This package provides two classes for managing locks in asynchronous (and
multi-threaded-like) environments: Mutex
and MutexRW
. Use these locks to
protect critical sections in your code, ensuring that only one task (or a
controlled number of tasks) can access a shared resource at a time.
Note: The
lock()
,lockRO()
, andlockRW()
methods return objects that implement a disposable interface via the[Symbol.dispose]
property. This allows you to use the upcomingusing
statement to automatically release locks at the end of a scope.
To install the package, you can use npm, yarn or pnpm:
npm install mutex-ts
# or
yarn add mutex-ts
# or
pnpm add mutex-ts
-
Mutex:
The
Mutex
class provides a simple mutual exclusion lock.You can manually acquire and release the lock with obtain():
import { Mutex } from 'mutex-ts'; const mutex = new Mutex(); // Obtain the lock const release = await mutex.obtain(); try { // Your critical section code here } finally { // Release the lock in a finally block release(); }
By using a
try...finally
block, you ensure that the lock is always released, even if an error occurs within the critical section.You can also use the
bypass
parameter to conditionally obtain a lock without waiting, which can be useful when you need to skip the lock within an already locked context determined at runtime:// Conditionally obtain the lock with bypass (non-blocking) const shouldBypass = someCondition(); // Determine at runtime const release = await mutex.obtain(shouldBypass); try { // Your critical section code here } finally { // Release the lock in a finally block release(); }
This allows you to make the decision to bypass the lock based on a runtime condition, ensuring flexibility in your locking strategy.
Alternatively, use the new
lock()
method to obtain a disposable lock object that automatically releases the lock when the scope ends. (This works in TypeScript with the upcomingusing
syntax.)import { Mutex } from 'mutex-ts'; async function criticalSectionUsing() { const mutex = new Mutex(); // Using the "using" pattern (requires proper TS configuration/polyfill) { using _ = await mutex.lock(); // Critical section code goes here. // When the block exits, _[Symbol.dispose]() is called, releasing the lock. } }
-
MutexRW:
The
MutexRW
class provides a more complex lock supporting multiple readers and a single writer.import { MutexRW } from 'mutex-ts'; const mutexRW = new MutexRW(); // Obtain a read lock const releaseRO = await mutexRW.obtainRO(); try { // Your read operation here } finally { // Release the read lock in a finally block releaseRO(); } // Obtain a write lock const releaseRW = await mutexRW.obtainRW(); try { // Your write operation here } finally { // Release the write lock in a finally block releaseRW(); }
Readers can obtain read locks simultaneously, but writers must wait until all readers release their locks. Using
try...finally
blocks ensures that locks are correctly released, preventing deadlocks or other synchronization issues.Alternatively, use the new
lockRO()
andlockRW()
methods to obtain a disposable lock object that automatically releases the lock when the scope ends. (This works in TypeScript with the upcomingusing
syntax.)import { MutexRW } from 'mutex-ts'; const mutexRW = new MutexRW(); { // Obtain a read lock using _ = await mutexRW.obtainRO(); // Automatic release of the read lock in the end of the block } { // Obtain a write lock using _ = await mutexRW.obtainRW(); // Automatic release of the write lock in the end of the block }
- Automatic Disposal: The objects returned by
lock()
,lockRO()
, andlockRW()
have a[Symbol.dispose]
property. When used with the upcoming using syntax, the corresponding lock is automatically released when the variable goes out of scope. - Bypass Option: Passing true to
obtain()
skips waiting for the lock, returning a release function (or no-op, if lock is not acquired). This can be useful when you want to conditionally enter a critical section. - Integration: Ensure your TypeScript configuration (or polyfill) supports
the
using
syntax and the disposable feature (i.e. having the correct"lib"
settings such as including"esnext"
or"esnext.disposable"
). - If you are using manual lock management, please make sure to handle lock
release in a
finally
block as shown in the examples to ensure proper synchronization and resource cleanup.
For further details, you can refer to the source code and comments within the package.
This project is licensed under the MIT License - see the LICENSE file for details.
- Dmitrii Baranov [email protected]