From 13ccfb3b3bf2aa4322f3f3a146dbbb73be53e253 Mon Sep 17 00:00:00 2001
From: Bernard Lambeau <blambeau@gmail.com>
Date: Tue, 21 Feb 2012 21:49:05 +0100
Subject: [PATCH] Keep a master offset in Match instead of slicing.

Instead of successively slicing strings in matches, we keep the
parsed text under @source, and an offset in the later in @offset.

Btw, that @offset might be very useful for keeping trace of the
localisation of the match in the source text (e.g. for subsequent
semantic passes; semantic error detection for instance).
---
 lib/citrus.rb      | 44 +++++++++++++++++++++++++-------------------
 test/match_test.rb | 17 ++++++++++++++++-
 2 files changed, 41 insertions(+), 20 deletions(-)

diff --git a/lib/citrus.rb b/lib/citrus.rb
index 1ce0387..ff534ac 100644
--- a/lib/citrus.rb
+++ b/lib/citrus.rb
@@ -636,7 +636,7 @@ def parse(string, options={})
         raise ParseError, input
       end
 
-      Match.new(string.slice(opts[:offset], length), events)
+      Match.new(string, events, opts[:offset])
     end
 
     # Tests whether or not this rule matches on the given +string+. Returns the
@@ -1239,14 +1239,11 @@ def to_citrus # :nodoc:
   # instantiated as needed. This class provides several convenient tree
   # traversal methods that help when examining and interpreting parse results.
   class Match
-    def initialize(string, events=[])
-      @string = string
+    def initialize(source, events=[], offset = 0)
+      @source = source
+      @offset = offset
 
       if events.length > 0
-        if events[-1] != string.length
-          raise ArgumentError, "Invalid events for length #{string.length}"
-        end
-
         elisions = []
 
         while events[0].elide?
@@ -1261,18 +1258,29 @@ def initialize(string, events=[])
         end
       else
         # Create a default stream of events for the given string.
-        events = [Rule.for(string), CLOSE, string.length]
+        events = [Rule.for(source), CLOSE, source.length]
       end
 
       @events = events
     end
 
+    # The main parsed text.
+    attr_reader :source
+
+    # The index of this match in the source text.
+    attr_reader :offset
+
     # The array of events for this match.
     attr_reader :events
 
     # Returns the length of this match.
     def length
-      @string.length
+      @events.last
+    end
+
+    # Returns the slice of the source text that this match captures.
+    def string
+      @string ||= @source[offset, length]
     end
 
     # Returns a hash of capture names to arrays of matches with that name,
@@ -1296,16 +1304,14 @@ def first
     # Allows methods of this match's string to be called directly and provides
     # a convenient interface for retrieving the first match with a given name.
     def method_missing(sym, *args, &block)
-      if @string.respond_to?(sym)
-        @string.__send__(sym, *args, &block)
+      if string.respond_to?(sym)
+        string.__send__(sym, *args, &block)
       else
         captures[sym].first
       end
     end
 
-    def to_s
-      @string
-    end
+    alias_method :to_s, :string
 
     # This alias allows strings to be compared to the string value of Match
     # objects. It is most useful in assertions in unit tests, e.g.:
@@ -1339,9 +1345,9 @@ def [](key, *args)
     def ==(other)
       case other
       when String
-        @string == other
+        string == other
       when Match
-        @string == other.to_s
+        string == other.to_s
       else
         super
       end
@@ -1350,7 +1356,7 @@ def ==(other)
     alias_method :eql?, :==
 
     def inspect
-      @string.inspect
+      string.inspect
     end
 
     # Prints the entire subtree of this match using the given +indent+ to
@@ -1372,7 +1378,7 @@ def dump(indent=' ')
           rule = stack.pop
 
           space = indent * (stack.size / 3)
-          string = @string.slice(os, event)
+          string = self.string.slice(os, event)
           lines[start] = "#{space}#{string.inspect} rule=#{rule}, offset=#{os}, length=#{event}"
 
           last_length = event unless last_length
@@ -1425,7 +1431,7 @@ def process_events!
             os = stack.pop
             start = stack.pop
 
-            match = Match.new(@string.slice(os, event), @events[start..index])
+            match = Match.new(source, @events[start..index], @offset + os)
             capture!(rule, match)
 
             if stack.size == 1
diff --git a/test/match_test.rb b/test/match_test.rb
index f45962a..0cce411 100644
--- a/test/match_test.rb
+++ b/test/match_test.rb
@@ -20,6 +20,14 @@ def test_match_inequality
     assert_equal(false, match2 == match1)
   end
 
+  def test_string
+    match1 = Match.new('abcdef')
+    assert_equal 'abcdef', match1.string
+
+    match2 = Match.new('abcdef', [Rule.for('bcd'), -1, 3], 1)
+    assert_equal 'bcd', match2.string
+  end
+
   def test_matches
     a = Rule.for('a')
     b = Rule.for('b')
@@ -59,11 +67,18 @@ def test_matches
       CLOSE, 3
     ]
 
-    match.matches.each do |m|
+    match.matches.each_with_index do |m, i|
       assert_equal(sub_events, m.events)
+      assert_equal(i*3, m.offset)
+      assert_equal(3, m.length)
+      assert_equal("abc", m.string)
       assert_equal("abc", m)
       assert(m.matches)
       assert_equal(3, m.matches.length)
+      m.matches.each_with_index do |m2,i2|
+        assert_equal(i*3+i2, m2.offset)
+        assert_equal(1, m2.length)
+      end
     end
   end