我想要一个有效的实用程序来生成唯一的字节序列。UUID是一个很好的候选，但UUID. randomuuid (). tostring()会生成44e128a5-ac7a-4c9a-be4c-224b6bf81b20这样的东西，这很好，但我更喜欢无虚线字符串。
我正在寻找一种有效的方法来生成随机字符串，只从字母数字字符(没有破折号或任何其他特殊符号)。
当前回答
一个简单的解决方案是
UUID.randomUUID().toString().replace("-", "")
(像现有的解决方案一样，只是它避免了String#replaceAll调用。这里不需要正则表达式替换，因此string# replace感觉更自然，尽管从技术上讲它仍然是用正则表达式实现的。由于生成UUID的开销比替换UUID的开销更大，因此在运行时不应该有显著差异。)
对于大多数场景来说，使用UUID类可能已经足够快了，不过我希望一些不需要后期处理的专门手写变体会更快。总之，整个计算的瓶颈通常是随机数生成器。对于UUID类，它使用securerrandom。
使用哪个随机数生成器也取决于应用程序。如果对安全性敏感，一般建议使用SecureRandom。否则，ThreadLocalRandom是一种替代方案(比securerrandom或旧的Random快，但不加密安全)。
其他回答
好吧，由于UUID在toString()上添加了连字符(虚线)，我们可以从Java自己的实现中窃取实现，将字节数组缩短为32并调整偏移量。
public static String special() {
UUID uuid = UUID.randomUUID();
return fastUUID(uuid.getLeastSignificantBits(), uuid.getMostSignificantBits());
}
private static String fastUUID(long lsb, long msb) {
byte[] buf = new byte[32];
formatUnsignedLong0(lsb, 4, buf, 20, 12);
formatUnsignedLong0(lsb >>> 48, 4, buf, 16, 4);
formatUnsignedLong0(msb, 4, buf, 12, 4);
formatUnsignedLong0(msb >>> 16, 4, buf, 8, 4);
formatUnsignedLong0(msb >>> 32, 4, buf, 0, 8);
return new String(buf, 0);
}
private static final char[] digits = {
'0' , '1' , '2' , '3' , '4' , '5' ,
'6' , '7' , '8' , '9' , 'a' , 'b' ,
'c' , 'd' , 'e' , 'f' , 'g' , 'h' ,
'i' , 'j' , 'k' , 'l' , 'm' , 'n' ,
'o' , 'p' , 'q' , 'r' , 's' , 't' ,
'u' , 'v' , 'w' , 'x' , 'y' , 'z'
};
private static void formatUnsignedLong0(long val, int shift, byte[] buf, int offset, int len) {
int charPos = offset + len;
int radix = 1 << shift;
int mask = radix - 1;
do {
buf[--charPos] = (byte)digits[((int) val) & mask];
val >>>= shift;
} while (charPos > offset);
}
运行:
public static void main(String[] args) {
IntStream.range(0, 100).forEach(i-> {
System.out.println(special());
});
}
生成:
23f57da8a2784bb5acca553030f82e61
a14427efd8d147fdae315c1cf20fc53c
ee972aa1cf85414ca364bef5c74a7e57
6877ef35eab94b9485c5dd7c8c5a8a56
578721476629422381c0f625e22564a8
dbf60f068b5443d7bc6e5280696fed9f
dd611e870700480d81d394dd2125316c
04d71b9379ef4db49c28e113485ea76d
fd4e8cf3f85a45ae8c1b9bfe3e489a4a
858c4e8297f849b784b65b6096dec4d4
b30a8ca318a349b486b5693814422555
351c2fab9bc1426fa3bb512484628f12
9ce59e01db38405aab82d46f2a236880
5ffb5acb547a4f15a4621b406391bc0d
541b5fba8ddb4f1ebbd59cdcd5f59f7c
77f9460c4baa43a7bbaaf7f2aff205bd
85fa5254305b4c72b1b7c0103aaee269
062d45aa86694b06aad841236b839341
7a265293560f4223ab8248fda502c89b
b748c13ac45747b99aead4b0a2d7d179
cbcbf623c75d407fa3c88cfc89a90ed4
da263eed8771496faebb6290527f77fa
22231088dec04cffa40fb79ff56c6453
594a66de4b874b3491649c5d033917f6
4f6802ebd0cc4a39b25a67191c3af09d
8878b7ab8aa445cdadbef0f7c70d3deb
