From ea810cbbbde4de3a2b93a38aed53d8994b3b44af Mon Sep 17 00:00:00 2001
From: Matthew Jason Benson <mbenson@apache.org>
Date: Fri, 11 Jan 2008 20:13:02 +0000
Subject: [PATCH] [LANG-192] addSplitByCharacterType and camelCase variant

git-svn-id: https://svn.apache.org/repos/asf/commons/proper/lang/trunk@611288 13f79535-47bb-0310-9956-ffa450edef68
---
 .../org/apache/commons/lang/StringUtils.java  | 79 +++++++++++++++++++
 .../apache/commons/lang/StringUtilsTest.java  | 48 ++++++++++-
 2 files changed, 126 insertions(+), 1 deletion(-)

diff --git a/src/java/org/apache/commons/lang/StringUtils.java b/src/java/org/apache/commons/lang/StringUtils.java
index c9c0092ee..bd72bb859 100644
--- a/src/java/org/apache/commons/lang/StringUtils.java
+++ b/src/java/org/apache/commons/lang/StringUtils.java
@@ -2607,6 +2607,85 @@ private static String[] splitWorker(String str, String separatorChars, int max,
         return (String[]) list.toArray(new String[list.size()]);
     }
 
+    /**
+     * <p>Splits a String by Character type as returned by
+     * <code>java.lang.Character.getType(char)</code>. Groups of contiguous
+     * characters of the same type are returned as complete tokens. 
+     * <pre>
+     * StringUtils.splitByCamelCase(null)         = null
+     * StringUtils.splitByCamelCase("")           = []
+     * StringUtils.splitByCamelCase("ab de fg")   = ["ab", " ", "de", " ", "fg"]
+     * StringUtils.splitByCamelCase("ab   de fg") = ["ab", "   ", "de", " ", "fg"]
+     * StringUtils.splitByCamelCase("ab:cd:ef")   = ["ab", ":", "cd", ":", "ef"]
+     * StringUtils.splitByCamelCase("fooBar")     = ["foo", "B", "ar"]
+     * StringUtils.splitByCamelCase("foo200Bar")  = ["foo", "200", "B", "ar"]
+     * StringUtils.splitByCamelCase("ASFRules")   = ["ASFR", "ules"]
+     * </pre>
+     * @param str the String to split, may be <code>null</code>
+     * @return an array of parsed Strings, <code>null</code> if null String input
+     * @since 2.4
+     */
+    public static String[] splitByCharacterType(String str) {
+        return splitByCharacterType(str, false);
+    }
+
+    /**
+     * <p>Splits a String by Character type as returned by
+     * <code>java.lang.Character.getType(char)</code>. Groups of contiguous
+     * characters of the same type are returned as complete tokens, with the
+     * following exception: if <code>camelCase</code> is <code>true</code>,
+     * the character of type <code>Character.UPPERCASE_LETTER</code>, if any,
+     * immediately preceding a token of type <code>Character.LOWERCASE_LETTER</code>
+     * will belong to the following token rather than to the preceding, if any,
+     * <code>Character.UPPERCASE_LETTER</code> token. 
+     * <pre>
+     * StringUtils.splitByCamelCase(null)         = null
+     * StringUtils.splitByCamelCase("")           = []
+     * StringUtils.splitByCamelCase("ab de fg")   = ["ab", " ", "de", " ", "fg"]
+     * StringUtils.splitByCamelCase("ab   de fg") = ["ab", "   ", "de", " ", "fg"]
+     * StringUtils.splitByCamelCase("ab:cd:ef")   = ["ab", ":", "cd", ":", "ef"]
+     * StringUtils.splitByCamelCase("fooBar")     = ["foo", "Bar"]
+     * StringUtils.splitByCamelCase("foo200Bar")  = ["foo", "200", "Bar"]
+     * StringUtils.splitByCamelCase("ASFRules")   = ["ASF", "Rules"]
+     * </pre>
+     * @param str the String to split, may be <code>null</code>
+     * @param camelCase whether to use so-called "camel-case" for letter types
+     * @return an array of parsed Strings, <code>null</code> if null String input
+     * @since 2.4
+     */
+    public static String[] splitByCharacterType(String str, boolean camelCase) {
+        if (str == null) {
+            return null;
+        }
+        if (str.length() == 0) {
+            return ArrayUtils.EMPTY_STRING_ARRAY;
+        }
+        char[] c = str.toCharArray();
+        List list = new ArrayList();
