From fa4c12cdc2547ad74ff1af6bcb4e7634e9d8a239 Mon Sep 17 00:00:00 2001
From: Victor Chang <vichang@google.com>
Date: Mon, 29 Mar 2021 23:52:43 +0100
Subject: [PATCH] ICU-21567 Avoid using regex in ULocale.getName()
No behavior change is expected
---
.../core/src/com/ibm/icu/util/ULocale.java | 35 ++++++++++++++++---
1 file changed, 31 insertions(+), 4 deletions(-)
diff --git a/icu4j/main/classes/core/src/com/ibm/icu/util/ULocale.java b/icu4j/main/classes/core/src/com/ibm/icu/util/ULocale.java
index 84472f5b301..c515ca1942b 100644
--- a/icu4j/main/classes/core/src/com/ibm/icu/util/ULocale.java
+++ b/icu4j/main/classes/core/src/com/ibm/icu/util/ULocale.java
@@ -28,7 +28,6 @@ import java.util.Objects;
import java.util.Set;
import java.util.TreeMap;
import java.util.TreeSet;
-import java.util.regex.Pattern;
import com.ibm.icu.impl.CacheBase;
import com.ibm.icu.impl.ICUData;
@@ -119,8 +118,6 @@ public final class ULocale implements Serializable, Comparable<ULocale> {
// using serialver from jdk1.4.2_05
private static final long serialVersionUID = 3715177670352309217L;
- private static final Pattern UND_PATTERN = Pattern.compile("^und(?=$|[_-])", Pattern.CASE_INSENSITIVE);
-
private static CacheBase<String, String, Void> nameCache = new SoftCache<String, String, Void>() {
@Override
protected String createInstance(String tmpLocaleID, Void unused) {
@@ -1144,11 +1141,41 @@ public final class ULocale implements Serializable, Comparable<ULocale> {
} else if ("root".equalsIgnoreCase(localeID)) {
tmpLocaleID = EMPTY_STRING;
} else {
- tmpLocaleID = UND_PATTERN.matcher(localeID).replaceFirst(EMPTY_STRING);
+ tmpLocaleID = stripLeadingUnd(localeID);
}
return nameCache.getInstance(tmpLocaleID, null /* unused */);
}
+ /**
+ * Strips out the leading "und" language code case-insensitively.
+ *
+ * @implNote Avoids creating new local non-primitive objects to reduce GC pressure.
+ */
+ private static String stripLeadingUnd(String localeID) {
+ int length = localeID.length();
+ if (length < 3) {
+ return localeID;
+ }
+
+ // If not starts with "und", return.
+ if (!localeID.regionMatches(/*ignoreCase=*/true, 0, "und", 0, /*len=*/3)) {
+ return localeID;
+ }
+
+ // The string is equals to "und" case-insensitively.
+ if (length == 3) {
+ return EMPTY_STRING;
+ }
+
+ // localeID must have a length >= 4
+ char separator = localeID.charAt(3);
+ if (separator == '-' || separator == '_') { // "und-*" or "und_*"
+ return localeID.substring(3);
+ }
+
+ return localeID;
+ }
+
/**
* Returns a string representation of this object.
* @return a string representation of the object.
--
2.40.0