Merge pull request #1038 from OwenSanzas/fix-xmltokener-unescapeentity

Fix input validation in XMLTokener.unescapeEntity()

Author: stleary

.gitignore

@@ -16,3 +16,6 @@ build
 /gradlew
 /gradlew.bat
 .gitmodules
+
+# ignore compiled class files
+*.class

src/main/java/org/json/XMLTokener.java

@@ -151,33 +151,108 @@ public Object nextEntity(@SuppressWarnings("unused") char ampersand) throws JSON
     /**
      * Unescape an XML entity encoding;
      * @param e entity (only the actual entity value, not the preceding & or ending ;
-     * @return
+     * @return the unescaped entity string
+     * @throws JSONException if the entity is malformed
      */
-    static String unescapeEntity(String e) {
+    static String unescapeEntity(String e) throws JSONException {
         // validate
         if (e == null || e.isEmpty()) {
             return "";
         }
         // if our entity is an encoded unicode point, parse it.
         if (e.charAt(0) == '#') {
-            int cp;
-            if (e.charAt(1) == 'x' || e.charAt(1) == 'X') {
-                // hex encoded unicode
-                cp = Integer.parseInt(e.substring(2), 16);
-            } else {
-                // decimal encoded unicode
-                cp = Integer.parseInt(e.substring(1));
+            if (e.length() < 2) {
+                throw new JSONException("Invalid numeric character reference: &#;");
             }
-            return new String(new int[] {cp},0,1);
-        } 
+            int cp = (e.charAt(1) == 'x' || e.charAt(1) == 'X')
+                ? parseHexEntity(e)
+                : parseDecimalEntity(e);
+            return new String(new int[] {cp}, 0, 1);
+        }
         Character knownEntity = entity.get(e);
-        if(knownEntity==null) {
+        if (knownEntity == null) {
             // we don't know the entity so keep it encoded
             return '&' + e + ';';
         }
         return knownEntity.toString();
     }
 
+    /**
+     * Parse a hexadecimal numeric character reference (e.g., "&#xABC;").
+     * @param e entity string starting with '#' (e.g., "#x1F4A9")
+     * @return the Unicode code point
+     * @throws JSONException if the format is invalid
+     */
+    private static int parseHexEntity(String e) throws JSONException {
+        // hex encoded unicode - need at least one hex digit after #x
+        if (e.length() < 3) {
+            throw new JSONException("Invalid hex character reference: missing hex digits in &#" + e.substring(1) + ";");
+        }
+        String hex = e.substring(2);
+        if (!isValidHex(hex)) {
+            throw new JSONException("Invalid hex character reference: &#" + e.substring(1) + ";");
+        }
+        try {
+            return Integer.parseInt(hex, 16);
+        } catch (NumberFormatException nfe) {
+            throw new JSONException("Invalid hex character reference: &#" + e.substring(1) + ";", nfe);
+        }
+    }
+
+    /**
+     * Parse a decimal numeric character reference (e.g., "&#123;").
+     * @param e entity string starting with '#' (e.g., "#123")
+     * @return the Unicode code point
+     * @throws JSONException if the format is invalid
+     */
+    private static int parseDecimalEntity(String e) throws JSONException {
+        String decimal = e.substring(1);
+        if (!isValidDecimal(decimal)) {
+            throw new JSONException("Invalid decimal character reference: &#" + decimal + ";");
+        }
+        try {
+            return Integer.parseInt(decimal);
+        } catch (NumberFormatException nfe) {
+            throw new JSONException("Invalid decimal character reference: &#" + decimal + ";", nfe);
+        }
+    }
+
+    /**
+     * Check if a string contains only valid hexadecimal digits.
+     * @param s the string to check
+     * @return true if s is non-empty and contains only hex digits (0-9, a-f, A-F)
+     */
+    private static boolean isValidHex(String s) {
+        if (s == null || s.isEmpty()) {
+            return false;
+        }
+        for (int i = 0; i < s.length(); i++) {
+            char c = s.charAt(i);
+            if (!((c >= '0' && c <= '9') || (c >= 'a' && c <= 'f') || (c >= 'A' && c <= 'F'))) {
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /**
+     * Check if a string contains only valid decimal digits.
+     * @param s the string to check
+     * @return true if s is non-empty and contains only digits (0-9)
+     */
+    private static boolean isValidDecimal(String s) {
+        if (s == null || s.isEmpty()) {
+            return false;
+        }
+        for (int i = 0; i < s.length(); i++) {
+            char c = s.charAt(i);
+            if (c < '0' || c > '9') {
+                return false;
+            }
+        }
+        return true;
+    }
+
 
     /**
      * <pre>{@code 

src/test/java/org/json/junit/XMLTest.java

@@ -1426,6 +1426,81 @@ public void clarifyCurrentBehavior() {
         assertEquals(jsonObject3.getJSONObject("color").getString("value"), "008E97");
     }
 
+    /**
+     * Tests that empty numeric character reference &#; throws JSONException.
+     * Previously threw StringIndexOutOfBoundsException.
+     * Related to issue #1035
+     */
+    @Test(expected = JSONException.class)
+    public void testEmptyNumericEntityThrowsJSONException() {
+        String xmlStr = "<a>&#;</a>";
+        XML.toJSONObject(xmlStr);
+    }
+
+    /**
+     * Tests that malformed decimal entity &#txx; throws JSONException.
+     * Previously threw NumberFormatException.
+     * Related to issue #1036
+     */
+    @Test(expected = JSONException.class)
+    public void testInvalidDecimalEntityThrowsJSONException() {
+        String xmlStr = "<a>&#txx;</a>";
+        XML.toJSONObject(xmlStr);
+    }
+
+    /**
+     * Tests that empty hex entity &#x; throws JSONException.
+     * Validates proper input validation for hex entities.
+     */
+    @Test(expected = JSONException.class)
+    public void testEmptyHexEntityThrowsJSONException() {
+        String xmlStr = "<a>&#x;</a>";
+        XML.toJSONObject(xmlStr);
+    }
+
+    /**
+     * Tests that invalid hex entity &#xGGG; throws JSONException.
+     * Validates hex digit validation.
+     */
+    @Test(expected = JSONException.class)
+    public void testInvalidHexEntityThrowsJSONException() {
+        String xmlStr = "<a>&#xGGG;</a>";
+        XML.toJSONObject(xmlStr);
+    }
+
+    /**
+     * Tests that valid decimal numeric entity &#65; works correctly.
+     * Should decode to character 'A'.
+     */
+    @Test
+    public void testValidDecimalEntity() {
+        String xmlStr = "<a>&#65;</a>";
+        JSONObject jsonObject = XML.toJSONObject(xmlStr);
+        assertEquals("A", jsonObject.getString("a"));
+    }
+
+    /**
+     * Tests that valid hex numeric entity &#x41; works correctly.
+     * Should decode to character 'A'.
+     */
+    @Test
+    public void testValidHexEntity() {
+        String xmlStr = "<a>&#x41;</a>";
+        JSONObject jsonObject = XML.toJSONObject(xmlStr);
+        assertEquals("A", jsonObject.getString("a"));
+    }
+
+    /**
+     * Tests that valid uppercase hex entity &#X41; works correctly.
+     * Should decode to character 'A'.
+     */
+    @Test
+    public void testValidUppercaseHexEntity() {
+        String xmlStr = "<a>&#X41;</a>";
+        JSONObject jsonObject = XML.toJSONObject(xmlStr);
+        assertEquals("A", jsonObject.getString("a"));
+    }
+
 }