File-based locks for Python on Linux and Windows
Project description
Locket implements a file-based lock that can be used by multiple processes provided they use the same path.
import locket
# Wait for lock
with locket.lock_file("path/to/lock/file"):
perform_action()
# Raise LockError if lock cannot be acquired immediately
with locket.lock_file("path/to/lock/file", timeout=0):
perform_action()
# Raise LockError if lock cannot be acquired after thirty seconds
with locket.lock_file("path/to/lock/file", timeout=30):
perform_action()
# Without context managers:
lock = locket.lock_file("path/to/lock/file")
try:
lock.acquire()
perform_action()
finally:
lock.release()
Locks largely behave as (non-reentrant) Lock instances from the threading module in the standard library. Specifically, their behaviour is:
Locks are uniquely identified by the file being locked, both in the same process and across different processes.
Locks are either in a locked or unlocked state.
When the lock is unlocked, calling acquire() returns immediately and changes the lock state to locked.
When the lock is locked, calling acquire() will block until the lock state changes to unlocked, or until the timeout expires.
If a process holds a lock, any thread in that process can call release() to change the state to unlocked.
Calling release() on an unlocked lock raises LockError.
Behaviour of locks after fork is undefined.
Installation
pip install locket
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for locket-1.0.0-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b6c819a722f7b6bd955b80781788e4a66a55628b858d347536b7e81325a3a5e3 |
|
MD5 | 9454e2c0701d6de9ce1bfa40ab78e9b8 |
|
BLAKE2b-256 | dbbc83e112abc66cd466c6b83f99118035867cecd41802f8d044638aa78a106e |