问题
这两天开发时，发现一个java字符串分割的坑，如下代码：
@Slf4j
public class StringTest {
private static final String SPLIT_SYMBOL = "|";
/**
* 错误示例
*/
@Test
public void testSplit() {
String str = "abc" + SPLIT_SYMBOL + "test123" + SPLIT_SYMBOL + "你好world";
String[] strArr = str.split(SPLIT_SYMBOL);
log.info("strArr.len: {}", strArr.length);
for (String s : strArr) {
log.info("strArr element: {}", s);
}
}
}
输出结果为：
09:21:55.453 [main] INFO com.practice.learn.devtraps.StringTest - strArr.len: 19
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: a
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: b
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: c
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: |
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: t
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: e
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: s
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: t
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 1
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 2
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 3
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: |
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 你
09:21:55.454 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 好
09:21:55.455 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: w
09:21:55.455 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: o
09:21:55.455 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: r
09:21:55.455 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: l
09:21:55.455 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: d
分析
经查才知道，因为String.split(String regex)处理时会把regex作为正在表达式判断，而我们知道"|" 是OR（或）操作，所以我们上面的程序解析之后肯定是错误的 （具体细节分析，我现在也没有完全看懂，后续学习正则表达式以及阅读并调试Java源代码后，再补充）。String类部分源代码如下（jdk版本：jdk-11.0.7）：
public final class String implements java.io.Serializable, Comparable<String>, CharSequence {
...
public String[] split(String regex) {
return split(regex, 0);
}
...
...
public String[] split(String regex, int limit) {
/* fastpath if the regex is a
(1)one-char String and this character is not one of the
RegEx's meta characters ".$|()[{^?*+\\", or
(2)two-char String and the first char is the backslash and
the second is not the ascii digit or ascii letter.
*/
char ch = 0;
if (((regex.length() == 1 &&
".$|()[{^?*+\\".indexOf(ch = regex.charAt(0)) == -1) ||
(regex.length() == 2 &&
regex.charAt(0) == '\\' &&
(((ch = regex.charAt(1))-'0')|('9'-ch)) < 0 &&
((ch-'a')|('z'-ch)) < 0 &&
((ch-'A')|('Z'-ch)) < 0)) &&
(ch < Character.MIN_HIGH_SURROGATE ||
ch > Character.MAX_LOW_SURROGATE))
{
...
...
String[] result = new String[resultSize];
return list.subList(0, resultSize).toArray(result);
}
return Pattern.compile(regex).split(this, limit);
}
...
}
解决方案
而解决方案也很简单，就是对特殊字符进行转义，即 “|” --> “\\|”，代码调整为：
@Slf4j
public class StringTest {
private static final String SPLIT_SYMBOL = "|";
private static final String SPLIT_SYMBOL_WITH_ESCAPE_CHARACTER = "\\" + SPLIT_SYMBOL;
/**
* 使用 '|' 等字段进行的分割，需要使用转义符
*/
@Test
public void testSplit() {
String str = "abc" + SPLIT_SYMBOL + "test123" + SPLIT_SYMBOL + "你好world";
String[] strArr = str.split(SPLIT_SYMBOL_WITH_ESCAPE_CHARACTER);
log.info("strArr.len: {}", strArr.length);
for (String s : strArr) {
log.info("strArr element: {}", s);
}
}
}
输出结果为：
09:43:21.060 [main] INFO com.practice.learn.devtraps.StringTest - strArr.len: 3
09:43:21.060 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: abc
09:43:21.060 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: test123
09:43:21.060 [main] INFO com.practice.learn.devtraps.StringTest - strArr element: 你好world
再深入一点
而手动增加’\\'并不是一个好idea，因为我记不住哪些字符需要转义，哪些不需要。所以最好的办法是由jdk自己封装并判断。而jdk开发者们，早就为我们想到了，即：
public final class Pattern implements java.io.Serializable
{
...
...
public static String quote(String s) {
int slashEIndex = s.indexOf("\\E");
if (slashEIndex == -1)
return "\\Q" + s + "\\E";
int lenHint = s.length();
lenHint = (lenHint < Integer.MAX_VALUE - 8 - lenHint) ?
(lenHint << 1) : (Integer.MAX_VALUE - 8);
StringBuilder sb = new StringBuilder(lenHint);
sb.append("\\Q");
int current = 0;
do {
sb.append(s, current, slashEIndex)
.append("\\E\\\\E\\Q");
current = slashEIndex + 2;
} while ((slashEIndex = s.indexOf("\\E", current)) != -1);
return sb.append(s, current, s.length())
.append("\\E")
.toString();
}
...
...
}
我们只需要调用Pattern.quote方法，将regex进行封装即可。针对上面的测试代码，调整起来很简单：
private static final String SPLIT_SYMBOL_WITH_ESCAPE_CHARACTER = Pattern.quote(SPLIT_SYMBOL);
运行结果没有任何问题。
结论
- 对java的字符串分割，心中要有一根弦：正则表达式。
- 使用Pattern.quote(String s)对分割符进行封装。