3 Commits

Author SHA1 Message Date
97a4ae2279 perf: RegexTools 的缓存改用 guava cache 2025-05-01 02:15:23 +08:00
af66cd2380 feat: RegexTools 新增重载方法,当将字符串视为正则表达式入参时,允许传对应的 flags 2025-05-01 02:08:23 +08:00
3b519105bf refactor!: 删除 RegexTools 中以 String[] 作为多个正则表达式入参的方法
字符串无法代表一个正则表达式,还需考虑正则表达式的 flag(s),所以当使用多个正则表达式时,更推荐使用 `Pattern[]`。
2025-04-30 22:57:47 +08:00
2 changed files with 228 additions and 181 deletions

View File

@@ -17,26 +17,40 @@
package xyz.zhouxy.plusone.commons.util;
import java.util.Arrays;
import java.util.Map;
import java.util.concurrent.ConcurrentHashMap;
import java.util.Objects;
import java.util.Optional;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import javax.annotation.Nonnull;
import javax.annotation.Nullable;
import com.google.common.cache.CacheBuilder;
import com.google.common.cache.CacheLoader;
import com.google.common.cache.LoadingCache;
/**
* 封装一些常用的正则操作,并可以缓存 {@link Pattern} 实例以复用(最多缓存大概 256 个)
* 封装一些常用的正则操作,并可以缓存 {@link Pattern} 实例以复用。
*
* @author <a href="http://zhouxy.xyz:3000/ZhouXY108">ZhouXY</a>
*
*/
public final class RegexTools {
private static final int DEFAULT_CACHE_INITIAL_CAPACITY = 64;
private static final int MAX_CACHE_SIZE = 256;
private static final Map<String, Pattern> PATTERN_CACHE
= new ConcurrentHashMap<>(DEFAULT_CACHE_INITIAL_CAPACITY);
private static final int DEFAULT_FLAG = 0;
private static final LoadingCache<RegexAndFlags, Pattern> PATTERN_CACHE = CacheBuilder
.newBuilder()
.maximumSize(MAX_CACHE_SIZE)
.build(new CacheLoader<RegexAndFlags, Pattern>() {
@SuppressWarnings("null")
public Pattern load(@Nonnull RegexAndFlags regexAndFlags) {
return regexAndFlags.compilePattern();
}
});
// ================================
// #region - getPattern
// ================================
/**
* 获取 {@link Pattern} 实例。
@@ -46,8 +60,20 @@ public final class RegexTools {
* @return {@link Pattern} 实例
*/
public static Pattern getPattern(final String pattern, final boolean cachePattern) {
return getPattern(pattern, DEFAULT_FLAG, cachePattern);
}
/**
* 获取 {@link Pattern} 实例。
*
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @return {@link Pattern} 实例
*/
public static Pattern getPattern(final String pattern, final int flags, final boolean cachePattern) {
AssertTools.checkNotNull(pattern);
return cachePattern ? cacheAndGetPatternInternal(pattern) : getPatternInternal(pattern);
return cachePattern ? cacheAndGetPatternInternal(pattern, flags) : getPatternInternal(pattern, flags);
}
/**
@@ -57,34 +83,29 @@ public final class RegexTools {
* @return {@link Pattern} 实例
*/
public static Pattern getPattern(final String pattern) {
return getPattern(pattern, DEFAULT_FLAG);
}
/**
* 获取 {@link Pattern} 实例,不缓存。
*
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @return {@link Pattern} 实例
*/
@Nonnull
public static Pattern getPattern(final String pattern, final int flags) {
AssertTools.checkNotNull(pattern);
return getPatternInternal(pattern);
return getPatternInternal(pattern, flags);
}
/**
* 将各个正则表达式转为 {@link Pattern} 实例。
*
* @param patterns 正则表达式
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @return {@link Pattern} 实例数组
*/
public static Pattern[] getPatterns(final String[] patterns, final boolean cachePattern) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
return cachePattern
? cacheAndGetPatternsInternal(patterns)
: getPatternsInternal(patterns);
}
// ================================
// #endregion - getPattern
// ================================
/**
* 将各个正则表达式转为 {@link Pattern} 实例,不缓存。
*
* @param patterns 正则表达式
* @return {@link Pattern} 实例数组
*/
public static Pattern[] getPatterns(final String[] patterns) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
return getPatternsInternal(patterns);
}
// ================================
