filter/xml fix for LOGSTASH-2246: extract non-ascii content with xpath

lib/logstash/filters/xml.rb

@@ -115,7 +115,7 @@ def filter(event)
           unless value.nil?
             matched = true
             event[xpath_dest] ||= []
-            event[xpath_dest] << value.to_s
+            event[xpath_dest] << value.to_str
           end
         end # XPath.each
       end # @xpath.each

spec/filters/xml.rb

@@ -1,3 +1,4 @@
+# encoding: utf-8
 require "test_utils"
 require "logstash/filters/xml"
 
@@ -154,4 +155,22 @@
     end
   end
 
+  describe "parse correctly non ascii content with xpath" do
+    config <<-CONFIG
+    filter {
+      xml {
+        source => "xmldata"
+        target => "data"
+        xpath => [ "/foo/key/text()", "xpath_field" ]
+      }
+    }
+    CONFIG
+
+    # Single value
+    sample("xmldata" => '<foo><key>Français</key></foo>') do
+      insist { subject["tags"] }.nil?
+      insist { subject["xpath_field"]} == ["Français"]
+    end
+  end
+
 end