commit fd0e6a9515ec8eb6a8fc2f4a9dbe8a476fa29519
author Marc Kupietz <kupietz@ids-mannheim.de> Wed Sep 09 18:07:29 2020 +0200
committer Marc Kupietz <kupietz@ids-mannheim.de> Thu Sep 24 20:20:48 2020 +0200
tree f3fbd3e61901c2d8d3a263415f29e76fc46d561b
parent 1e882fb0d540f28f88576c61d200ad60ddc7bc88

Do not escape double quoutes inside raw_text elements

This is not necessary and breaks tokenization compatibility.

Change-Id: Ib43733cf7264ee07b010a3478e8c4b728f7bd708

script/tei2korapxml

diff --git a/script/tei2korapxml b/script/tei2korapxml
index ab1975c..eed322d 100755
--- a/script/tei2korapxml
+++ b/script/tei2korapxml
@@ -20,7 +20,7 @@
   unshift @INC, "$FindBin::Bin/../lib";
 };
 
-use KorAP::XML::TEI qw!remove_xml_comments escape_xml!;
+use KorAP::XML::TEI qw!remove_xml_comments escape_xml escape_xml_minimal!;
 use KorAP::XML::TEI::Tokenizer::External;
 use KorAP::XML::TEI::Tokenizer::Conservative;
 use KorAP::XML::TEI::Tokenizer::Aggressive;
@@ -364,7 +364,7 @@
             };
 
             # Encode and escape data
-            $data = escape_xml(encode( "UTF-8", $data ));
+            $data = escape_xml_minimal(encode( "UTF-8", $data ));
             # note: the index still refers to the 'single character'-versions,
             # which are counted as 1 (search for '&' in data.xml and see
             # corresponding indices in $_tokens_file)