Właściwie chcę krótszego unikalnego identyfikatora opartego na znaczniku czasu, dlatego wypróbowałem poniższy program.
Można to zgadnąć z nanosecond + ( endians.length * endians.length ) kombinacji.
public class TimStampShorterUUID {
private static final Character [] endians =
{'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',
'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',
'0', '1', '2', '3', '4', '5', '6', '7', '8', '9'
};
private static ThreadLocal<Character> threadLocal = new ThreadLocal<Character>();
private static AtomicLong iterator = new AtomicLong(-1);
public static String generateShorterTxnId() {
int firstLetter = ThreadLocalRandom.current().nextInt(0, (endians.length));
Character secondLetter = threadLocal.get();
if (secondLetter == null) {
synchronized (threadLocal) {
if (secondLetter == null) {
threadLocal.set(endians[(int) (iterator.incrementAndGet() % endians.length)]);
}
}
secondLetter = threadLocal.get();
}
return "" + endians[firstLetter] + secondLetter + System.nanoTime();
}
public static void main(String[] args) {
Map<String, String> uniqueKeysTestMap = new ConcurrentHashMap<>();
Thread t1 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t2 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t3 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t4 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t5 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t6 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
Thread t7 = new Thread() {
@Override
public void run() {
while(true) {
String time = generateShorterTxnId();
String result = uniqueKeysTestMap.put(time, "");
if(result != null) {
System.out.println("failed! - " + time);
}
}
}
};
t1.start();
t2.start();
t3.start();
t4.start();
t5.start();
t6.start();
t7.start();
}
}
UPDATE : Ten kod będzie działał na pojedynczej JVM, ale powinniśmy pomyśleć o rozproszonej JVM, stąd myślę o dwóch rozwiązaniach, jednym z DB, a drugim bez DB.
z DB
Nazwa firmy (nazwa skrócona 3 znaki) ---- Random_Number ---- Specyficzne dla klucza redis COUNTER
(3 ) -------------------------- ---------------------- (2 znaki) ---------------- (11 znaków)
bez DB
IPADDRESS ---- THREAD_NUMBER ---- INCR_NUMBER ---- epoka milisekundy
(5 znaków) ----------------- (2 znaki) --------- -------------- (2 znaki) ----------------- (6 znaków)
poinformuje Cię po zakończeniu kodowania.