Locks implements locking/synchronization mechanisms that have traditionally been used for protecting shared memory between multiple threads. JavaScript is inherently single threaded and does not suffer from these security and stability issues. However, because of its asynchronous eventy nature JavaScript can still benefit from making particular operations wait for the completion of others.
Node.js users:
npm install locks
Component users:
component install Wizcorp/locks
Accessing the locks module:
var locks = require('locks');
Please note that all the examples below will also demonstrate how to unlock each time. But in general, this matters:
// unlocking will give waiting processes a chance to get the lock and continue
myLock.unlock();
Mutex locks are the most basic locks which aim to prevent the simultaneous access to a resource by more than one actor at a time. more info
Creating a Mutex Lock:
var mutex = locks.createMutex();
Waiting to lock:
mutex.lock(function () {
console.log('We got the lock!');
// do stuff
mutex.unlock();
});
Waiting to lock, with timeout:
mutex.timedLock(5000, function (error) {
if (error) {
console.log('Could not get the lock within 5 seconds, so gave up');
} else {
console.log('We got the lock!');
// do stuff
mutex.unlock();
}
});
Checking if a lock is held:
if (mutex.isLocked) {
console.log('Something has the lock.');
}
Optimistic attempt to lock:
if (mutex.tryLock()) {
console.log('We got the lock!');
// do stuff
mutex.unlock();
} else {
console.log('Could not get the lock at this time');
}
Read/Write Locks are used to allow many actors to read from a resource, as long as nothing is writing to it. That also means that only one actor may be writing at any given time. more info
Creating a Read/Write Lock:
var rwlock = locks.createReadWriteLock();
Waiting to read lock:
rwlock.readLock(function () {
console.log('We may now read from a resource!');
// do stuff
rwlock.unlock();
});
Waiting to write lock:
rwlock.writeLock(function () {
console.log('We may now write to a resource!');
// do stuff
rwlock.unlock();
});
Waiting to read lock, with timeout:
rwlock.timedReadLock(5000, function (error) {
if (error) {
console.log('Could not get the lock within 5 seconds, so gave up');
} else {
console.log('We may now read from a resource!');
// do stuff
rwlock.unlock();
}
});
Waiting to write lock, with timeout:
rwlock.timedWriteLock(5000, function (error) {
if (error) {
console.log('Could not get the lock within 5 seconds, so gave up');
} else {
console.log('We may now write to a resource!');
// do stuff
rwlock.unlock();
}
});
Checking if a lock is held:
if (rwlock.isWriteLocked) {
console.log('Something has the write lock.');
}
else if (rwlock.isReadLocked) {
console.log('The read lock is held one or more times.');
}
Optimistic attempt to read lock:
if (rwlock.tryReadLock()) {
console.log('We may now read from a resource!');
// do stuff
rwlock.unlock();
} else {
console.log('Could not get the lock at this time');
}
Optimistic attempt to write lock:
if (rwlock.tryWriteLock()) {
console.log('We may now write to a resource!');
// do stuff
rwlock.unlock();
} else {
console.log('Could not get the lock at this time');
}
Condition variables allow synchronization between processes based on values.
Creating a Condition Variable:
var initialValue = 'hello world';
var cond = locks.createCondVariable(initialValue);
Waiting for a condition to be met:
cond.wait(
function conditionTest(value) {
return value.indexOf('こんにちは') !== -1;
},
function whenConditionMet() {
console.log('Our welcome message is in Japanese!');
}
);
Setting the value on a Condition Variable:
cond.set('こんにちは世界!');
Semaphores solve the problem of sharing a limited set of resources. more info
Creating a Semaphore:
var initialValue = 3; // amount of resources available
var sem = locks.createSemaphore(initialValue);
Claiming and releasing a resource:
sem.wait(function () {
console.log('We may now access one resource!');
// do stuff
// release the resource
sem.signal();
});
MIT