2c3ed0154f0c4ddbad498b7ae928b9bc
cac1dcaa80e54e2db3248987d2dbda4a
f9a3567e6dd54bf5900444c8b1c03815
f0d25d7b615a4495b51c01ab15093a88
243e45926311437c8b26cede2dc7de25
e4738c50e4cd448fbac252571c0907df
261d3593cc054569bcd645885d22c2ee
64a4796356a04cc4a09506aeb6f5b8fd
9aeebcbdde074ed69738589ca9bde0f1
ec040c956861466b84ed7f7cec601be0
18bd233781e44e7cb152800db4c4edc7
1b7b251df1244e8db46a45c186aada2b
3e32f644c9074cb3bbb15c5be1d9b95e
625309e3ffd14a90bfbd6d48142ac60e
664f0cf347ce4767add576da584526e7
fe3893fd376849fe9fed00e328e61470
254ce1441bbf4a7eae1cdf8d288e61e8
90896c6b309a49f48cc3b7a1570e1846
5f47acd1319245648098c1aec9b95f23
f798033052614b9eae8da7eba4ba3475
3471c4320e4e431eb1fa9f5eb5cb21e0
855f473fed034b1fa17f4f65b850e03b
1245de826d0d4373bdd4cf2157792954
543a8b16efca4fa2b5263315e8b21660
2dc186d699274257922853d783c0ec13
a92e6b1783db4b49a4aaa424b9e1b03f
16773feb48054cf0942a2a27204b3572
1e58da2107ac4ee39e28a93b32e1df1d
67622c19498d4178a1bab6b19087f2c2
412b6b4474fc43ccbeb1e7707b6420ee
7d0fd632913c425eb5f087600ccea870
439687baddb44852a43048b04d38427c
8b2dcc4e50464429a18b11e4aacf51a2
2fccb1c832894fe6b0b61bbdf175cd39
6d224b3d6e8747319fcf01b0309d8a0f
b4982e3b4b594cb4b334c95c2c96355e
c47fbaf90d1d4e9caf211f93b742631f
9440271e8ba6447d9a008e89a93016a6
8d24274b6a3f436a88362438aa6a221d
60452bd3f71747ed8c3706abb2235bd5
6fa93f2ee30740b89496439dd7227a4f
cc17504cf80641f882c8665ae166ba44
743efe8eac6e47a789928da4fb5b6f70
4c4d2df3461448c4a3e934cf4a7ea74f
b231eb3fb46240d38157764e8906aa7b
a234ae65f7ed48f6b1887644eed36cc2
c9cd5ed3df3f4a27957b45498f0c48ef
3eb2fbbb0a714bb7986aef3ee34f0254
d15968e605f0440c9e740e3f4e498a9f
63a8d50e8db24b91a13d4ac2fb6f7d5b
5377df9296154c57926672ca8b3c9478
a4db4a3a9d5148648a23aa7f4f77f1e0
d0aee355a2ba42de89d659385514b0fa
e92e7702481a4575a66d59c061459c5b
1b6c542d8f994d85a1312ab2cf4545ce
88e347a515474ec59013673e5402b97f
2187d9b2dc2b4d96baceade5ae99db44
4d641e69ca5b4acf90f8afe238d8a940
9c0f4c101c434831b928114c5fc0c401
140e16f6cf134785a98ae9baee5b9e7b
4dac5910f4d047e1b213c058e2230bf3
fb50a7e6333f49e4b469234426d5002f
c96c5f2fa167458eaa6d01997d90a980
1e79721e587c4a92aa55cdf8195c8c55
0da27fc5d8384ce299197b4e06cda1d4
a5e32d9cf5834e86b3fe02bc0e3104d6
2dc1826647594b1fb728de67d3df363c
0276371815254198bd22cc76f901b332
bf9d77b7b4a64e7a97ade2a62af1f8e0
268cce3249f64895b6b47e86cf296e5b
d523201fc950435f803bf89d5f042c45
607a4306b90b467f8b19c2c943bc92ef
adfa9fb63a874ca1ad746ff573f03f28
fe88132c70d141e8839ce9e7f0308750
实际上应该和java一样高效(更高效)。
这个实用工具类生成带有或不带破折号的String类型的uuid。
package your.package.name;
import java.security.SecureRandom;
import java.util.Random;
/**
* Utility class that creates random-based UUIDs.
