Add local entity expansion limit methods (#202)

GitHub: fix GH-192 Add local entity expansion limit methods. - `REXML::Document#entity_expansion_limit=` - `REXML::Document#entity_expansion_text_limit=` - `REXML::Parsers::SAX2Parser#entity_expansion_limit=` - `REXML::Parsers::SAX2Parser#entity_expansion_text_limit=` - `REXML::Parsers::StreamParser#entity_expansion_limit=` - `REXML::Parsers::StreamParser#entity_expansion_text_limit=` - `REXML::Parsers::PullParser#entity_expansion_limit=` - `REXML::Parsers::PullParser#entity_expansion_text_limit=` --------- Co-authored-by: Sutou Kouhei <kou@clear-code.com>
ruby · Aug 26, 2024 · caec187 · caec187
1 parent 1c694d1
commit caec187
Show file tree

Hide file tree

Showing 12 changed files with 83 additions and 88 deletions.
diff --git a/lib/rexml/attribute.rb b/lib/rexml/attribute.rb
@@ -148,8 +148,9 @@ def to_s
     # have been expanded to their values
     def value
       return @unnormalized if @unnormalized
-      @unnormalized = Text::unnormalize( @normalized, doctype )
-      @unnormalized
+
+      @unnormalized = Text::unnormalize(@normalized, doctype,
+                                        entity_expansion_text_limit: @element&.document&.entity_expansion_text_limit)
     end
 
     # The normalized value of this attribute.  That is, the attribute with

diff --git a/lib/rexml/document.rb b/lib/rexml/document.rb
@@ -91,6 +91,8 @@ class Document < Element
     #
     def initialize( source = nil, context = {} )
       @entity_expansion_count = 0
+      @entity_expansion_limit = Security.entity_expansion_limit
+      @entity_expansion_text_limit = Security.entity_expansion_text_limit
       super()
       @context = context
       return if source.nil?
@@ -431,10 +433,12 @@ def Document::entity_expansion_text_limit
     end
 
     attr_reader :entity_expansion_count
+    attr_writer :entity_expansion_limit
+    attr_accessor :entity_expansion_text_limit
 
     def record_entity_expansion
       @entity_expansion_count += 1
-      if @entity_expansion_count > Security.entity_expansion_limit
+      if @entity_expansion_count > @entity_expansion_limit
         raise "number of entity expansions exceeded, processing aborted."
       end
     end

diff --git a/lib/rexml/entity.rb b/lib/rexml/entity.rb
@@ -71,9 +71,12 @@ def Entity::matches? string
     # Evaluates to the unnormalized value of this entity; that is, replacing
     # &ent; entities.
     def unnormalized
-      document.record_entity_expansion unless document.nil?
+      document&.record_entity_expansion
+
       return nil if @value.nil?
-      @unnormalized = Text::unnormalize(@value, parent)
+
+      @unnormalized = Text::unnormalize(@value, parent,
+                                        entity_expansion_text_limit: document&.entity_expansion_text_limit)
     end
 
     #once :unnormalized

diff --git a/lib/rexml/parsers/baseparser.rb b/lib/rexml/parsers/baseparser.rb
@@ -164,6 +164,8 @@ def initialize( source )
         @listeners = []
         @prefixes = Set.new
         @entity_expansion_count = 0
+        @entity_expansion_limit = Security.entity_expansion_limit
+        @entity_expansion_text_limit = Security.entity_expansion_text_limit
       end
 
       def add_listener( listener )
@@ -172,6 +174,8 @@ def add_listener( listener )
 
       attr_reader :source
       attr_reader :entity_expansion_count
+      attr_writer :entity_expansion_limit
+      attr_writer :entity_expansion_text_limit
 
       def stream=( source )
         @source = SourceFactory.create_from( source )
@@ -585,7 +589,7 @@ def unnormalize( string, entities=nil, filter=nil )
               end
               re = Private::DEFAULT_ENTITIES_PATTERNS[entity_reference] || /&#{entity_reference};/
               rv.gsub!( re, entity_value )
-              if rv.bytesize > Security.entity_expansion_text_limit
+              if rv.bytesize > @entity_expansion_text_limit
                 raise "entity expansion has grown too large"
               end
             else
@@ -627,7 +631,7 @@ def pop_namespaces_restore
 
