PDFBOX-5747: Surrogate pairs with combining diacritics are incorrectly ordered on text extraction

- Changed TextPosition.insertDiacritic() to preserve surrogate pairs
- Added unit test
- Included example test PDF file attached to PDFBOX-5747

Author: reckart

pdfbox/src/main/java/org/apache/pdfbox/text/TextPosition.java

@@ -689,7 +689,7 @@ public void mergeDiacritic(TextPosition diacritic)
                 break;
             }
             float currCharXEnd = currCharXStart + widths[i];
-
+                        
              // this is the case where there is an overlap of the diacritic character with the
              // current character and the previous character. If no previous character, just append
              // the diacritic after the current one
@@ -759,16 +759,25 @@ private void insertDiacritic(int i, TextPosition diacritic)
         float[] widths2 = new float[widths.length + 1];
         System.arraycopy(widths, 0, widths2, 0, i);
 
+        // First we add a zero-width entry for the diacritic in the widths array
+        widths2[i] = widths[i];
+        widths2[i + 1] = 0;
+        System.arraycopy(widths, i + 1, widths2, i + 2, widths.length - i - 1);
+
         // Unicode combining diacritics always go after the base character, regardless of whether
         // the string is in presentation order or logical order
         sb.append(unicode.charAt(i));
-        widths2[i] = widths[i];
+        
+        // If a surrogate starts at the current position, make sure we preserve it
+        if (i < unicode.length() - 1 && Character.isSurrogatePair(unicode.charAt(i), unicode.charAt(i + 1))) {
+            sb.append(unicode.charAt(i + 1));
+            i++;
+        }
+
         sb.append(combineDiacritic(diacritic.getUnicode()));
-        widths2[i + 1] = 0;
 
         // get the rest of the string
         sb.append(unicode.substring(i + 1));
-        System.arraycopy(widths, i + 1, widths2, i + 2, widths.length - i - 1);
 
         unicode = sb.toString();
         widths = widths2;

pdfbox/src/test/resources/input/PDFBOX-5747-unicode-surrogate-with-diacritic.pdf

@@ -0,0 +1,3 @@
+Firefox file:///home/pablo/invchar
+𝑋̂
+1 of 1 18/12/2023, 12:49

pdfbox/src/test/resources/input/PDFBOX-5747-unicode-surrogate-with-diacritic.pdf-sorted.txt

@@ -0,0 +1,3 @@
+𝑋̂
+Firefox file:///home/pablo/invchar
+1 of 1 18/12/2023, 12:49