{
    "info": {
        "author": "Mateu C\u00e0naves",
        "author_email": "mcanaves@hundredrooms.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 3 - Alpha",
            "Environment :: Web Environment",
            "Intended Audience :: Developers",
            "License :: OSI Approved :: BSD License",
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 3",
            "Programming Language :: Python :: 3.6",
            "Programming Language :: Python :: 3.7",
            "Topic :: Internet :: WWW/HTTP"
        ],
        "description": "# hr-prometheus\n\n[![Build Status](https://travis-ci.com/HundredRooms/hr-prometheus.svg?branch=master)](https://travis-ci.com/HundredRooms/hr-prometheus)\n[![codecov](https://codecov.io/gh/HundredRooms/hr-prometheus/branch/master/graph/badge.svg)](https://codecov.io/gh/HundredRooms/hr-prometheus)\n[![PyPI version](https://badge.fury.io/py/hr-prometheus.svg)](https://badge.fury.io/py/hr-prometheus)\n\nPrometheus integration for aiohttp projects.\n\nhr-prometheus adds support for providing [aiohttp](https://aiohttp.readthedocs.io/en/stable/) applications metrics to [prometheus](https://prometheus.io/docs/introduction/overview/). It is implemented as a aiohttp middleware.\n\nCurrently, it exports the following metrics via the /metrics endpoint by default:\n\n- request_latency: Elapsed time per request in seconds.\n  - Labels exported: method (HTTP method), path\n- request_count: Number of requests received.\n  - Labels exported: method (HTTP method), path, status (HTTP status)\n- requests_in_progress: In progress requests.\n  - Labels exported: method (HTTP method), path\n\nDefault request behaviour can be modified by passing a custom `RequestMonitor` to the middleware. You can find out how to do it in [advanced section](#advance-usage).\n\n## Installation\n\n```shell\npip install hr-prometheus\n```\n\n## Usage\n\nBriefly, the following is all you need to do to measure and export prometheus metrics from your aiohttp web application:\n\n```python\nfrom aiohttp import web\nfrom hr_prometheus import hrprometheus_middleware, hrprometheus_view\n\napp = web.Application()\napp.router.add_get(\"/metrics\", hrprometheus_view)\napp.middlewares.append(hrprometheus_middleware())\n```\n\n### Advanced usage\n\n#### Custom monitors\n\nTo modify the default behavior you simply need to create a new monitor that inherits from the `BaseRequestMonitor` and pass the class to the middleware.\n\nThis class provides two public methods. `update_init_metrics` and `update_end_metrics`.\nThese methods are executed at the beginning and end of a request respectively. Simply add the metrics you want at each point.\n\nHere's an example taken from the default monitor.\n\n```python\nfrom aiohttp import web\nfrom hr_prometheus import hrprometheus_middleware, hrprometheus_view\nfrom hr_prometheus.monitors import BaseRequestMonitor\n\n\nclass RequestMonitor(BaseRequestMonitor):\n    REQUEST_COUNT = Counter(\n        \"request_count\", \"Number of requests received\", [\"method\", \"path\", \"status\"]\n    )\n    REQUEST_LATENCY = Histogram(\n        \"request_latency\", \"Elapsed time per request\", [\"method\", \"path\"]\n    )\n    REQUEST_IN_PROGRESS = Gauge(\n        \"requests_in_progress\", \"Requests in progress\", [\"method\", \"path\"]\n    )\n\n    def update_init_metrics(self):\n        self.REQUEST_IN_PROGRESS.labels(*self.request_description).inc()\n\n    def update_end_metrics(self):\n        resp_time = time.time() - self.init_time\n        self.REQUEST_COUNT.labels(*self.request_description, self.response_status).inc()\n        self.REQUEST_LATENCY.labels(*self.request_description).observe(resp_time)\n        self.REQUEST_IN_PROGRESS.labels(*self.request_description).dec()\n\n\napp = web.Application()\napp.router.add_get(\"/metrics\", hrprometheus_view)\napp.middlewares.append(hrprometheus_middleware(RequestMonitor))\n```\n\n#### Grouping dynamic routes\nIn aiohttp you may define dynamic routes by parametrizing the route path (e.g. `/v1/resource/{resource_id}`). If you are interested in grouping the different values for a given parameter under the same metrics you can do so by specifying the fixed parameters for a named route to the request monitor (you can do so trhough the middleware for convinience).