From 6c009fd7820c74bac251565bfee7999bb831ba38 Mon Sep 17 00:00:00 2001 From: Stephen Colebourne Date: Sat, 19 Jul 2003 18:09:33 +0000 Subject: [PATCH] Fix split to use whitespace, remove StringTokenizer Performance tune some methods git-svn-id: https://svn.apache.org/repos/asf/jakarta/commons/proper/lang/trunk@137452 13f79535-47bb-0310-9956-ffa450edef68 --- .../org/apache/commons/lang/StringUtils.java | 270 +++++++++++------- 1 file changed, 168 insertions(+), 102 deletions(-) diff --git a/src/java/org/apache/commons/lang/StringUtils.java b/src/java/org/apache/commons/lang/StringUtils.java index e5eb0288a..bf2881d59 100644 --- a/src/java/org/apache/commons/lang/StringUtils.java +++ b/src/java/org/apache/commons/lang/StringUtils.java @@ -56,7 +56,6 @@ import java.util.ArrayList; import java.util.Iterator; import java.util.List; -import java.util.StringTokenizer; import org.apache.commons.lang.math.NumberUtils; @@ -99,10 +98,25 @@ * @author Arun Mammen Thomas * @author Gary Gregory * @since 1.0 - * @version $Id: StringUtils.java,v 1.64 2003/07/19 00:22:50 scolebourne Exp $ + * @version $Id: StringUtils.java,v 1.65 2003/07/19 18:09:33 scolebourne Exp $ */ public class StringUtils { - + // Performance testing notes (JDK 1.4, Jul03, scolebourne) + // Whitespace: + // Character.isWhitespace() is faster than WHITESPACE.indexOf() + // where WHITESPACE is a string of all whitespace characters + // + // Character access: + // String.charAt(n) versus toCharArray(), then array[n] + // String.charAt(n) is about 15% worse for a 10K string + // They are about equal for a length 50 string + // String.charAt(n) is about 4 times better for a length 3 string + // String.charAt(n) is best bet overall + // + // Append: + // String.concat about twice as fast as StringBuffer.append + // (not sure who tested this) + /** *

The maximum size to which the padding constant(s) can expand.

*/ @@ -121,7 +135,6 @@ public class StringUtils { *

Used for efficient space padding. The length of each String expands as needed.

*/ private final static String[] padding = new String[Character.MAX_VALUE]; - // String.concat about twice as fast as StringBuffer.append /** *

StringUtils instances should NOT be constructed in @@ -181,7 +194,7 @@ public static String clean(String str) { * * @see java.lang.String#trim() * @param str the String to be trimmed, may be null - * @return the trimmed text, null if null String input + * @return the trimmed string, null if null String input */ public static String trim(String str) { return (str == null ? null : str.trim()); @@ -206,7 +219,7 @@ public static String trim(String str) { * @see java.lang.String#trim() * @param str the String to be trimmed, may be null * @return the trimmed String, - * null if a whitespace, empty or null String input + * null if only chars <= 32, empty or null String input */ public static String trimToNull(String str) { String ts = trim(str); @@ -231,7 +244,7 @@ public static String trimToNull(String str) { * * @see java.lang.String#trim() * @param str the String to be trimmed, may be null - * @return the trimmed String, or an empty String if null input + * @return the trimmed String, or an empty String if null input */ public static String trimToEmpty(String str) { return (str == null ? "" : str.trim()); @@ -287,8 +300,8 @@ public static String deleteWhitespace(String str) { if (str == null) { return null; } - StringBuffer buffer = new StringBuffer(); int sz = str.length(); + StringBuffer buffer = new StringBuffer(sz); for (int i = 0; i < sz; i++) { if (!Character.isWhitespace(str.charAt(i))) { buffer.append(str.charAt(i)); @@ -992,13 +1005,16 @@ public static String mid(String str, int pos, int len) { * separator. * Whitespace is defined by {@link Character#isWhitespace(char)}.

* - *

The separator is not included in the returned String array.