// #region - matches
// ================================
/**
* 判断 {@code input} 是否匹配 {@code pattern}。
@@ -132,11 +153,21 @@ public final class RegexTools {
*/
public static boolean matches(@Nullable final CharSequence input, final String pattern,
final boolean cachePattern) {
AssertTools.checkNotNull(pattern);
Pattern p = cachePattern
? cacheAndGetPatternInternal(pattern)
: getPatternInternal(pattern);
return matchesInternal(input, p);
return matches(input, pattern, DEFAULT_FLAG, cachePattern);
}
/**
* 判断 {@code input} 是否匹配 {@code pattern}。
*
* @param input 输入
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @return 判断结果
*/
public static boolean matches(@Nullable final CharSequence input, final String pattern, final int flags,
final boolean cachePattern) {
return matchesInternal(input, getPattern(pattern, flags, cachePattern));
}
/**
@@ -147,69 +178,29 @@ public final class RegexTools {
* @return 判断结果
*/
public static boolean matches(@Nullable final CharSequence input, final String pattern) {
AssertTools.checkNotNull(pattern);
return matchesInternal(input, getPatternInternal(pattern));
return matches(input, pattern, DEFAULT_FLAG);
}
/**
* 判断 {@code input} 是否匹配 {@code patterns} 中的一个
* 判断 {@code input} 是否匹配 {@code pattern}。不缓存 {@link Pattern} 实例
*
* @param input 输入
* @param patterns 正则表达式
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @param input 输入
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @return 判断结果
*/
public static boolean matchesOne(@Nullable final CharSequence input, final String[] patterns,
final boolean cachePattern) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
final Pattern[] patternSet = cachePattern
? cacheAndGetPatternsInternal(patterns)
: getPatternsInternal(patterns);
return matchesOneInternal(input, patternSet);
public static boolean matches(@Nullable final CharSequence input,
final String pattern, final int flags) {
return matchesInternal(input, getPattern(pattern, flags));
}
/**
* 判断 {@code input} 是否匹配 {@code patterns} 中的一个。不缓存 {@link Pattern} 实例。
*
* @param input 输入
* @param patterns 正则表达式
* @return 判断结果
*/
public static boolean matchesOne(@Nullable final CharSequence input, final String[] patterns) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
final Pattern[] patternSet = getPatternsInternal(patterns);
return matchesOneInternal(input, patternSet);
}
// ================================
// #endregion - matches
// ================================
/**
* 判断 {@code input} 是否匹配全部正则。
*
* @param input 输入
* @param patterns 正则表达式
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @return 判断结果
*/
public static boolean matchesAll(@Nullable final CharSequence input, final String[] patterns,
final boolean cachePattern) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
final Pattern[] patternSet = cachePattern
? cacheAndGetPatternsInternal(patterns)
: getPatternsInternal(patterns);
return matchesAllInternal(input, patternSet);
}
/**
* 判断 {@code input} 是否匹配全部正则。不缓存 {@link Pattern} 实例。
*
* @param input 输入
* @param patterns 正则表达式
* @return 判断结果
*/
public static boolean matchesAll(@Nullable final CharSequence input, final String[] patterns) {
AssertTools.checkArgument(ArrayTools.isAllElementsNotNull(patterns));
final Pattern[] patternSet = getPatternsInternal(patterns);
return matchesAllInternal(input, patternSet);
}
// ================================
// #region - getMatcher
// ================================
/**
* 生成 Matcher。
@@ -233,12 +224,21 @@ public final class RegexTools {
* @return 结果
*/
public static Matcher getMatcher(final CharSequence input, final String pattern, boolean cachePattern) {
AssertTools.checkNotNull(input);
AssertTools.checkNotNull(pattern);
final Pattern p = cachePattern
