From 44107948b94682334c8cd3e55f7e8c9c28218c88 Mon Sep 17 00:00:00 2001
From: Ko van der Sloot <Ko.vanderSloot@ziggo.nl>
Date: Fri, 21 Apr 2023 11:22:39 +0200
Subject: [PATCH] fixed a problem comparable to
 https://github.com/proycon/python-ucto/issues/16 (for XmlInput)

---
 src/tokenize.cxx | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/tokenize.cxx b/src/tokenize.cxx
index dd63350..7c910a1 100644
--- a/src/tokenize.cxx
+++ b/src/tokenize.cxx
@@ -1847,6 +1847,7 @@ namespace Tokenizer {
   }
 
   folia::Document *TokenizerClass::tokenize_folia( const string& infile_name ){
+    reset(); // when starting a new inputfile, we must reset provenance et.al.
     if ( inputclass == outputclass
 	 && !doWordCorrection ){
       DBG << "ucto: --filter=NO is automatically set. inputclass equals outputclass!"