HBASE-27219 Change JONI encoding in RegexStringComparator (#4632)

Signed-off-by: Andrew Purtell <apurtell@apache.org>
apache · Jul 19, 2022 · ee56bca · ee56bca
1 parent 25c375e
commit ee56bca
Showing 1 changed file with 4 additions and 2 deletions.
diff --git a/hbase-client/src/main/java/org/apache/hadoop/hbase/filter/RegexStringComparator.java b/hbase-client/src/main/java/org/apache/hadoop/hbase/filter/RegexStringComparator.java
@@ -26,7 +26,7 @@
 import org.apache.yetus.audience.InterfaceAudience;
 import org.jcodings.Encoding;
 import org.jcodings.EncodingDB;
-import org.jcodings.specific.UTF8Encoding;
+import org.jcodings.specific.NonStrictUTF8Encoding;
 import org.joni.Matcher;
 import org.joni.Option;
 import org.joni.Regex;
@@ -312,7 +312,9 @@ public byte[] toByteArray() {
    * NOTE: Only the {@link Pattern} flags CASE_INSENSITIVE, DOTALL, and MULTILINE are supported.
    */
   static class JoniRegexEngine implements Engine {
-    private Encoding encoding = UTF8Encoding.INSTANCE;
+    // When using UTF8Encoding, an infinite loop can occur if an invalid UTF8 is encountered.
+    // Use NonStrictUTF8Encoding instead of UTF8Encoding to avoid the issue.
+    private Encoding encoding = NonStrictUTF8Encoding.INSTANCE;
     private String regex;
     private Regex pattern;