Fix errors around strings with lone high or low surrogates (#1673)

This fixes some errors around treatment of strings with lone high/low
surrogates.

pkl-core/src/main/java/org/pkl/core/ast/expression/binary/SubscriptNode.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2024-2025 Apple Inc. and the Pkl project authors. All rights reserved.
+ * Copyright © 2024-2026 Apple Inc. and the Pkl project authors. All rights reserved.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -44,7 +44,9 @@ public abstract class SubscriptNode extends BinaryExpressionNode {
           .build();
     }
 
-    if (Character.isHighSurrogate(receiver.charAt(charIndex))) {
+    if (Character.isHighSurrogate(receiver.charAt(charIndex))
+        && charIndex < receiver.length() - 1
+        && Character.isLowSurrogate(receiver.charAt(charIndex + 1))) {
       return receiver.substring(charIndex, charIndex + 2);
     }
     return receiver.substring(charIndex, charIndex + 1);

pkl-core/src/main/java/org/pkl/core/runtime/VmUtils.java

@@ -675,13 +675,14 @@ public final class VmUtils {
     var charOffset = startIndex;
 
     while (charOffset < length && codePointOffset > 0) {
-      if (Character.isHighSurrogate(string.charAt(charOffset++))
+      var ch = string.charAt(charOffset);
+      charOffset++;
+      if (Character.isHighSurrogate(ch)
           && charOffset < length
-          && !Character.isLowSurrogate(string.charAt(charOffset++))) {
-        codePointOffset -= 2;
-      } else {
-        codePointOffset -= 1;
+          && Character.isLowSurrogate(string.charAt(charOffset))) {
+        charOffset++;
       }
+      codePointOffset--;
     }
 
     return codePointOffset != 0 ? -1 : charOffset;
@@ -692,13 +693,14 @@ public final class VmUtils {
     var charOffset = string.length();
 
     while (charOffset > 0 && codePointOffset > 0) {
-      if (Character.isLowSurrogate(string.charAt(--charOffset))
+      charOffset--;
+      char ch = string.charAt(charOffset);
+      if (Character.isLowSurrogate(ch)
           && charOffset > 0
-          && !Character.isHighSurrogate(string.charAt(--charOffset))) {
-        codePointOffset -= 2;
-      } else {
-        codePointOffset -= 1;
+          && Character.isHighSurrogate(string.charAt(charOffset - 1))) {
+        charOffset--;
       }
+      codePointOffset--;
     }
 
     return codePointOffset != 0 ? -1 : charOffset;

pkl-core/src/main/java/org/pkl/core/stdlib/base/StringNodes.java

@@ -220,7 +220,9 @@ public final class StringNodes {
       var charIndex = VmUtils.codePointOffsetToCharOffset(self, index);
       if (charIndex == -1 || charIndex == self.length()) return VmNull.withoutDefault();
 
-      if (Character.isHighSurrogate(self.charAt(charIndex))) {
+      if (Character.isHighSurrogate(self.charAt(charIndex))
+          && charIndex < self.length() - 1
+          && Character.isLowSurrogate(self.charAt(charIndex + 1))) {
         return self.substring(charIndex, charIndex + 2);
       }
       return self.substring(charIndex, charIndex + 1);

pkl-core/src/test/files/LanguageSnippetTests/input/api/string.pkl

@@ -415,6 +415,7 @@ examples {
     str1.substring(2, 3)
     str1.substring(2, 4)
     str1.substring(0, 7)
+    "\u{D800}hello".substring(0, 2)
     module.catch(() -> str1.substring(-1, 4))
     module.catch(() -> str1.substring(1, 8))
     module.catch(() -> str1.substring(3, 2))
@@ -436,6 +437,15 @@ examples {
     str1.getOrNull(3)
     str1.getOrNull(6)
     str1.getOrNull(7)
+    "🏀".getOrNull(0)
+    "🏀".getOrNull(1)
+    "\u{D800}".getOrNull(0)
+    "\u{D800}".getOrNull(1)
+    "\u{D800}h".getOrNull(0)
+    "\u{D800}h".getOrNull(1)
+    "\u{D800}h".getOrNull(2)
+    "h\u{D800}".getOrNull(0)
+    "h\u{D800}".getOrNull(1)
   }
 
   ["toCodePoints()"] {

pkl-core/src/test/files/LanguageSnippetTests/input/basic/string.pkl

@@ -59,6 +59,13 @@ examples {
     str3[2]
     module.catch(() -> str3[-1])
     module.catch(() -> str3[4])
+
+    "🏀"[0]
+    "\u{D800}"[0]
+    "\u{D800}h"[0]
+    "\u{D800}h"[1]
+    "h\u{D800}"[0]
+    "h\u{D800}"[1]
   }
 
   ["dollar sign has no special meaning"] {

pkl-core/src/test/files/LanguageSnippetTests/output/api/string.pcf

@@ -353,6 +353,7 @@ examples {
     "c"
     "cd"
     "abcdefg"
+    "?h"
     "Character index `-1` is out of range `0`..`7`. String: \"abcdefg\""
     "Character index `8` is out of range `1`..`7`. String: \"abcdefg\""
     "Character index `2` is out of range `3`..`7`. String: \"abcdefg\""
@@ -372,6 +373,15 @@ examples {
     "d"
     "g"
     null
+    "🏀"
+    null
+    "?"
+    null
+    "?"
+    "h"
+    null
+    "h"
+    "?"
   }
   ["toCodePoints()"] {
     List(97, 98, 99, 100, 101, 102, 103)

pkl-core/src/test/files/LanguageSnippetTests/output/basic/string.pcf

@@ -51,6 +51,12 @@ examples {
     "i"
     "Character index `-1` is out of range `0`..`3`. String: \"this\""
     "Character index `4` is out of range `0`..`3`. String: \"this\""
+    "🏀"
+    "?"
+    "?"
+    "h"
+    "h"
+    "?"
   }
   ["dollar sign has no special meaning"] {
     "123$"