+ *

The separator is not included in the returned String array. + * Adjacent separators are treated as one separator.

* *

A null input String returns null.

* * 
-     * StringUtils.split(null)      = null
-     * StringUtils.split("abc def") = ["abc", "def"]
+     * StringUtils.split(null)       = null
+     * StringUtils.split("")         = []
+     * StringUtils.split("abc def")  = ["abc", "def"]
+     * StringUtils.split("abc  def") = ["abc", "def"]
      *
* * @param str the String to parse, may be null @@ -1009,17 +1025,19 @@ public static String[] split(String str) { } /** - *

Splits the provided text into an array, using the specified separator.

+ *

Splits the provided text into an array, separator specified. + * This is an alternative to using StringTokenizer.

* *

The separator is not included in the returned String array. - * Adjacent separators will cause an empty String to be returned ("").

+ * Adjacent separators are treated as one separator.

* *

A null input String returns null.

* * 
-     * StringUtils.split(null, '.')      = null
-     * StringUtils.split("a.b.c", '.') = ["a", "b", "c"]
-     * StringUtils.split("a..b.c", '.') = ["a", "", "b", "c"]
+     * StringUtils.split(null, '.')     = null
+     * StringUtils.split("", '.')       = []
+     * StringUtils.split("a.b.c", '.')  = ["a", "b", "c"]
+     * StringUtils.split("a..b.c", '.') = ["a", "b", "c"]
      * StringUtils.split("a:b:c", '.')  = ["a:b:c"]
      *
* @@ -1029,34 +1047,52 @@ public static String[] split(String str) { * @return an array of parsed Strings, null if null String input */ public static String[] split(String str, char separatorChar) { + // Performance tuned for 2.0 (JDK1.4) + if (str == null) { return null; } - char[] chars = str.toCharArray(); - List list = new ArrayList(); - int start = 0; - for (int i = 0; i < chars.length; i++) { - if (chars[i] == separatorChar) { - list.add(str.substring(start, i)); - start = i + 1; - } + int len = str.length(); + if (len == 0) { + return ArrayUtils.EMPTY_STRING_ARRAY; + } + List list = new ArrayList(); + int i =0, start = 0; + boolean match = false; + while (i < len) { + if (str.charAt(i) == separatorChar) { + if (match) { + list.add(str.substring(start, i)); + match = false; + } + start = ++i; + continue; + } + match = true; + i++; + } + if (match) { + list.add(str.substring(start, i)); } - list.add(str.substring(start)); return (String[]) list.toArray(new String[list.size()]); } /** - *

Splits the provided text into an array, using the specified separators.

+ *

Splits the provided text into an array, separators specified. + * This is an alternative to using StringTokenizer.

* - *

The separator is not included in the returned String array.

+ *

The separator is not included in the returned String array. + * Adjacent separators are treated as one separator.

* *

A null input String returns null. * A null separatorChars splits on whitespace.

* * 
      * StringUtils.split(null, null)      = null
+     * StringUtils.split("", null)        = []
      * StringUtils.split("abc def", null) = ["abc", "def"]
      * StringUtils.split("abc def", " ")  = ["abc", "def"]
+     * StringUtils.split("abc  def", " ") = ["abc", "def"]
      * StringUtils.split("ab:cd:ef", ":") = ["ab", "cd", "ef"]
      *
* @@ -1070,74 +1106,104 @@ public static String[] split(String str, String separatorChars) { } /** - *

Splits the provided text into a array, based on a given separator.

+ *

Splits the provided text into an array, separators specified. + * This is an alternative to using StringTokenizer.

* - *

The separator is not included in the returned String array. The - * maximum number of splits to perfom can be controlled. A null - * separator will cause parsing to be on whitespace.

- * - *

This is useful for quickly splitting a String directly into - * an array of tokens, instead of an enumeration of tokens (as - * StringTokenizer does).

+ *

The separator is not included in the returned String array. + * Adjacent separators are treated as one separator.

* *

A null input String returns null. * A null separatorChars splits on whitespace.

