Skip to main content

Profile code executed by pytest

Project description

pytest-line-profiler

PyPI version

Python versions

See Build Status on Travis CI

line-by-line profiling for code executed by pytest, using line-profiler.

Why?

Line profiler is a wonderful tool to easily identify bottlenecks inside specific functions of your code, and quantity the improvements after a refactor.

Using it is straightforward but required to instrument the functions you want to profile with a "virtual decorator" and then execute "a trigger script" (some code that call the decorated functions somehow) via kernprof.py which works as a python wrapper that undertand that decorator and register the functions to be profiled.

Altought it does its job, is a bit invasive: you need to have an special "instrumented" version of your code, and execute it in a way that potentially clashes with the way you do normally (for instance, through a shortcut command from your editor, another script, etc.)

Moreover, frequently in real case scenarios, "a trigger script" isn't just a simple function call. You need to prepare input data, connect to external resources, etc. And that's exactly what a test can do, right?

Installation

You can install "pytest-line-profiler" via pip from PyPI:

$ pip install pytest-line-profiler

Usage

Mark your test passing the functions you wants to profile as positional arguments, like @pytest.mark.line_profile.with_args(function1, function2, [...])

If your test exercises any of those functions, you'll get the profile result as a report.

For example, consider this example

import pytest

def f(i):
    return i * 10

def g(n=10):
    return sum(f(i) for i in range(10))


@pytest.mark.line_profile.with_args(f, g)
def test_as_mark():
    assert g() == 450

After that test is executed, you'll get the stats from the line profiler instance.

============ Line Profile result for tests/test_line_profiler.py::test_as_mark ============
Timer unit: 1e-06 s

Total time: 4e-06 s
File: /home/tin/lab/pytest-line-profiler/tests/test_line_profiler.py
Function: f at line 4

Line #      Hits         Time  Per Hit   % Time  Line Contents
==============================================================
     4                                           def f(i):
     5        10          4.0      0.4    100.0      return i * 10

Total time: 3e-05 s
File: /home/tin/lab/pytest-line-profiler/tests/test_line_profiler.py
Function: g at line 7

Line #      Hits         Time  Per Hit   % Time  Line Contents
==============================================================
     7                                           def g(n=10):
     8         1         30.0     30.0    100.0      return sum(f(i) for i in range(10))

Alternatively, you can run any test passing the function/s to profile from the command line

$ pytest --line-profile path.to.function_to_be profiled [...] 

Contributing

Contributions are very welcome. Tests can be run with pytest, please ensure the coverage at least stays the same before you submit a pull request.

License

Distributed under the terms of the MIT license, "pytest-line-profiler" is free and open source software

Issues

If you encounter any problems, please file an issue along with a detailed description.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Built Distribution

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page