       def record_entity_expansion(delta=1)
         @entity_expansion_count += delta
-        if @entity_expansion_count > Security.entity_expansion_limit
+        if @entity_expansion_count > @entity_expansion_limit
           raise "number of entity expansions exceeded, processing aborted."
         end
       end

diff --git a/lib/rexml/parsers/pullparser.rb b/lib/rexml/parsers/pullparser.rb
@@ -51,6 +51,14 @@ def entity_expansion_count
         @parser.entity_expansion_count
       end
 
+      def entity_expansion_limit=( limit )
+        @parser.entity_expansion_limit = limit
+      end
+
+      def entity_expansion_text_limit=( limit )
+        @parser.entity_expansion_text_limit = limit
+      end
+
       def each
         while has_next?
           yield self.pull

diff --git a/lib/rexml/parsers/sax2parser.rb b/lib/rexml/parsers/sax2parser.rb
@@ -26,6 +26,14 @@ def entity_expansion_count
         @parser.entity_expansion_count
       end
 
+      def entity_expansion_limit=( limit )
+        @parser.entity_expansion_limit = limit
+      end
+
+      def entity_expansion_text_limit=( limit )
+        @parser.entity_expansion_text_limit = limit
+      end
+
       def add_listener( listener )
         @parser.add_listener( listener )
       end

diff --git a/lib/rexml/parsers/streamparser.rb b/lib/rexml/parsers/streamparser.rb
@@ -18,6 +18,14 @@ def entity_expansion_count
         @parser.entity_expansion_count
       end
 
+      def entity_expansion_limit=( limit )
+        @parser.entity_expansion_limit = limit
+      end
+
+      def entity_expansion_text_limit=( limit )
+        @parser.entity_expansion_text_limit = limit
+      end
+
       def parse
         # entity string
         while true

diff --git a/lib/rexml/text.rb b/lib/rexml/text.rb
@@ -268,7 +268,8 @@ def inspect
     #   u = Text.new( "sean russell", false, nil, true )
     #   u.value   #-> "sean russell"
     def value
-      @unnormalized ||= Text::unnormalize( @string, doctype )
+      @unnormalized ||= Text::unnormalize(@string, doctype,
+                                          entity_expansion_text_limit: document&.entity_expansion_text_limit)
     end
 
     # Sets the contents of this text node.  This expects the text to be
@@ -411,11 +412,12 @@ def Text::normalize( input, doctype=nil, entity_filter=nil )
     end
 
     # Unescapes all possible entities
-    def Text::unnormalize( string, doctype=nil, filter=nil, illegal=nil )
+    def Text::unnormalize( string, doctype=nil, filter=nil, illegal=nil, entity_expansion_text_limit: nil )
+      entity_expansion_text_limit ||= Security.entity_expansion_text_limit
       sum = 0
       string.gsub( /\r\n?/, "\n" ).gsub( REFERENCE ) {
         s = Text.expand($&, doctype, filter)
-        if sum + s.bytesize > Security.entity_expansion_text_limit
+        if sum + s.bytesize > entity_expansion_text_limit
           raise "entity expansion has grown too large"
         else
           sum += s.bytesize

diff --git a/test/test_document.rb b/test/test_document.rb
@@ -31,16 +31,6 @@ def test_new
     end
 
     class EntityExpansionLimitTest < Test::Unit::TestCase
-      def setup
-        @default_entity_expansion_limit = REXML::Security.entity_expansion_limit
-        @default_entity_expansion_text_limit = REXML::Security.entity_expansion_text_limit
-      end
-
-      def teardown
-        REXML::Security.entity_expansion_limit = @default_entity_expansion_limit
-        REXML::Security.entity_expansion_text_limit = @default_entity_expansion_text_limit
-      end
-
       class GeneralEntityTest < self
         def test_have_value
           xml = <<XML
@@ -64,9 +54,8 @@ def test_have_value
             doc.root.children.first.value
           end
 
