers will be possible

          m:fuzzy/pattern/;



• Single letter flags can be ``chained'':

          s:ewi/cat/feline/



• User-defined modifiers can also take arguments:

          m:fuzzy('bare')/pattern/;



• Hence parentheses are no longer valid regex delimiters