Lucene.Net.Index.Term: Reverted to original implementation, since Utf8ToString() does not throw an exception on invalid text.
diff --git a/src/Lucene.Net/Index/Term.cs b/src/Lucene.Net/Index/Term.cs
index 73b8407..1cf8dd7 100644
--- a/src/Lucene.Net/Index/Term.cs
+++ b/src/Lucene.Net/Index/Term.cs
@@ -1,5 +1,6 @@
using Lucene.Net.Support;
using System;
+using System.Text;
namespace Lucene.Net.Index
{
@@ -94,10 +95,11 @@
/// </summary>
public static string ToString(BytesRef termText)
{
+ // the term might not be text, but usually is. so we make a best effort
+ Encoding decoder = new UTF8Encoding(false, true);
try
{
- // LUCENENET specific: termText already has this handy UTF8ToString method, so we're using that instead of Encoding.UTF8.GetBytes()
- return termText.Utf8ToString();
+ return decoder.GetString(termText.Bytes, termText.Offset, termText.Length);
}
catch
{