? cacheAndGetPatternInternal(pattern)
: getPatternInternal(pattern);
return p.matcher(input);
return getMatcher(input, pattern, DEFAULT_FLAG, cachePattern);
}
/**
* 生成 Matcher。
*
* @param input 输入
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @param cachePattern 是否缓存 {@link Pattern} 实例
* @return 结果
*/
public static Matcher getMatcher(final CharSequence input,
final String pattern, final int flags, boolean cachePattern) {
return getMatcher(input, getPattern(pattern, flags, cachePattern));
}
/**
@@ -249,70 +249,56 @@ public final class RegexTools {
* @return 结果
*/
public static Matcher getMatcher(final CharSequence input, final String pattern) {
AssertTools.checkNotNull(input);
AssertTools.checkNotNull(pattern);
return getPatternInternal(pattern).matcher(input);
return getMatcher(input, pattern, DEFAULT_FLAG);
}
// ========== internal methods ==========
/**
* 生成 Matcher。不缓存 {@link Pattern} 实例。
*
* @param input 输入
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @return 结果
*/
public static Matcher getMatcher(final CharSequence input, final String pattern, final int flags) {
AssertTools.checkNotNull(input);
AssertTools.checkNotNull(pattern);
return getPatternInternal(pattern, flags).matcher(input);
}
// ================================
// #endregion - getMatcher
// ================================
// ================================
// #region - internal methods
// ================================
/**
* 获取 {@link Pattern} 实例。
*
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @return {@link Pattern} 实例
*/
@Nonnull
private static Pattern cacheAndGetPatternInternal(final String pattern) {
if (PATTERN_CACHE.size() < MAX_CACHE_SIZE) {
return PATTERN_CACHE.computeIfAbsent(pattern, Pattern::compile);
}
Pattern result = PATTERN_CACHE.get(pattern);
if (result != null) {
return result;
}
return Pattern.compile(pattern);
private static Pattern cacheAndGetPatternInternal(final String pattern, final int flags) {
final RegexAndFlags regexAndFlags = new RegexAndFlags(pattern, flags);
return PATTERN_CACHE.getUnchecked(regexAndFlags);
}
/**
* 获取 {@link Pattern} 实例,不缓存。
*
* @param pattern 正则表达式
* @param flags 正则表达式匹配标识
* @return {@link Pattern} 实例
*/
@Nonnull
private static Pattern getPatternInternal(final String pattern) {
Pattern result = PATTERN_CACHE.get(pattern);
if (result == null) {
result = Pattern.compile(pattern);
}
return result;
}
/**
* 将各个正则表达式转为 {@link Pattern} 实例。
*
* @param patterns 正则表达式
* @return {@link Pattern} 实例数组
*/
@Nonnull
private static Pattern[] cacheAndGetPatternsInternal(final String[] patterns) {
return Arrays.stream(patterns)
.map(RegexTools::cacheAndGetPatternInternal)
.toArray(Pattern[]::new);
}
/**
* 将各个正则表达式转为 {@link Pattern} 实例。
*
* @param patterns 正则表达式
* @return {@link Pattern} 实例数组
*/
@Nonnull
private static Pattern[] getPatternsInternal(final String[] patterns) {
return Arrays.stream(patterns)
.map(RegexTools::getPatternInternal)
.toArray(Pattern[]::new);
private static Pattern getPatternInternal(final String pattern, final int flags) {
final RegexAndFlags regexAndFlags = new RegexAndFlags(pattern, flags);
return Optional.ofNullable(PATTERN_CACHE.getIfPresent(regexAndFlags))
.orElseGet(regexAndFlags::compilePattern);
}
/**
@@ -352,8 +338,49 @@ public final class RegexTools {
.allMatch(pattern -> pattern.matcher(input).matches());
}
// ================================
// #endregion - internal methods
// ================================
private RegexTools() {
// 不允许实例化
throw new IllegalStateException("Utility class");
}
// ================================
// #region - RegexAndFlags
// ================================
private static final class RegexAndFlags {
private final String regex;
private final int flags;
