Skip to main content

Tornado Async Utilities

Project description

Maintained license PythonVersions status PyPI Requirements Status Build Status Coverage Status

tor-async-util is a set of utilities that are useful when implementing RESTful APIs using Tornado’s Asynchronous and non-Blocking I/O.

Features

  • when async curl httpclient is used, it’s useful to know if libcurl was compiled with an async dns resolver - see is_libcurl_compiled_with_async_dns_resolver()

  • instead of CTRL+C generating an unfriendly stack trace install a signal handler - see install_sigint_handler()

  • a default request handler which generates a RESTful API friendly not found response - see DefaultRequestHandler()

  • an abstract base class from which all request handler classes can be derived to provide

  • read and write json requests and responses optionally verifying each against a jsonschema - see RequestHandler.get_json_request_body() and RequestHandler.write_and_verify()

  • accessing decoded BASIC auth credentials - see RequestHandler.get_basic_auth_creds()

  • augment Tornado’s default set_status() with support for additional status codes - see RequestHandler.set_status()

  • override Tornado’s default write_error() so a json response body is generated rather than the default HTML response body - see RequestHandler.write_error()

  • thin wrapper around ConfigParser.ConfigParser to parse ini files for things settings such as logging levels, keyczar crypters and keyczar signers - see Config

  • core implementations of /_version, /_noop and /_health endpoints include async health checkers - see generate_version_response(), generate_noop_response(), generate_health_check_response() and AsyncHealthCheck

  • this explains that the time_info attribute of a tornado response object contains timing details of the phases of a request which is available when using the cURL http client. a description of these timing details can be found at here. these timings are very, very helpful in understanding latencies from interactions between microservices - see write_http_client_response_to_log(). an example of what the logs look like is below

2016-01-23T03:45:53.362+00:00 INFO async_docker_remote_api 'Remote Docker API' took 3.42 ms to
respond with 200 to 'GET' against >>>http://127.0.0.1:4243/containers/cid/logs?stdout=1<<< - timing
detail: q=0.13 ms n=0.03 ms c=0.04 ms p=1.65 ms s=1.66 ms t=1.83 ms r=0.00 ms
  • integration tests often run database installer(s), start up service(s) and then direct various requests at the service(s). when the tests fail it’s very useful to output the logs associated with the installers and services. The nose plug-in tor_async_util.nose_plugins.FileCapture is used in integration tests to identify the files that should be displayed on test failure. in order for tor_async_util.nose_plugins.FileCapture to work as desired it must be registered prior to running tests. tor_async_util_nosetests.py is responsible for registering tor_async_util.nose_plugins.FileCapture and is as a replacement for nosetests as per the instructions documented here

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

tor_async_util-1.14.0.tar.gz (16.1 kB view hashes)

Uploaded Source

Built Distribution

tor_async_util-1.14.0-py2-none-any.whl (18.1 kB view hashes)

Uploaded Python 2

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page