Merge pull request #167 from bulasevich/GR-59497

[Backport][GR-59497] TRegex: fix OracleDB flavor incorrectly parsing escaped surrogate pairs.

Author: zakkak

regex/src/com.oracle.truffle.regex.test/src/com/oracle/truffle/regex/tregex/test/OracleDBTests.java

@@ -1387,6 +1387,7 @@ public void generatedTests() {
         test("()(a*\\1+)*", "", "aaa", 0, true, 0, 3, 0, 0, 3, 3);
         test("(a(\\2b|)?)+\\1c", "", "aaabaaac", 0, true, 0, 8, 5, 6, 6, 6);
         test("((|ab)+?w\\Z|^c)de()d", "", "ffffff", 0, false);
+        test("\u0282\\\ud807\udfdd+\u1cf2", "", "\u0282\ud807\udfdd\ud807\udfdd\u1cf2", 0, true, 0, 13);
         test("(a{1100,1100})\\1", "i", "a".repeat(2400), 0, true, 0, 2200, 0, 1100);
 
         /* GENERATED CODE END - KEEP THIS MARKER FOR AUTOMATIC UPDATES */

regex/src/com.oracle.truffle.regex/src/com/oracle/truffle/regex/tregex/parser/flavors/OracleDBRegexLexer.java

@@ -424,7 +424,7 @@ protected int parseCustomEscapeChar(char c, boolean inCharClass) {
         } else {
             // outside character classes, all escaped characters are simply treated as literals,
             // there are no escape sequences in oracleDB
-            return c;
+            return Character.isHighSurrogate(c) ? finishSurrogatePair(c) : c;
         }
     }