ICU-13219 Address Mark's comment

unicode-org · Nov 21, 2023 · 503df57 · 503df57
1 parent 9cf2b52
commit 503df57
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 5 deletions.
diff --git a/icu4c/source/common/rbbi.cpp b/icu4c/source/common/rbbi.cpp
@@ -109,6 +109,10 @@ RuleBasedBreakIterator::RuleBasedBreakIterator(UDataMemory* udm, UBool isPhraseB
         // "thai-arab" to "[[:scx=thai:][:scx=arab:]]"
         UnicodeString udxs(u'[');
         for (size_t i = 0; i < items; i++) {
+            if (i > 0 && dxs[i*5-1] != u'-') {
+                status = U_ILLEGAL_ARGUMENT_ERROR;
+                return;
+            }
             // Special handling of zyyy
             // "The code Zyyy (Common) can be specified to exclude all scripts"
             if (uprv_strncmp(dxs+i*5, "zyyy", 4) == 0) {

diff --git a/icu4j/main/core/src/main/java/com/ibm/icu/text/RuleBasedBreakIterator.java b/icu4j/main/core/src/main/java/com/ibm/icu/text/RuleBasedBreakIterator.java
@@ -102,13 +102,11 @@ public static RuleBasedBreakIterator getInstanceFromCompiledRules(InputStream is
      * @param phraseBreaking a flag indicating if phrase breaking is required.
      * @param dxValues Dictionary break script exclusions.
      * @throws IOException if there is an error while reading the rules from the buffer.
-     *         IllegalArgumentException if the dxValues is not null nor a String in the supported
-     *         format.
      * @see    #compileRules(String, OutputStream)
      * @internal
      */
     /* package-potected */ static RuleBasedBreakIterator getInstanceFromCompiledRules(
-            ByteBuffer bytes, boolean phraseBreaking, String dxValues) throws IOException, IllegalArgumentException {
+            ByteBuffer bytes, boolean phraseBreaking, String dxValues) throws IOException {
         RuleBasedBreakIterator instance = getInstanceFromCompiledRules(bytes);
         instance.fPhraseBreaking = phraseBreaking;
         instance.fDX = makeExcludedDictionaryBreakUnicodeSet(dxValues);
@@ -118,11 +116,10 @@ public static RuleBasedBreakIterator getInstanceFromCompiledRules(InputStream is
     /**
      * Crate a UnicodeSet for the Dictionary Break Script Exclusions.
      * @param dxValues Dictionary break script exclusions, a string of Script code joined by "-".
-     * @throws IOException if there is an error while constr the rules from the buffer.
      * @internal
      */
     private static UnicodeSet makeExcludedDictionaryBreakUnicodeSet(
-        String dxs) throws IllegalArgumentException {
+        String dxs) {
         if (dxs == null) {
             return null;
         }
@@ -133,6 +130,9 @@ private static UnicodeSet makeExcludedDictionaryBreakUnicodeSet(
         StringBuilder builder = new StringBuilder("[");
         int items = 1 + (dxs.length() / 5);
         for (int i = 0; i < items; i++) {
+            if (i > 0 && dxs.charAt(i*5-1) != '-') {
+                throw new IllegalArgumentException("Incorrect value for dx key: " + dxs);
+            }
             String script = dxs.substring(i*5, i*5+4);
             // Special handling of zyyy
             // "The code Zyyy (Common) can be specified to exclude all scripts"
@@ -142,6 +142,10 @@ private static UnicodeSet makeExcludedDictionaryBreakUnicodeSet(
             builder.append("[:scx=").append(script).append(":]");
         }
         builder.append("]");
+        // The UnicodeSet constructor will catch malformed dx values below.
+        // For example, if the locale is "en-u-dx-abc-defgh", dxs is "abc-defgh"
+        // and builder.toString() return "[[:scx=abc-:][:scx=efgh:]]" and causes
+        // UnicodeSet constructor to throw IllegalArgumentException
         return new UnicodeSet(builder.toString());
     }