Semafy provides synchronization and concurrency management across execution agents (main thread, web workers). It contains a robust set of tools modeled after C++ synchronization primitives, offering control and flexibility for managing shared resources and states.

• Mutexes: Exclusive and shared locking to protect data from concurrent access.
• Semaphores: Control access to a finite number of resources.
• ConditionVariable: Allows agents to wait for certain conditions to occur.
• Barriers: Ensures agents reach a certain point before any can proceed.
• Utilities: Manage async code with lock, tryLock, lockGuard, and callOnce.
• Error Handling: Specific error classes to enhance debuggability and reliability.
• Platform Agnostic: Works in any browser or server-side environment that supports SharedArrayBuffer.

NPM:

npm install semafy

Yarn:

yarn add semafy

• BasicLockable: A base interface that provides exclusive blocking for agents.

• Lockable: Extends BasicLockable to include attempted locking.

• SharedLockable: Provides shared blocking semantics for agents.

• SharedResource: Represents a shared resource that is backed by a SharedArrayBuffer.

• SharedTimedLockable: Extends SharedLockable to include timed blocking.

• TimedLockable: Extends Lockable to include timed blocking.

• Mutex: Provides essential mutex operations including lock, unlock, and tryLock.

• TimedMutex: A timed variant that supports timed operations including tryLockFor and tryLockUntil.

• RecursiveMutex: Allows multiple locks from the same agent.

• RecursiveTimedMutex: A timed variant that supports timed operations.

• SharedMutex: Allows multiple readers or exclusive writer access, facilitating reader-writer scenarios.

• SharedTimedMutex: A timed variant that supports timed operations.

• lock(): Sequentially acquires the given locks. If any lock fails, the process is stopped, and any acquired locks are released in reverse order.

• lockGuard(): Locks a mutex before calling a callback function, ensuring the mutex is unlocked afterwards.

• MultiLock: Wraps multiple BasicLockable objects to create a multi-lock. Calls to lock, unlock, etc will acquire / release locks on all of the wrapped objects.

• SharedLock: Wraps a SharedLockable object (e.g. SharedMutex) to create a shared lock. Calls to lock, unlock, etc will acquire / release a shared lock instead of an exclusive lock.

• tryLock(): Tries to sequentially acquire the given locks. If any lock fails, the process is stopped, and any acquired locks are released in reverse order.

• UniqueLock: Wraps a BasicLockable object to create a unique lock. Calls to lock, unlock, etc will acquire / release a lock on the wrapped object.

• callOnce(): Executes a callback function at most once, based on the state of a provided OnceFlag.

• OnceFlag: Represents a flag that can be set exactly once, shared across different execution agents.

• ConditionVariable: Allows agents to wait for specific conditions, tightly integrated with mutexes for state management.

• CountingSemaphore: Manages access to a finite number of resources, allowing multiple entities to hold the semaphore concurrently up to a maximum count.

• Latch: Allows agents to wait until a set of operations has been completed. Ideal for scenarios where multiple tasks must reach a common point before proceeding. For example, the completion of multiple data loading operations before data processing begins.

• LockError: A generic error related to errors in lock acquisition, release and management.

• MultiLockError: Occurs when attempting to acquire multiple locks simultaneously.

• MultiUnlockError: Occurs when attempting multiple unlocks simultaneously.

• OwnershipError: Occurs when attempting to unlock an unacquired mutex.

• RelockError: Occurs when attempting to lock an already acquired mutex. Prevents deadlocks from occurring.

• TimeoutError: Occurs when an operation exceeds a set time, such as when using tryLockFor or tryLockUntil.

Browser security requirements for using shared memory must be met. Please see SharedArrayBuffer > Security Requirements for details.

Contributions are welcome!

• Bug Reports: Please use the GitHub issue tracker to report any bugs. Include a detailed description and any relevant code snippets or logs.

• Feature Requests: Please submit feature requests as issues, clearly describing the feature and its potential benefits.

• Pull Requests: Please ensure your code adheres to the existing style of the project and include any necessary tests and documentation.

For more information, check out the contributor's guide.