private RegexAndFlags(String regex, int flags) {
this.regex = regex;
this.flags = flags;
}
private final Pattern compilePattern() {
return Pattern.compile(regex, flags);
}
@Override
public int hashCode() {
return Objects.hash(regex, flags);
}
@Override
public boolean equals(@Nullable Object obj) {
if (this == obj)
return true;
if (!(obj instanceof RegexAndFlags))
return false;
RegexAndFlags other = (RegexAndFlags) obj;
return Objects.equals(regex, other.regex) && flags == other.flags;
}
}
// ================================
// #endregion - RegexAndFlags
// ================================
}

View File

@@ -33,19 +33,42 @@ public
class RegexToolsTests {
@Test
void getPattern_CachePatternTrue_ReturnsCachedPattern() {
void getPattern_SameRegexAndFlag_CachePatternIsTrue_ReturnsCachedPattern() {
String pattern = "abc";
Pattern cachedPattern = RegexTools.getPattern(pattern, true);
Pattern patternFromCache = RegexTools.getPattern(pattern, true);
Pattern patternFromCache = RegexTools.getPattern(pattern);
assertSame(cachedPattern, patternFromCache, "Pattern should be cached");
Pattern cachedPatternWithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE, true);
Pattern patternFromCacheWithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE);
assertSame(cachedPatternWithFlag, patternFromCacheWithFlag, "Pattern should be cached");
}
@Test
void getPattern_CachePatternFalse_ReturnsNewPattern() {
String pattern = "getPattern_CachePatternFalse_ReturnsNewPattern";
void getPattern_SameRegexAndFlag_CachePatternFalse_ReturnsNewPattern() {
String pattern = "getPattern_SameRegexAndFlag_CachePatternFalse_ReturnsNewPattern";
Pattern pattern1 = RegexTools.getPattern(pattern, false);
Pattern pattern2 = RegexTools.getPattern(pattern, false);
Pattern pattern3 = RegexTools.getPattern(pattern);
assertNotSame(pattern1, pattern2, "Pattern should not be cached");
assertNotSame(pattern1, pattern3, "Pattern should not be cached");
assertNotSame(pattern2, pattern3, "Pattern should not be cached");
Pattern pattern1WithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE, false);
Pattern pattern2WithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE, false);
Pattern pattern3WithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE);
assertNotSame(pattern1WithFlag, pattern2WithFlag, "Pattern should not be cached");
assertNotSame(pattern1WithFlag, pattern3WithFlag, "Pattern should not be cached");
assertNotSame(pattern2WithFlag, pattern3WithFlag, "Pattern should not be cached");
}
@Test
void getPattern_SameRegexAndDifferentFlag_ReturnsNewPattern() {
String pattern = "getPattern_SameRegexAndDifferentFlag_CachePatternFalse_ReturnsNewPattern";
Pattern pattern1WithFlag = RegexTools.getPattern(pattern, Pattern.CASE_INSENSITIVE, true);
Pattern pattern2WithFlag = RegexTools.getPattern(pattern, 0, true);
assertNotSame(pattern1WithFlag, pattern2WithFlag, "Patterns should not be the same");
}
@Test
@@ -53,52 +76,38 @@ class RegexToolsTests {
assertThrows(NullPointerException.class, () -> {
RegexTools.getPattern(null, true);
});
}
@Test
void getPatterns_CachePatternTrue_ReturnsCachedPatterns() {
String[] patterns = {"abc", "def"};
Pattern[] cachedPatterns = RegexTools.getPatterns(patterns, true);
Pattern[] patternsFromCache = RegexTools.getPatterns(patterns, true);
assertSame(cachedPatterns[0], patternsFromCache[0]);
assertSame(cachedPatterns[1], patternsFromCache[1]);
}
@Test
void getPatterns_CachePatternFalse_ReturnsNewPatterns() {
String[] patterns = {"getPatterns_CachePatternFalse_ReturnsNewPatterns1", "getPatterns_CachePatternFalse_ReturnsNewPatterns2"};
Pattern[] patterns1 = RegexTools.getPatterns(patterns, false);
Pattern[] patterns2 = RegexTools.getPatterns(patterns, false);
assertNotSame(patterns1[0], patterns2[0]);
