[commons-lang] branch master updated: LANG-1172: Support dash as a delimiter in locales (#766)

[gg...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

ggregory pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/commons-lang.git


The following commit(s) were added to refs/heads/master by this push:
     new 35fb9fc  LANG-1172: Support dash as a delimiter in locales (#766)
35fb9fc is described below

commit 35fb9fca83af4ffbaf6f1163a927d8751938fda3
Author: Clemens Wolff <cl...@justamouse.com>
AuthorDate: Sat Mar 5 19:21:25 2022 -0500

    LANG-1172: Support dash as a delimiter in locales (#766)
    
    * LANG-1172: Support dash as a delimiter in locales
    
    * Extract constants for dash and underscore
    
    * Prioritize old delimiter in comparison order
---
 src/main/java/org/apache/commons/lang3/LocaleUtils.java     | 13 +++++++++----
 src/test/java/org/apache/commons/lang3/LocaleUtilsTest.java |  6 +++---
 2 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/src/main/java/org/apache/commons/lang3/LocaleUtils.java b/src/main/java/org/apache/commons/lang3/LocaleUtils.java
index 587d3a7..69f5eae 100644
--- a/src/main/java/org/apache/commons/lang3/LocaleUtils.java
+++ b/src/main/java/org/apache/commons/lang3/LocaleUtils.java
@@ -36,6 +36,8 @@ import java.util.concurrent.ConcurrentMap;
  * @since 2.2
  */
 public class LocaleUtils {
+    private static final char UNDERSCORE = '_';
+    private static final char DASH = '-';
 
     // class to avoid synchronization (Init on demand)
     static class SyncAvoid {
@@ -248,7 +250,9 @@ public class LocaleUtils {
             return new Locale(str);
         }
 
-        final String[] segments = str.split("_", -1);
+        final String[] segments = str.indexOf(UNDERSCORE) != -1
+            ? str.split(String.valueOf(UNDERSCORE), -1)
+            : str.split(String.valueOf(DASH), -1);
         final String language = segments[0];
         if (segments.length == 2) {
             final String country = segments[1];
@@ -289,6 +293,7 @@ public class LocaleUtils {
      *   LocaleUtils.toLocale("")           = new Locale("", "")
      *   LocaleUtils.toLocale("en")         = new Locale("en", "")
      *   LocaleUtils.toLocale("en_GB")      = new Locale("en", "GB")
+     *   LocaleUtils.toLocale("en-GB")      = new Locale("en", "GB")
      *   LocaleUtils.toLocale("en_001")     = new Locale("en", "001")
      *   LocaleUtils.toLocale("en_GB_xxx")  = new Locale("en", "GB", "xxx")   (#)
      * </pre>
@@ -300,7 +305,7 @@ public class LocaleUtils {
      * <p>This method validates the input strictly.
      * The language code must be lowercase.
      * The country code must be uppercase.
-     * The separator must be an underscore.
+     * The separator must be an underscore or a dash.
      * The length must be correct.
      * </p>
      *
@@ -325,7 +330,7 @@ public class LocaleUtils {
             throw new IllegalArgumentException("Invalid locale format: " + str);
         }
         final char ch0 = str.charAt(0);
-        if (ch0 == '_') {
+        if (ch0 == UNDERSCORE || ch0 == DASH) {
             if (len < 3) {
                 throw new IllegalArgumentException("Invalid locale format: " + str);
             }
@@ -340,7 +345,7 @@ public class LocaleUtils {
             if (len < 5) {
                 throw new IllegalArgumentException("Invalid locale format: " + str);
             }
-            if (str.charAt(3) != '_') {
+            if (str.charAt(3) != ch0) {
                 throw new IllegalArgumentException("Invalid locale format: " + str);
             }
             return new Locale(StringUtils.EMPTY, str.substring(1, 3), str.substring(4));
diff --git a/src/test/java/org/apache/commons/lang3/LocaleUtilsTest.java b/src/test/java/org/apache/commons/lang3/LocaleUtilsTest.java
index d8bd84d..112f5ba 100644
--- a/src/test/java/org/apache/commons/lang3/LocaleUtilsTest.java
+++ b/src/test/java/org/apache/commons/lang3/LocaleUtilsTest.java
@@ -170,12 +170,11 @@ public class LocaleUtilsTest  {
     @Test
     public void testToLocale_2Part() {
         assertValidToLocale("us_EN", "us", "EN");
+        assertValidToLocale("us-EN", "us", "EN");
         //valid though doesn't exist
         assertValidToLocale("us_ZH", "us", "ZH");
 
         assertThrows(
-                IllegalArgumentException.class, () -> LocaleUtils.toLocale("us-EN"), "Should fail as not underscore");
-        assertThrows(
                 IllegalArgumentException.class,
                 () -> LocaleUtils.toLocale("us_En"),
                 "Should fail second part not uppercase");
@@ -203,6 +202,7 @@ public class LocaleUtilsTest  {
     @Test
     public void testToLocale_3Part() {
         assertValidToLocale("us_EN_A", "us", "EN", "A");
+        assertValidToLocale("us-EN-A", "us", "EN", "A");
         // this isn't pretty, but was caused by a jdk bug it seems
         // http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4210525
         if (SystemUtils.isJavaVersionAtLeast(JAVA_1_4)) {
@@ -214,7 +214,7 @@ public class LocaleUtilsTest  {
         }
 
         assertThrows(
-                IllegalArgumentException.class, () -> LocaleUtils.toLocale("us_EN-a"), "Should fail as not underscore");
+                IllegalArgumentException.class, () -> LocaleUtils.toLocale("us_EN-a"), "Should fail as no consistent delimiter");
         assertThrows(
                 IllegalArgumentException.class, () -> LocaleUtils.toLocale("uu_UU_"), "Must be 3, 5 or 7+ in length");
     }