Ignore rexml examples on ruby/spec

Merged: https://github.com/ruby/ruby/pull/2832

spec/ruby/library/rexml/element/inspect_spec.rb

@@ -1,27 +1,30 @@
-require 'rexml/document'
 require_relative '../../../spec_helper'
 
-describe "REXML::Element#inspect" do
+ruby_version_is ''...'2.8' do
+  require 'rexml/document'
 
-  before :each do
-    @name = REXML::Element.new "name"
-  end
+  describe "REXML::Element#inspect" do
 
-  it "returns the node as a string" do
-    @name.inspect.should == "<name/>"
-  end
+    before :each do
+      @name = REXML::Element.new "name"
+    end
 
-  it "inserts '...' if the node has children" do
-    e = REXML::Element.new "last_name"
-    @name << e
-    @name.inspect.should == "<name> ... </>"
-    # This might make more sense but differs from MRI's default behavior
-    # @name.inspect.should == "<name> ... </name>"
-  end
+    it "returns the node as a string" do
+      @name.inspect.should == "<name/>"
+    end
 
-  it "inserts the attributes in the string" do
-    @name.add_attribute "language"
-    @name.attributes["language"] = "english"
-    @name.inspect.should == "<name language='english'/>"
+    it "inserts '...' if the node has children" do
+      e = REXML::Element.new "last_name"
+      @name << e
+      @name.inspect.should == "<name> ... </>"
+      # This might make more sense but differs from MRI's default behavior
+      # @name.inspect.should == "<name> ... </name>"
+    end
+
+    it "inserts the attributes in the string" do
+      @name.add_attribute "language"
+      @name.attributes["language"] = "english"
+      @name.inspect.should == "<name language='english'/>"
+    end
   end
 end