\n\nHere is an example of an api returning neighbouring cells from a matrix\n```python\nfrom aiohttp import web\nfrom hr_prometheus import hrprometheus_middleware\nfrom my_project.views import get_cell_neighbour_view\n\nmiddleware = hrprometheus_middleware(fixed_routes_parameter={\"get_cell_neighbour\": [\"cell_id\"]})\napp = web.Application(middlewares=[middleware])\napp.add_route(\"GET\", \"/cell_neighbour/{cell_id}/direction/{direction}\", get_cell_neighbour_view, name=\"get_cell_neighbour\")\n```\nThis way requests with path `/cell_neighbour/1948/direction/north` and `/cell_neighbour/874/direction/north` are both collapsed into `\"/cell_neighbour/{cell_id}/direction/north\"`\n\nThis is especially useful when you have a wide range of possible values for a path parameter and you are only interested in the overall monitoring, thus avoiding metrics namespace pollution.",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/hundredrooms/hr-prometheus",
        "keywords": "",
        "license": "BSD",
        "maintainer": "",
        "maintainer_email": "",
        "name": "hr-prometheus",
        "package_url": "https://pypi.org/project/hr-prometheus/",
        "platform": "",
        "project_url": "https://pypi.org/project/hr-prometheus/",
        "project_urls": {
            "Homepage": "https://github.com/hundredrooms/hr-prometheus"
        },
        "release_url": "https://pypi.org/project/hr-prometheus/0.2.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Prometheus integration for aiohttp framework.",
        "version": "0.2.0"
    },
    "last_serial": 4637656,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d201606596bc99984b2d36b79dcd0fa2",
                    "sha256": "1950924e7295293eeec5fd430af47306890b24717dffe726d26e91b92a2e02d7"
                },
                "downloads": -1,
                "filename": "hr-prometheus-0.1.0.tar.gz",
                "has_sig": false,
                "md5_digest": "d201606596bc99984b2d36b79dcd0fa2",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3562,
                "upload_time": "2018-12-18T13:31:12",
                "url": "https://files.pythonhosted.org/packages/54/93/a1cb23166fd5a1f5fe06c55027ea5f2cb7f1bc9b19a5552cf3f80aaa9075/hr-prometheus-0.1.0.tar.gz"
            }
        ],
        "0.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "30602924d2100067e18f6ff264469cb2",
                    "sha256": "184af6deccfa94e29a0fd564a5f51ddec722841004b629dab8de0d3484c77c30"
                },
                "downloads": -1,
                "filename": "hr-prometheus-0.1.1.tar.gz",
                "has_sig": false,
                "md5_digest": "30602924d2100067e18f6ff264469cb2",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3568,
                "upload_time": "2018-12-18T14:02:49",
                "url": "https://files.pythonhosted.org/packages/f0/3f/bdebdb0aa5f16c251f272783ba64b0a6b94a1a1c72450a416776b9cde2fe/hr-prometheus-0.1.1.tar.gz"
            }
        ],
        "0.2.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "9606a8b0efcc1881baf1eaa88d12af00",
                    "sha256": "877f069cecd64eaba7b241e85c1847e1dc335344f83a32b029b64a6ec717750a"
                },
                "downloads": -1,
                "filename": "hr-prometheus-0.2.0.tar.gz",
                "has_sig": false,
                "md5_digest": "9606a8b0efcc1881baf1eaa88d12af00",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 4552,
                "upload_time": "2018-12-27T11:45:40",
                "url": "https://files.pythonhosted.org/packages/d8/9b/09316b5388ce4f471517ef94562779fb0b8f9339349c3709bcbda68332a7/hr-prometheus-0.2.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "9606a8b0efcc1881baf1eaa88d12af00",
                "sha256": "877f069cecd64eaba7b241e85c1847e1dc335344f83a32b029b64a6ec717750a"
            },
            "downloads": -1,
            "filename": "hr-prometheus-0.2.0.tar.gz",
            "has_sig": false,
            "md5_digest": "9606a8b0efcc1881baf1eaa88d12af00",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 4552,
            "upload_time": "2018-12-27T11:45:40",
            "url": "https://files.pythonhosted.org/packages/d8/9b/09316b5388ce4f471517ef94562779fb0b8f9339349c3709bcbda68332a7/hr-prometheus-0.2.0.tar.gz"
        }
    ]
}