* * 
-     * StringUtils.split(null, null, 0)       = null
-     * StringUtils.split("ab de fg", null, 0) = ["ab", "cd", "ef"]
-     * StringUtils.split("ab:cd:ef", ":", 0)  = ["ab", "cd", "ef"]
-     * StringUtils.split("ab:cd:ef", ":", 2)  = ["ab", "cdef"]
+     * StringUtils.split(null, null, 0)         = null
+     * StringUtils.split("", null, 0)           = []
+     * StringUtils.split("ab de fg", null, 0)   = ["ab", "cd", "ef"]
+     * StringUtils.split("ab   de fg", null, 0) = ["ab", "cd", "ef"]
+     * StringUtils.split("ab:cd:ef", ":", 0)    = ["ab", "cd", "ef"]
+     * StringUtils.split("ab:cd:ef", ":", 2)    = ["ab", "cdef"]
      *
* * @param str the String to parse, may be null * @param separatorChars the characters used as the delimiters, * null splits on whitespace * @param max the maximum number of elements to include in the - * array. A zero or negative value implies no limit. + * array. A zero or negative value implies no limit * @return an array of parsed Strings, null if null String input */ public static String[] split(String str, String separatorChars, int max) { + // Performance tuned for 2.0 (JDK1.4) + // Direct code is quicker than StringTokenizer. + // Also, StringTokenizer uses isSpace() not isWhitespace() + if (str == null) { return null; } - StringTokenizer tok = null; + int len = str.length(); + if (len == 0) { + return ArrayUtils.EMPTY_STRING_ARRAY; + } + List list = new ArrayList(); + int sizePlus1 = 1; + int i =0, start = 0; + boolean match = false; if (separatorChars == null) { - // Null separator means we're using StringTokenizer's default - // delimiter, which comprises all whitespace characters. - - // TODO: StringTokenizer uses isSpace() not isWhitespace() - tok = new StringTokenizer(str); - } else { - tok = new StringTokenizer(str, separatorChars); - } - - int listSize = tok.countTokens(); - if (max > 0 && listSize > max) { - listSize = max; - } - - String[] list = new String[listSize]; - int i = 0; - int lastTokenBegin = 0; - int lastTokenEnd = 0; - while (tok.hasMoreTokens()) { - if (max > 0 && i == listSize - 1) { - // In the situation where we hit the max yet have - // tokens left over in our input, the last list - // element gets all remaining text. - String endToken = tok.nextToken(); - lastTokenBegin = str.indexOf(endToken, lastTokenEnd); - list[i] = str.substring(lastTokenBegin); - break; - } else { - list[i] = tok.nextToken(); - lastTokenBegin = str.indexOf(list[i], lastTokenEnd); - lastTokenEnd = lastTokenBegin + list[i].length(); + // Null separator means use whitespace + while (i < len) { + if (Character.isWhitespace(str.charAt(i))) { + if (match) { + if (sizePlus1++ == max) { + i = len; + } + list.add(str.substring(start, i)); + match = false; + } + start = ++i; + continue; + } + match = true; + i++; + } + } else if (separatorChars.length() == 1) { + // Optimise 1 character case + char sep = separatorChars.charAt(0); + while (i < len) { + if (str.charAt(i) == sep) { + if (match) { + if (sizePlus1++ == max) { + i = len; + } + list.add(str.substring(start, i)); + match = false; + } + start = ++i; + continue; + } + match = true; + i++; + } + } else { + // standard case + while (i < len) { + if (separatorChars.indexOf(str.charAt(i)) >= 0) { + if (match) { + if (sizePlus1++ == max) { + i = len; + } + list.add(str.substring(start, i)); + match = false; + } + start = ++i; + continue; + } + match = true; + i++; } - i++; } - return list; + if (match) { + list.add(str.substring(start, i)); + } + return (String[]) list.toArray(new String[list.size()]); } // Joining @@ -1792,41 +1858,47 @@ public static String escape(String str) { * null if null String input */ public static String repeat(String str, int repeat) { + // Performance tuned for 2.0 (JDK1.4) + if (str == null) { return null; } if (repeat <= 0) { return ""; } - int inputLength = str.length(); + int inputLength; + if (repeat == 1 || (inputLength = str.length()) == 0) { + return str; + } if (inputLength == 1 && repeat <= PAD_LIMIT) { return padding(repeat, str.charAt(0)); } - char[] input = str.toCharArray(); - char[] output = new char[repeat * inputLength]; + int outputLength = inputLength * repeat; switch (inputLength) { case 1: - char ch = input[0]; + char ch = str.charAt(0); + char[] output1 = new char[outputLength]; for (int i = repeat - 1; i >= 0; i--) { - output[i] = ch; + output1[i] = ch; } - break; + return new String(output1); case 2: - char ch0 = input[0]; - char ch1 = input[1]; + char ch0 = str.charAt(0); + char ch1 = str.charAt(1); + char[] output2 = new char[outputLength]; for (int i = repeat * 2 - 2; i >= 0; i--,i--) { - output[i] = ch0; - output[i + 1] = ch1; + output2[i] = ch0; + output2[i + 1] = ch1; } - break; + return new String(output2); default: - for (int i = repeat - 1; i >= 0; i--) { - System.arraycopy(input, 0, output, i * inputLength, inputLength); - } - break; + StringBuffer buf = new StringBuffer(outputLength); + for (int i = 0; i < repeat; i++) { + buf.append(str); + } + return buf.toString(); } - return new String(output); } /** @@ -2210,8 +2282,7 @@ public static String center(String str, int size, String padStr) { * This is similar to {@link String#trim()} but instead removes whitespace. * Whitespace is defined by {@link Character#isWhitespace(char)}.