*
*/
public abstract class RandomUuidStringCreator {
private static final int RANDOM_VERSION = 4;
/**
* Returns a random-based UUID as String.
*
* It uses a thread local {@link SecureRandom}.
*
* @return a random-based UUID string
*/
public static String getRandomUuid() {
return getRandomUuid(SecureRandomLazyHolder.SECURE_RANDOM);
}
/**
* Returns a random-based UUID as String WITH dashes.
*
* It uses a thread local {@link SecureRandom}.
*
* @return a random-based UUID string
*/
public static String getRandomUuidWithDashes() {
return format(getRandomUuid());
}
/**
* Returns a random-based UUID String.
*
* It uses any instance of {@link Random}.
*
* @return a random-based UUID string
*/
public static String getRandomUuid(Random random) {
long msb = 0;
long lsb = 0;
// (3) set all bit randomly
if (random instanceof SecureRandom) {
// Faster for instances of SecureRandom
final byte[] bytes = new byte[16];
random.nextBytes(bytes);
msb = toNumber(bytes, 0, 8); // first 8 bytes for MSB
lsb = toNumber(bytes, 8, 16); // last 8 bytes for LSB
} else {
msb = random.nextLong(); // first 8 bytes for MSB
lsb = random.nextLong(); // last 8 bytes for LSB
}
// Apply version and variant bits (required for RFC-4122 compliance)
msb = (msb & 0xffffffffffff0fffL) | (RANDOM_VERSION & 0x0f) << 12; // apply version bits
lsb = (lsb & 0x3fffffffffffffffL) | 0x8000000000000000L; // apply variant bits
// Convert MSB and LSB to hexadecimal
String msbHex = zerofill(Long.toHexString(msb), 16);
String lsbHex = zerofill(Long.toHexString(lsb), 16);
// Return the UUID
return msbHex + lsbHex;
}
/**
* Returns a random-based UUID as String WITH dashes.
*
* It uses a thread local {@link SecureRandom}.
*
* @return a random-based UUID string
*/
public static String getRandomUuidWithDashes(Random random) {
return format(getRandomUuid(random));
}
private static long toNumber(final byte[] bytes, final int start, final int length) {
long result = 0;
for (int i = start; i < length; i++) {
result = (result << 8) | (bytes[i] & 0xff);
}
return result;
}
private static String zerofill(String string, int length) {
return new String(lpad(string.toCharArray(), length, '0'));
}
private static char[] lpad(char[] chars, int length, char fill) {
int delta = 0;
int limit = 0;
if (length > chars.length) {
delta = length - chars.length;
limit = length;
} else {
delta = 0;
limit = chars.length;
}
char[] output = new char[chars.length + delta];
for (int i = 0; i < limit; i++) {
if (i < delta) {
output[i] = fill;
} else {
output[i] = chars[i - delta];
}
}
return output;
}
private static String format(String string) {
char[] input = string.toCharArray();
char[] output = new char[36];
System.arraycopy(input, 0, output, 0, 8);
System.arraycopy(input, 8, output, 9, 4);
System.arraycopy(input, 12, output, 14, 4);
System.arraycopy(input, 16, output, 19, 4);
System.arraycopy(input, 20, output, 24, 12);
output[8] = '-';
output[13] = '-';
output[18] = '-';
output[23] = '-';
return new String(output);
}
// Holds lazy secure random
private static class SecureRandomLazyHolder {
static final Random SECURE_RANDOM = new SecureRandom();
}
/**
* For tests!
*/
public static void main(String[] args) {
System.out.println("// Using `java.security.SecureRandom` (DEFAULT)");
System.out.println("RandomUuidCreator.getRandomUuid()");
System.out.println();
for (int i = 0; i < 5; i++) {
System.out.println(RandomUuidStringCreator.getRandomUuid());
}
System.out.println();
System.out.println("// Using `java.util.Random` (FASTER)");
System.out.println("RandomUuidCreator.getRandomUuid(new Random())");
System.out.println();
Random random = new Random();
for (int i = 0; i < 5; i++) {
System.out.println(RandomUuidStringCreator.getRandomUuid(random));
}
}
}
输出如下:
// Using `java.security.SecureRandom` (DEFAULT)
RandomUuidStringCreator.getRandomUuid()
'f553ca75657b4b5d85bedf1082785a0b'
'525ecc389e934f209b97d0f0db09d9c6'
'93ec6425bb04499ab47b790fd013ab0d'
'c2d438c620ea4cd5baafd448f9fe945b'
'fb4bc5734931415e94e78da62cb5fe0d'
// Using `java.util.Random` (FASTER)
RandomUuidStringCreator.getRandomUuid(new Random())
'051360b5c92d40fbbb89b40842adbacc'
'a993896538aa43faacbcfd83f913f38b'
'720684d22c584d5299cb03cdbc1912d2'
'82cf94ea296a4a138a92825a0068d4a1'
'a7eda46a215c4e55be3aa957ba74ca9c'
uuid-creator中有一个编解码器可以更有效地做到这一点:Base16Codec。例子:
// Returns a base-16 string
// It is much faster than doing `uuid.toString().replaceAll("-", "")`.