assertNotSame(patterns1[1], patterns2[1]);
}
@Test
void getPatterns_NullPatterns_ThrowsException() {
assertThrows(IllegalArgumentException.class, () -> {
RegexTools.getPatterns(null, true);
assertThrows(NullPointerException.class, () -> {
RegexTools.getPattern(null, Pattern.CASE_INSENSITIVE, true);
});
}
@Test
void matches_InputMatchesPattern_ReturnsTrue() {
String pattern = "abc";
assertTrue(RegexTools.matches("abc", pattern), "Input should match pattern");
assertFalse(RegexTools.matches("ABC", pattern), "Input should match pattern");
assertTrue(RegexTools.matches("ABC", pattern, Pattern.CASE_INSENSITIVE), "Input should match pattern");
Pattern compiledPattern = Pattern.compile(pattern);
assertTrue(RegexTools.matches("abc", compiledPattern), "Input should match pattern");
assertFalse(RegexTools.matches("ABC", compiledPattern), "Input should match pattern");
assertTrue(RegexTools.matches("abc", pattern, true), "Input should match pattern");
Pattern cachedPattern1 = RegexTools.getPattern(pattern);
Pattern cachedPattern2 = RegexTools.getPattern(pattern);
assertSame(cachedPattern1, cachedPattern2, "Cached pattern should be the same");
}
@Test
void matches_InputDoesNotMatchPattern_ReturnsFalse() {
String pattern = "abc";
Pattern compiledPattern = Pattern.compile(pattern);
assertFalse(RegexTools.matches("abcd", compiledPattern), "Input should not match pattern");
assertFalse(RegexTools.matches("abcd", pattern), "Input should not match pattern");
}
@Test
void matches_NullInput_ReturnsFalse() {
String pattern = "abc";
Pattern compiledPattern = Pattern.compile(pattern);
assertFalse(RegexTools.matches(null, compiledPattern), "Null input should return false");
assertFalse(RegexTools.matches(null, pattern), "Null input should return false");
}
@Test
@@ -119,6 +128,7 @@ class RegexToolsTests {
compiledPatterns[i] = Pattern.compile(patterns[i]);
}
assertFalse(RegexTools.matchesOne("xyz", compiledPatterns), "Input should not match any pattern");
assertFalse(RegexTools.matchesOne(null, compiledPatterns), "Input should not match any pattern");
}
@Test
@@ -139,30 +149,40 @@ class RegexToolsTests {
compiledPatterns[i] = Pattern.compile(patterns[i]);
}
assertFalse(RegexTools.matchesAll("abc", compiledPatterns), "Input should not match all patterns");
assertFalse(RegexTools.matchesAll(null, compiledPatterns), "Input should not match all patterns");
}
@Test
void getMatcher_ValidInputAndPattern_ReturnsMatcher() {
String pattern = "abc";
Pattern compiledPattern = Pattern.compile(pattern);
Matcher matcher = RegexTools.getMatcher("abc", compiledPattern);
assertNotNull(matcher, "Matcher should not be null");
Matcher matcher1 = RegexTools.getMatcher("abc", pattern);
assertNotNull(matcher1, "Matcher should not be null");
assertTrue(matcher1.matches(), "Should be matches");
Matcher matcher2 = RegexTools.getMatcher("ABC", pattern, true);
assertNotNull(matcher2, "Matcher should not be null");
assertFalse(matcher2.matches(), "Should be matches");
Pattern cachedPattern = RegexTools.getPattern(pattern);
Pattern patternFromCache = RegexTools.getPattern(pattern);
assertSame(cachedPattern, patternFromCache);
}
@Test
void getMatcher_NullInput_ThrowsException() {
String pattern = "abc";
Pattern compiledPattern = Pattern.compile(pattern);
assertThrows(NullPointerException.class, () -> {
RegexTools.getMatcher(null, compiledPattern);
RegexTools.getMatcher(null, pattern);
});
}
@Test
void getMatcher_NullPattern_ThrowsException() {
final Pattern pattern = null;
assertThrows(NullPointerException.class, () -> {
RegexTools.getMatcher("abc", pattern);
RegexTools.getMatcher("abc", (String) null);
});
assertThrows(NullPointerException.class, () -> {
RegexTools.getMatcher("abc", (Pattern) null);
});
}