* - *

If the input String is null, null - * is returned.

+ *

A null input String returns null.

* * 
      * StringUtils.strip(null)     = null
@@ -2234,8 +2305,7 @@ public static String strip(String str) {
      * This is similar to {@link String#trim()} but allows the characters
      * to be stripped to be controlled.

      *
-     * 
If the input String is null, null
-     * is returned.

+     * 
A null input String returns null.

      * 
      * 
If the stripChars String is null, whitespace is
      * stripped as defined by {@link Character#isWhitespace(char)}.
@@ -2262,8 +2332,7 @@ public static String strip(String str, String stripChars) {
     /**
      * 
Strips any of a set of characters from the start of a String.

      *
-     * 
If the input String is null, null
-     * is returned.

+     * 
A null input String returns null.

      * 
      * 
If the stripChars String is null, whitespace is
      * stripped as defined by {@link Character#isWhitespace(char)}.

@@ -2302,8 +2371,7 @@ public static String stripStart(String str, String stripChars) {
     /**
      * 
Strips any of a set of characters from the end of a String.

      *
-     * 
If the input String is null, null
-     * is returned.

+     * 
A null input String returns null.

      * 
      * 
If the stripChars String is null, whitespace is
      * stripped as defined by {@link Character#isWhitespace(char)}.

@@ -2570,9 +2638,8 @@ public static String capitaliseAllWords(String str) {
         }
         StringBuffer buffer = new StringBuffer(strLen);
         boolean whitespace = true;
-        char[] strChars = str.toCharArray();
         for (int i = 0; i < strLen; i++) {
-            char ch = strChars[i];
+            char ch = str.charAt(i);
             if (Character.isWhitespace(ch)) {
                 buffer.append(ch);
                 whitespace = true;
@@ -2609,9 +2676,8 @@ public static String uncapitaliseAllWords(String str) {
         }
         StringBuffer buffer = new StringBuffer(strLen);
         boolean whitespace = true;
-        char[] strChars = str.toCharArray();
         for (int i = 0; i < strLen; i++) {
-            char ch = strChars[i];
+            char ch = str.charAt(i);
             if (Character.isWhitespace(ch)) {
                 buffer.append(ch);
                 whitespace = true;