+ * Most users of this class do not need to call this method. This method + * will be called automatically by other (public) methods when required. + *
+ * This method exists to allow subclasses to add code before or after the + * tokenization. For example, a subclass could alter the character array, + * offset or count to be parsed, or call the tokenizer multiple times on + * multiple strings. It is also be possible to filter the results. + *
+ * StrTokenizer will always pass a zero offset and a count
+ * equal to the length of the array to this method, however a subclass
+ * may pass other values, or even an entirely different array.
*
- * @return array containing the tokens.
+ * @param chars the character array being tokenized, may be null
+ * @param offset the start position within the character array, must be valid
+ * @param count the number of characters to tokenize, must be valid
+ * @return the modifiable list of String tokens, unmodifiable if null array or zero count
*/
- private String[] readTokens() {
- if (chars == null) {
- return ArrayUtils.EMPTY_STRING_ARRAY;
- }
- int len = chars.length;
- if (len == 0) {
- return ArrayUtils.EMPTY_STRING_ARRAY;
+ protected List tokenize(char[] chars, int offset, int count) {
+ if (chars == null || count == 0) {
+ return Collections.EMPTY_LIST;
}
StrBuilder buf = new StrBuilder();
List tokens = new ArrayList();
- int start = 0;
+ int pos = offset;
// loop around the entire buffer
- while (start >= 0 && start < len) {
+ while (pos >= 0 && pos < count) {
// find next token
- start = readNextToken(chars, start, len, buf, tokens);
+ pos = readNextToken(chars, pos, count, buf, tokens);
// handle case where end of string is a delimiter
- if (start >= len) {
+ if (pos >= count) {
addToken(tokens, "");
}
}
- return (String[]) tokens.toArray(new String[tokens.size()]);
+ return tokens;
}
/**
@@ -1058,10 +1068,7 @@ public StrTokenizer setIgnoreEmptyTokens(boolean ignoreEmptyTokens) {
* @return the string content being parsed
*/
public String getContent() {
- if (text == null) {
- text = new String(chars);
- }
- return text;
+ return new String(chars);
}
//-----------------------------------------------------------------------
diff --git a/src/test/org/apache/commons/lang/text/StrTokenizerTest.java b/src/test/org/apache/commons/lang/text/StrTokenizerTest.java
index 47c3aa22c..3ab8d3d82 100644
--- a/src/test/org/apache/commons/lang/text/StrTokenizerTest.java
+++ b/src/test/org/apache/commons/lang/text/StrTokenizerTest.java
@@ -17,6 +17,7 @@
package org.apache.commons.lang.text;
import java.util.Arrays;
+import java.util.Collections;
import java.util.List;
import java.util.NoSuchElementException;
@@ -531,7 +532,7 @@ void testEmpty(StrTokenizer tokenizer) {
public void testGetContent() {
String input = "a b c \"d e\" f ";
StrTokenizer tok = new StrTokenizer(input);
- assertSame(input, tok.getContent());
+ assertEquals(input, tok.getContent());
tok = new StrTokenizer(input.toCharArray());
assertEquals(input, tok.getContent());
@@ -804,4 +805,29 @@ public void testIteration() {
assertEquals(false, tkn.hasNext());
}
+ //-----------------------------------------------------------------------
+ public void testTokenizeSubclassInputChange() {
+ StrTokenizer tkn = new StrTokenizer("a b c d e") {
+ protected List tokenize(char[] chars, int offset, int count) {
+ return super.tokenize("w x y z".toCharArray(), 2, 5);
+ }
+ };
+ assertEquals("x", tkn.next());
+ assertEquals("y", tkn.next());
+ }
+
+ //-----------------------------------------------------------------------
+ public void testTokenizeSubclassOutputChange() {
+ StrTokenizer tkn = new StrTokenizer("a b c") {
+ protected List tokenize(char[] chars, int offset, int count) {
+ List list = super.tokenize(chars, offset, count);
+ Collections.reverse(list);
+ return list;
+ }
+ };
+ assertEquals("c", tkn.next());
+ assertEquals("b", tkn.next());
+ assertEquals("a", tkn.next());
+ }
+
}