From: Przemyslaw Daniel <przemyslaw_daniel@...>
Date: 2007-07-12T19:40:56+09:00
Subject: Re: awk regexp search

> Note that this basic framework does not support awk's /regex1/,/regex2/
> notation that captures lines between (and including) the lines matching
> those regular expressions.
> 
>> Best regards,
>> baptiste
> --Greg

how about something like this simple implementation

# AWK in Ruby

module AWK
  class ClassAwk
      def initialize(filename = "")
        @NR = 0;            # record number
        @NF = 0;            # field number
        @FS = /\s+/;        # field separator
        @line = "";         # line matched
        @fields = [];       # fields of macthed line
        @trace = [];        # regexp trace

        if (filename == "")
          @file = ARGF;
        else
          @file = File.open(filename, "r").close;;
        end
      end

      #NOTE: every rule has to be in separate line
      #to get unique rule id
      def rule(regexp1, regexp2 = nil)
        msg = "regexp parameter must be Regexp";
        raise ArgumentError, msg unless regexp1.kind_of?(Regexp);

        if regexp2 == nil
          if @line =~ regexp1
            @fields = @line.split(@FS)
            yield
          end
        else
          raise ArgumentError, msg unless regexp2.kind_of?(Regexp);
          rule_id = /.+:([0-9]+)/.match(caller.first.to_s)[1].to_i;

          @trace[rule_id] = true if @line =~ regexp1;
          if @trace[rule_id]
            @fields = @line.split(@FS)
            yield
          end
          @trace[rule_id] = false if @line =~ regexp2;
        end
      end

      def analyze()
        @NR = 0;
        ARGF.each { |@line|
          @line = @line.chop;
          @NR += 1;
          yield
        }
      end

      #get paricular field
      def getField(index)
        output = "";

        if (index == 0)
          output = @line;
        else
          if index - 1 < @fields.length
            return @fields[index - 1];
          end
        end
      end

      #get NR (record number)
      def getNR
        return @NR;
      end

      #get number of fileds
      def getNF
        return @fields.length;
      end
  end
end

and an example how to use it:

require "awk.rb"

awk = AWK::ClassAwk.new();

awk.analyze() {
  awk.rule(/start1/, /stop1/) {
    print "1, NR:", awk.getNR(),", ";
    print "NF: ", awk.getNF(),", ", awk.getField(0), "\n";
  };
  awk.rule(/start2/, /stop2/) {
    print "2, NR:", awk.getNR(),", ";
    print "NF: ", awk.getNF(),", ", awk.getField(0), "\n";
  };
  awk.rule(/start1/) {
     print awk.getField(0);
  };
}





-- 
Posted via http://www.ruby-forum.com/.