{
    "info": {
        "author": "Jason Moiron",
        "author_email": "jmoiron@jmoiron.net",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 4 - Beta",
            "Intended Audience :: Developers",
            "License :: OSI Approved :: MIT License",
            "Operating System :: POSIX"
        ],
        "description": "gaspar\n======\n\nGaspar is a library for creating small, simple TCP daemons that parallelize CPU\nintensive work with a simple and low-overhead request/response pattern.\n\nIt does this by forking off a number of worker processes, using eventlet to\nhandle incoming requests and the 0MQ push/pull message pattern to load\nballance work across the worker processes and receive responses.\n\n\nrunning servers\n---------------\n\nGaspar uses the terms ``producer`` and ``consumer`` for the process that receives\nincoming requests and the processes that actually handle those requests.  In the\n0MQ documentation, these are called ``ventilator`` and ``sink``, and various\nother terms are used throughout distributed systems literature.  To use Gaspar,\nyou need only to create a producer and a consumer, and then start the producer::\n\n    >>> import gaspar\n    >>> def echo(message): return message\n    >>> consumer = gaspar.Consumer(handler=echo)\n    >>> producer = gaspar.Producer(consumer, 10123)\n    >>> producer.start()\n\nThe ``start`` call will block, and at this point you will have a server which is\nlistening on port ``10123``, receiving requests, sending them to a number of workers\n(default is the # of CPUs on your machine), and then replying based on the echo\nhandler.\n\n\nrequests\n--------\n\nGaspar's default Producer takes requests on a simple naked TCP port.  These\nrequests are:\n\n   * a 4-byte unsigned int in network-byte order (``struct.pack('!I', N)``)\n   * a string of that length\n\nThe reply is simply a string followed by the termination of the socket.  The\nconvenience function ``gaspar.client.request(\"host:port\", message)`` will send a\nrequest and return the reply synchronously.  It uses the basic ``socket``\nlibraries, so you can \"green\" it safely with eventlet or gevent's monkey\npatching methods.\n\n``gaspar.client`` also provides a function called ``pack`` which takes a string\nand returns a new string with the 4-byte message length prepended.  If you\nare using a gaspar daemon with async frameworks that are not greenlet based,\nyou can use this to cover that aspect of the client protocol.\n\nlimitations\n-----------\n\nformless request/response\n~~~~~~~~~~~~~~~~~~~~~~~~~\n\nGaspar requests and responses are just strings.  There is no standard way to\nserialize multiple arguments or return multiple values.  Because the nature of\nthe work being farmed out to such a daemon could be defeated by the wrong\ncalling semantics, these details are left to the ``Consumer`` implementation\nand to postprocessing client responses.\n\nsingle-server operation\n~~~~~~~~~~~~~~~~~~~~~~~\n\nAlthough the technologies in use (TCP and 0MQ) would allow for daemons to be\nspread across systems, this wasn't an original design goal of Gaspar and it\nis not currently supported.\n\n\nwhy shouldn't I use celery?\n---------------------------\n\nThe major \"advantages\" of Gaspar over Celery are its small size, conceptual\nsimplicity, and infrastructureless operation.  The purpose of Gaspar was to\nmake it very easy to remove CPU bound processes from a tight event based I/O\nloop (like eventlet, gevent, tornado, et al), turn it into I/O wait, and\nspread that work across multiple cores.\n\nCelery serves a much broader range of purposes, is a lot more sophisticated,\nand has features like delayed and recurrent execution that Gaspar lacks.  If\nyou have a number of assorted tasks you need to execute asynchronously, Celery\nis very good at this.  If you have an asynchronous worker that has a few very\nCPU-intensive tasks that are blocking the event loop, Gaspar allows you to\nfarm that work out to daemons with very little code.",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "UNKNOWN",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "http://github.com/jmoiron/gaspar",
        "keywords": "eventlet zmq parallel prefork",
        "license": "MIT",
        "maintainer": null,
        "maintainer_email": null,
        "name": "gaspar",
        "package_url": "https://pypi.org/project/gaspar/",
        "platform": "UNKNOWN",
        "project_url": "https://pypi.org/project/gaspar/",
        "project_urls": {
            "Download": "UNKNOWN",
            "Homepage": "http://github.com/jmoiron/gaspar"
        },
        "release_url": "https://pypi.org/project/gaspar/1.0/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "gaspar eventlet zmq parallel worker",
        "version": "1.0"
    },
    "last_serial": 792278,
    "releases": {
        "1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "5770d4b14cd45d6e1c07d6885708b43d",
                    "sha256": "b12b2dbf6c5dd9288343085ef7f12c491fb9a7ec36a4f2674e033bbf03285f39"
                },
                "downloads": -1,
                "filename": "gaspar-1.0.tar.gz",
                "has_sig": false,
                "md5_digest": "5770d4b14cd45d6e1c07d6885708b43d",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 7342,
                "upload_time": "2012-01-04T00:23:56",
                "url": "https://files.pythonhosted.org/packages/5d/85/a960c3c2e7d4fa127887cbbcb9f26fc4e8785ba5cdb5131a3f7260f6640b/gaspar-1.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "5770d4b14cd45d6e1c07d6885708b43d",
                "sha256": "b12b2dbf6c5dd9288343085ef7f12c491fb9a7ec36a4f2674e033bbf03285f39"
            },
            "downloads": -1,
            "filename": "gaspar-1.0.tar.gz",
            "has_sig": false,
            "md5_digest": "5770d4b14cd45d6e1c07d6885708b43d",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 7342,
            "upload_time": "2012-01-04T00:23:56",
            "url": "https://files.pythonhosted.org/packages/5d/85/a960c3c2e7d4fa127887cbbcb9f26fc4e8785ba5cdb5131a3f7260f6640b/gaspar-1.0.tar.gz"
        }
    ]
}