A thread-safe queue object which is interchangeable with the stdlib Queue. Any overflow goes into a compressed file to keep excessive amounts of queued items out of memory
filequeue is a Python library that provides a thread-safe queue which is a subclass of queue.Queue from the stdlib, filequeue.FileQueue.
filequeue.FileQueue is a drop in, swap out replacement for queue.Queue and will overflow onto disk if the number of items exceeds the specified buffersize, maxsize, instead of blocking or raising Full like the regular queue.Queue.
There is also filequeue.PriorityFileQueue and filequeue.LifoFileQueue implementations, as counterparts to queue.PriorityQueue and queue.LifoQueue.
Note filequeue.FileQueue and filequeue.LifoFileQueue will only have identical behaviour as queue.Queue and queue.LifoQueue respectively if they are initialised with maxsize=0 (the default). See __init__ docstring for details (help(FileQueue))
Note filequeue.PriorityFileQueue won’t currently work exactly the same as a straight out replacement for queue.PriorityQueue. The interface is very slightly different (extra optional kw argument on put and __init__), although it will work it won’t behave exactly the same if using as a swap out replacement. It might still be useful to people though and hopefully I’ll be able to address this in a future version.
- Python 2.5+ or Python 3.x
The motivation came from wanting to queue a lot of work, without consuming lots of memory.
The interface of filequeue.FileQueue matches that of queue.Queue (or Queue.Queue in python 2). With the idea being that most people will use queue.Queue, and can then swap in a filequeue.FileQueue only if the memory usage becomes an issue. (Same applies for filequeue.LifoFileQueue)
Made available as-is under the BSD Licence.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size filequeue-0.4.1.tar.gz (4.6 kB)||File type Source||Python version None||Upload date||Hashes View|