From eeb45e153d047d1e123d24a5a3ba783958dfb709 Mon Sep 17 00:00:00 2001
From: NAITOH Jun <naitoh@gmail.com>
Date: Sun, 14 Jan 2024 21:48:27 +0900
Subject: [PATCH] Change `@scanner.match` to respond `nil`/`@scanner` in order
 to improve processing speed.

---
 lib/rexml/parsers/baseparser.rb       |  3 +--
 lib/rexml/source.rb                   | 17 ++++++-------
 test/parse/test_entity_declaration.rb | 36 +++++++++++++++++++++++++++
 3 files changed, 45 insertions(+), 11 deletions(-)
 create mode 100644 test/parse/test_entity_declaration.rb

diff --git a/lib/rexml/parsers/baseparser.rb b/lib/rexml/parsers/baseparser.rb
index 5ce7b8d9..65bad260 100644
--- a/lib/rexml/parsers/baseparser.rb
+++ b/lib/rexml/parsers/baseparser.rb
@@ -274,8 +274,7 @@ def pull_event
             return [ :elementdecl, @source.match( ELEMENTDECL_PATTERN, true )[1] ]
 
           when ENTITY_START
-            match = @source.match( ENTITYDECL, true ).compact
-            match[0] = :entitydecl
+            match = [:entitydecl, *@source.match( ENTITYDECL, true ).captures.compact]
             ref = false
             if match[1] == '%'
               ref = true
diff --git a/lib/rexml/source.rb b/lib/rexml/source.rb
index 4de68731..390d0ad5 100644
--- a/lib/rexml/source.rb
+++ b/lib/rexml/source.rb
@@ -70,11 +70,10 @@ def read
 
     def match(pattern, cons=false)
       if cons
-        @scanner.scan(pattern)
+        @scanner.scan(pattern).nil? ? nil : @scanner
       else
-        @scanner.check(pattern)
+        @scanner.check(pattern).nil? ? nil : @scanner
       end
-      @scanner.matched? ? [@scanner.matched, *@scanner.captures] : nil
     end
 
     # @return true if the Source is exhausted
@@ -161,24 +160,24 @@ def read
 
     def match( pattern, cons=false )
       if cons
-        @scanner.scan(pattern)
+        md = @scanner.scan(pattern)
       else
-        @scanner.check(pattern)
+        md = @scanner.check(pattern)
       end
-      while !@scanner.matched? and @source
+      while md.nil? and @source
         begin
           @scanner << readline
           if cons
-            @scanner.scan(pattern)
+            md = @scanner.scan(pattern)
           else
-            @scanner.check(pattern)
+            md = @scanner.check(pattern)
           end
         rescue
           @source = nil
         end
       end
 
-      @scanner.matched? ? [@scanner.matched, *@scanner.captures] : nil
+      md.nil? ? nil : @scanner
     end
 
     def empty?
diff --git a/test/parse/test_entity_declaration.rb b/test/parse/test_entity_declaration.rb
new file mode 100644
index 00000000..e15deec6
--- /dev/null
+++ b/test/parse/test_entity_declaration.rb
@@ -0,0 +1,36 @@
+# frozen_string_literal: false
+require 'test/unit'
+require 'rexml/document'
+
+module REXMLTests
+  class TestParseEntityDeclaration < Test::Unit::TestCase
+    private
+    def xml(internal_subset)
+      <<-XML
+<!DOCTYPE r SYSTEM "urn:x-henrikmartensson:test" [
+#{internal_subset}
+]>
+<r/>
+      XML
+    end
+
+    def parse(internal_subset)
+      REXML::Document.new(xml(internal_subset)).doctype
+    end
+
+    def test_empty
+      exception = assert_raise(REXML::ParseException) do
+        parse(<<-INTERNAL_SUBSET)
+<!ENTITY>
+        INTERNAL_SUBSET
+      end
+      assert_equal(<<-DETAIL.chomp, exception.to_s)
+Malformed notation declaration: name is missing
+Line: 5
+Position: 72
+Last 80 unconsumed characters:
+ <!ENTITY>  ]> <r/>
+      DETAIL
+    end
+  end
+end