UuidCodec<String> codec = new Base16Codec();
String string = codec.encode(UUID.randomUUID());
我刚刚复制了UUID toString()方法，并更新了它以删除“-”。它会比其他解决方案更快更直接
public String generateUUIDString(UUID uuid) {
return (digits(uuid.getMostSignificantBits() >> 32, 8) +
digits(uuid.getMostSignificantBits() >> 16, 4) +
digits(uuid.getMostSignificantBits(), 4) +
digits(uuid.getLeastSignificantBits() >> 48, 4) +
digits(uuid.getLeastSignificantBits(), 12));
}
/** Returns val represented by the specified number of hex digits. */
private String digits(long val, int digits) {
long hi = 1L << (digits * 4);
return Long.toHexString(hi | (val & (hi - 1))).substring(1);
}
用法:
generateUUIDString（UUID.randomUUID（））
使用反射的另一个实现
public String generateString(UUID uuid) throws NoSuchMethodException, InvocationTargetException, IllegalAccessException {
if (uuid == null) {
return "";
}
Method digits = UUID.class.getDeclaredMethod("digits", long.class, int.class);
digits.setAccessible(true);
return ( (String) digits.invoke(uuid, uuid.getMostSignificantBits() >> 32, 8) +
digits.invoke(uuid, uuid.getMostSignificantBits() >> 16, 4) +
digits.invoke(uuid, uuid.getMostSignificantBits(), 4) +
digits.invoke(uuid, uuid.getLeastSignificantBits() >> 48, 4) +
digits.invoke(uuid, uuid.getLeastSignificantBits(), 12));
}
一个简单的解决方案是
UUID.randomUUID().toString().replace("-", "")
(像现有的解决方案一样，只是它避免了String#replaceAll调用。这里不需要正则表达式替换，因此string# replace感觉更自然，尽管从技术上讲它仍然是用正则表达式实现的。由于生成UUID的开销比替换UUID的开销更大，因此在运行时不应该有显著差异。)
对于大多数场景来说，使用UUID类可能已经足够快了，不过我希望一些不需要后期处理的专门手写变体会更快。总之，整个计算的瓶颈通常是随机数生成器。对于UUID类，它使用securerrandom。
使用哪个随机数生成器也取决于应用程序。如果对安全性敏感，一般建议使用SecureRandom。否则，ThreadLocalRandom是一种替代方案(比securerrandom或旧的Random快，但不加密安全)。
我很惊讶看到这么多字符串取代UUID的想法。这个怎么样:
UUID temp = UUID.randomUUID();
String uuidString = Long.toHexString(temp.getMostSignificantBits())
+ Long.toHexString(temp.getLeastSignificantBits());
这是最快的方法，因为整个UUID的toString()已经更加昂贵，更不用说正则表达式，它必须被解析和执行或替换为空字符串。
推荐文章
- 为特定的代码行禁用特定的Checkstyle规则?
- Java中生成UUID字符串的有效方法(UUID. randomuuid ().toString()不带破折号)
- malformedurlexception:没有协议
- 为什么字符集名称不是常量?
- Maven依赖项失败，出现501错误
- lambda表达式每次执行时都在堆上创建一个对象吗?
- 使用流相加bigdecimal
- Java抽象接口
- 无法识别的SSL消息，明文连接?异常
- 编码字符串为UTF-8
- 用于双值的assertEquals的delta或epsilon参数的含义
- 在Linux中安装Java SDK后，我可以在哪里找到它?
- 如何从字典中获得一个随机值?
- Java等价于c#扩展方法
- 当使用JDBC连接到postgres时，是否可以指定模式?