+        int tokenStart = 0;
+        int currentType = Character.getType(c[tokenStart]);
+        for (int pos = tokenStart + 1; pos < c.length; pos++) {
+            int type = Character.getType(c[pos]);
+            if (type == currentType) {
+                continue;
+            }
+            if (camelCase && type == Character.LOWERCASE_LETTER
+                    && currentType == Character.UPPERCASE_LETTER) {
+                int newTokenStart = pos - 1;
+                if (newTokenStart != tokenStart) {
+                    list.add(new String(c, tokenStart, newTokenStart - tokenStart));
+                    tokenStart = newTokenStart;
+                }
+            } else {
+                list.add(new String(c, tokenStart, pos - tokenStart));
+                tokenStart = pos;
+            }
+            currentType = type;
+        }
+        list.add(new String(c, tokenStart, c.length - tokenStart));
+        return (String[]) list.toArray(new String[list.size()]);
+    }
+
     // Joining
     //-----------------------------------------------------------------------
     /**
diff --git a/src/test/org/apache/commons/lang/StringUtilsTest.java b/src/test/org/apache/commons/lang/StringUtilsTest.java
index dedfc98f5..f87e60403 100644
--- a/src/test/org/apache/commons/lang/StringUtilsTest.java
+++ b/src/test/org/apache/commons/lang/StringUtilsTest.java
@@ -847,7 +847,53 @@ private void innerTestSplitPreserveAllTokens(char separator, String sepStr, char
         assertEquals(msg, "a", res[0]);
         assertEquals(msg, str.substring(2), res[1]);
     }
-    
+
+    public void testSplitByCharacterType() {
+        assertNull(StringUtils.splitByCharacterType(null));
+        assertEquals(0, StringUtils.splitByCharacterType("").length);
+        assertNull(StringUtils.splitByCharacterType(null, true));
+        assertEquals(0, StringUtils.splitByCharacterType("", true).length);
+
+        final boolean camelCase = true;
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", " ", "de", " ",
+                "fg" }, StringUtils.splitByCharacterType("ab de fg")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", " ", "de", " ",
+                "fg" }, StringUtils.splitByCharacterType("ab de fg", camelCase)));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", "   ", "de", " ",
+                "fg" }, StringUtils.splitByCharacterType("ab   de fg")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", "   ", "de", " ",
+                "fg" }, StringUtils.splitByCharacterType("ab   de fg", camelCase)));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", ":", "cd", ":",
+                "ef" }, StringUtils.splitByCharacterType("ab:cd:ef")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ab", ":", "cd", ":",
+                "ef" }, StringUtils.splitByCharacterType("ab:cd:ef", camelCase)));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "foo", "B", "ar" },
+                StringUtils.splitByCharacterType("fooBar")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "foo", "Bar" },
+                StringUtils.splitByCharacterType("fooBar", camelCase)));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "foo", "200", "B", "ar" },
+                StringUtils.splitByCharacterType("foo200Bar")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "foo", "200", "Bar" },
+                StringUtils.splitByCharacterType("foo200Bar", camelCase)));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ASFR", "ules" },
+                StringUtils.splitByCharacterType("ASFRules")));
+
+        assertTrue(ArrayUtils.isEquals(new String[] { "ASF", "Rules" },
+                StringUtils.splitByCharacterType("ASFRules", camelCase)));
+
+    }
+
     public void testDeprecatedDeleteSpace_String() {
         assertEquals(null, StringUtils.deleteSpaces(null));
         assertEquals("", StringUtils.deleteSpaces(""));