From 1abde1749987775f2d8cb32d9d1cdd73a9b6cf22 Mon Sep 17 00:00:00 2001
From: khituras <chew@gmx.net>
Date: Tue, 20 Feb 2024 08:33:26 +0100
Subject: [PATCH] Version 2.6.2.

Add exception handling. Although the issue observed was eventually solved in the xml-mapper.
---
 jcore-jsbd-ae/pom.xml                              |  2 ++
 .../jcore/ae/jsbd/main/SentenceAnnotator.java      | 14 ++++++++++----
 2 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/jcore-jsbd-ae/pom.xml b/jcore-jsbd-ae/pom.xml
index fcde4c4f9..79b53ef66 100644
--- a/jcore-jsbd-ae/pom.xml
+++ b/jcore-jsbd-ae/pom.xml
@@ -14,6 +14,8 @@
         <version>2.6.1</version>
     </parent>
 
+    <version>2.6.2-SNAPSHOT</version>
+
     <build>
         <plugins>
             <plugin>
diff --git a/jcore-jsbd-ae/src/main/java/de/julielab/jcore/ae/jsbd/main/SentenceAnnotator.java b/jcore-jsbd-ae/src/main/java/de/julielab/jcore/ae/jsbd/main/SentenceAnnotator.java
index 583db41a1..2acfe21e8 100644
--- a/jcore-jsbd-ae/src/main/java/de/julielab/jcore/ae/jsbd/main/SentenceAnnotator.java
+++ b/jcore-jsbd-ae/src/main/java/de/julielab/jcore/ae/jsbd/main/SentenceAnnotator.java
@@ -202,10 +202,16 @@ public void process(JCas aJCas) throws AnalysisEngineProcessException {
                         while (start < end && (Character.isWhitespace(documentText.getCodensedText().charAt(start))))
                             ++start;
 
-                        // get the string between the current annotation borders and recognized sentences
-                        String textSpan = documentText.getCodensedText().substring(start, end);
-                        if (!StringUtils.isBlank(textSpan))
-                            doSegmentation(documentText, textSpan, start);
+                            String textSpan;
+                        try {
+                            // get the string between the current annotation borders and recognized sentences
+                            textSpan = documentText.getCodensedText().substring(start, end);
+                        } catch (Exception e) {
+                            LOGGER.error("Document text boundary error. Tried to get substring from {} to {} for text \"{}\". Text condensation is {}", start, end, documentText.getCodensedText(), cutAwayTypes != null && !cutAwayTypes.isEmpty());
+                            throw e;
+                        }
+                            if (!StringUtils.isBlank(textSpan))
+                                doSegmentation(documentText, textSpan, start);
                     }
 
                 } catch (ClassNotFoundException e) {