fix TestAutomaton.testRandomFinite test bug that was trying to StringsToAutomaton.build a Collection of BytesRefs containing a too-massive (> 1000 UTF-8 bytes) term; corrected the exception message to make it clear the limit is in UTF-8 bytes, not java (UTF-16) characters

lucene/core/src/java/org/apache/lucene/util/automaton/StringsToAutomaton.java

@@ -269,7 +269,7 @@ final class StringsToAutomaton {
       throw new IllegalArgumentException(
           "This builder doesn't allow terms that are larger than "
               + Automata.MAX_STRING_UNION_TERM_LENGTH
-              + " characters, got "
+              + " UTF-8 bytes, got "
               + current);
     }
     assert stateRegistry != null : "Automaton already built.";

lucene/core/src/test/org/apache/lucene/util/automaton/TestAutomaton.java

@@ -790,9 +790,18 @@ public class TestAutomaton extends LuceneTestCase {
     return null;
   }
 
+  private static boolean hasMassiveTerm(Collection<BytesRef> terms) {
+    for (BytesRef term : terms) {
+      if (term.length > Automata.MAX_STRING_UNION_TERM_LENGTH) {
+        return true;
+      }
+    }
+    return false;
+  }
+
   private Automaton unionTerms(Collection<BytesRef> terms) {
     Automaton a;
-    if (random().nextBoolean()) {
+    if (random().nextBoolean() || hasMassiveTerm(terms)) {
       if (VERBOSE) {
         System.out.println("TEST: unionTerms: use union");
       }

lucene/core/src/test/org/apache/lucene/util/automaton/TestStringsToAutomaton.java

@@ -103,7 +103,7 @@ public class TestStringsToAutomaton extends LuceneTestCase {
             .startsWith(
                 "This builder doesn't allow terms that are larger than "
                     + Automata.MAX_STRING_UNION_TERM_LENGTH
-                    + " characters"));
+                    + " UTF-8 bytes"));
 
     byte[] b1k = ArrayUtil.copyOfSubArray(b10k, 0, 1000);
     build(Collections.singleton(new BytesRef(b1k)), false); // no exception