Bump org.apache.pdfbox to 3.0.4 and guard against empty unicode strings (#3271)

Fixes: #3265

* Bump org.apache.pdfbox to 3.0.4

* Guard against empty unicode strings

Auto-cherry-pick to 1.0.x

Signed-off-by: David Frizelle <[email protected]>

---------

Signed-off-by: David Frizelle <[email protected]>

Author: dafriz

document-readers/pdf-reader/src/main/java/org/springframework/ai/reader/pdf/layout/CharacterFactory.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2023-2024 the original author or authors.
+ * Copyright 2023-2025 the original author or authors.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -94,7 +94,7 @@ private double numberOfSpacesBetweenTwoCharacters(final TextPosition textPositio
 
 	private char getCharacterFromTextPosition(final TextPosition textPosition) {
 		String string = textPosition.getUnicode();
-		char character = string.charAt(0);
+		char character = !string.isEmpty() ? string.charAt(0) : '\0';
 		return character;
 	}
 

pom.xml

@@ -280,7 +280,7 @@
 		<protobuf-java.version>3.25.2</protobuf-java.version>
 
 		<!-- readers/writer/stores dependencies-->
-		<pdfbox.version>3.0.3</pdfbox.version>
+		<pdfbox.version>3.0.4</pdfbox.version>
 		<pgvector.version>0.1.6</pgvector.version>
 		<sap.hanadb.version>2.20.11</sap.hanadb.version>
 		<coherence.version>24.09</coherence.version>