-          REXML::Security.entity_expansion_limit = 100
-          assert_equal(100, REXML::Security.entity_expansion_limit)
           doc = REXML::Document.new(xml)
+          doc.entity_expansion_limit = 100
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             doc.root.children.first.value
           end
@@ -95,9 +84,8 @@ def test_empty_value
             doc.root.children.first.value
           end
 
-          REXML::Security.entity_expansion_limit = 100
-          assert_equal(100, REXML::Security.entity_expansion_limit)
           doc = REXML::Document.new(xml)
+          doc.entity_expansion_limit = 100
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             doc.root.children.first.value
           end
@@ -118,12 +106,12 @@ def test_with_default_entity
 </member>
 XML
 
-          REXML::Security.entity_expansion_limit = 4
           doc = REXML::Document.new(xml)
+          doc.entity_expansion_limit = 4
           assert_equal("\na\na a\n<\n", doc.root.children.first.value)
 
-          REXML::Security.entity_expansion_limit = 3
           doc = REXML::Document.new(xml)
+          doc.entity_expansion_limit = 3
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             doc.root.children.first.value
           end
@@ -142,8 +130,8 @@ def test_entity_expansion_text_limit
 <member>&a;</member>
           XML
 
-          REXML::Security.entity_expansion_text_limit = 90
           doc = REXML::Document.new(xml)
+          doc.entity_expansion_text_limit = 90
           assert_equal(90, doc.root.children.first.value.bytesize)
         end
       end

diff --git a/test/test_pullparser.rb b/test/test_pullparser.rb
@@ -157,16 +157,6 @@ def test_peek
     end
 
     class EntityExpansionLimitTest < Test::Unit::TestCase
-      def setup
-        @default_entity_expansion_limit = REXML::Security.entity_expansion_limit
-        @default_entity_expansion_text_limit = REXML::Security.entity_expansion_text_limit
-      end
-
-      def teardown
-        REXML::Security.entity_expansion_limit = @default_entity_expansion_limit
-        REXML::Security.entity_expansion_text_limit = @default_entity_expansion_text_limit
-      end
-
       class GeneralEntityTest < self
         def test_have_value
           source = <<-XML
@@ -206,22 +196,21 @@ def test_empty_value
 </member>
           XML
 
-          REXML::Security.entity_expansion_limit = 100000
           parser = REXML::Parsers::PullParser.new(source)
+          parser.entity_expansion_limit = 100000
           while parser.has_next?
             parser.pull
           end
           assert_equal(11111, parser.entity_expansion_count)
 
-          REXML::Security.entity_expansion_limit = @default_entity_expansion_limit
           parser = REXML::Parsers::PullParser.new(source)
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             while parser.has_next?
               parser.pull
             end
           end
           assert do
-            parser.entity_expansion_count > @default_entity_expansion_limit
+            parser.entity_expansion_count > REXML::Security.entity_expansion_limit
           end
         end
 
@@ -239,14 +228,14 @@ def test_with_default_entity
 </member>
           XML
 
-          REXML::Security.entity_expansion_limit = 4
           parser = REXML::Parsers::PullParser.new(source)
+          parser.entity_expansion_limit = 4
           while parser.has_next?
             parser.pull
           end
 
-          REXML::Security.entity_expansion_limit = 3
           parser = REXML::Parsers::PullParser.new(source)
+          parser.entity_expansion_limit = 3
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             while parser.has_next?
               parser.pull
@@ -255,7 +244,7 @@ def test_with_default_entity
         end
 
         def test_with_only_default_entities
-          member_value = "&lt;p&gt;#{'A' * @default_entity_expansion_text_limit}&lt;/p&gt;"
+          member_value = "&lt;p&gt;#{'A' * REXML::Security.entity_expansion_text_limit}&lt;/p&gt;"
           source = <<-XML
 <?xml version="1.0" encoding="UTF-8"?>
 <member>
@@ -276,11 +265,11 @@ def test_with_only_default_entities
             end
           end
 
-          expected_value = "<p>#{'A' * @default_entity_expansion_text_limit}</p>"
+          expected_value = "<p>#{'A' * REXML::Security.entity_expansion_text_limit}</p>"
           assert_equal(expected_value, events['member'].strip)
           assert_equal(0, parser.entity_expansion_count)
           assert do
-            events['member'].bytesize > @default_entity_expansion_text_limit
+            events['member'].bytesize > REXML::Security.entity_expansion_text_limit
           end
         end
 
@@ -296,8 +285,8 @@ def test_entity_expansion_text_limit
 <member>&a;</member>
           XML
 
-          REXML::Security.entity_expansion_text_limit = 90
           parser = REXML::Parsers::PullParser.new(source)
+          parser.entity_expansion_text_limit = 90
           events = {}
           element_name = ''
           while parser.has_next?

diff --git a/test/test_sax.rb b/test/test_sax.rb
@@ -100,16 +100,6 @@ def test_sax2
     end
 
     class EntityExpansionLimitTest < Test::Unit::TestCase
-      def setup
-        @default_entity_expansion_limit = REXML::Security.entity_expansion_limit
-        @default_entity_expansion_text_limit = REXML::Security.entity_expansion_text_limit
-      end
-
-      def teardown
-        REXML::Security.entity_expansion_limit = @default_entity_expansion_limit
-        REXML::Security.entity_expansion_text_limit = @default_entity_expansion_text_limit
-      end
-
       class GeneralEntityTest < self
         def test_have_value
           source = <<-XML
@@ -147,18 +137,17 @@ def test_empty_value
 </member>
           XML
 
-          REXML::Security.entity_expansion_limit = 100000
           sax = REXML::Parsers::SAX2Parser.new(source)
+          sax.entity_expansion_limit = 100000
           sax.parse
           assert_equal(11111, sax.entity_expansion_count)
 
-          REXML::Security.entity_expansion_limit = @default_entity_expansion_limit
           sax = REXML::Parsers::SAX2Parser.new(source)
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             sax.parse
           end
           assert do
-            sax.entity_expansion_count > @default_entity_expansion_limit
+            sax.entity_expansion_count > REXML::Security.entity_expansion_limit
           end
         end
 
@@ -176,19 +165,19 @@ def test_with_default_entity
 </member>
           XML
 
-          REXML::Security.entity_expansion_limit = 4
           sax = REXML::Parsers::SAX2Parser.new(source)
+          sax.entity_expansion_limit = 4
           sax.parse
 
-          REXML::Security.entity_expansion_limit = 3
           sax = REXML::Parsers::SAX2Parser.new(source)
+          sax.entity_expansion_limit = 3
           assert_raise(RuntimeError.new("number of entity expansions exceeded, processing aborted.")) do
             sax.parse
           end
         end
 
         def test_with_only_default_entities
-          member_value = "&lt;p&gt;#{'A' * @default_entity_expansion_text_limit}&lt;/p&gt;"
+          member_value = "&lt;p&gt;#{'A' * REXML::Security.entity_expansion_text_limit}&lt;/p&gt;"
           source = <<-XML
 <?xml version="1.0" encoding="UTF-8"?>
 <member>
@@ -203,11 +192,11 @@ def test_with_only_default_entities
           end
           sax.parse
 
-          expected_value = "<p>#{'A' * @default_entity_expansion_text_limit}</p>"
+          expected_value = "<p>#{'A' * REXML::Security.entity_expansion_text_limit}</p>"
           assert_equal(expected_value, text_value.strip)
           assert_equal(0, sax.entity_expansion_count)
           assert do
-            text_value.bytesize > @default_entity_expansion_text_limit
+            text_value.bytesize > REXML::Security.entity_expansion_text_limit
           end
         end
 
@@ -223,8 +212,8 @@ def test_entity_expansion_text_limit
 <member>&a;</member>
           XML
 
-          REXML::Security.entity_expansion_text_limit = 90
           sax = REXML::Parsers::SAX2Parser.new(source)
+          sax.entity_expansion_text_limit = 90
           text_size = nil
           sax.listen(:characters, ["member"]) do |text|
             text_size = text.size