Contributing¶
This documentation is for people who want to contribute code to the project, whether fixing a small bug, adding support for a new device, or even discussing a completely new device type.
In short¶
Reporting issues¶
We use the github issue tracker at https://github.com/python-microscope/microscope/issues. When reporting an issue please include as much information as you can from:
- Steps to reproduce issue
Include information so that we can try it ourselves. Don’t just say “camera fails when setting exposure time”. Instead, include the code to reproduce the issue and the error message.
- Operating system
MacOS 10.15, Ubuntu 18.04, Windows 7, Windows 10, etc…
- Python version
Also include the python minor version number, i.e, Python 3.7.3 or 3.6.2. On command line, this is the output of
python --version
.
- Device, SDK, and firmware information
Include the device model, revision number, and serial number. Also include the firmware and the device SDK version number.
- Pyro version
If the issue only happens in the network, please also include the version of the Pyro package.
Requesting support for new device¶
To request support for a new device, or even to support for a feature of an already supported device, open a new issue on the issue tracker.
If there’s already an open issue requesting the same, please leave a comment so that we know more people want it.
Fixing an issue¶
To fix an issue, including adding support for a new device, please do the following:
Open a new issue (if not already opened)
On the commit message, refer to the issue number
Comment on the issue what branch or commit to pull to fix it.
Why the non-standard procedure?¶
This procedure to fix an issue is not very standard on github projects. However, it prevents spreading the discussion of the issue over multiple pages and enables one to find that discussion from the git history.
Coding standards¶
Code format style¶
Let us not discuss over code formatting style. Code formatting is
handled by Black. Simply run
black
at the root of the project after making changes and before
making a commit:
black .
Docstrings¶
The API section of the documentation is generated from the inlined docstrings. It requires the Google Python docstrings format minus the type description on the list of arguments since those are defined in the type annotations. It looks like this:
def func(arg1: str, arg2: int) -> bool:
"""One-line summary.
Extended description of function.
Args:
arg1: Description of arg1. This can be a multi-line
description, no problem.
arg2: Description of arg2.
Returns:
Description of return value
"""
return True
Commit messages¶
The first line of the commit message have a one-line summary of the change. This needs to mention the class, function, or module where relevant changes happen. If there’s an associated issue number, make reference to it. If it is a “fixup” commit, make reference to it. Some examples for changes:
limited to a method function:
TheClassName.enable: fix for firmware older than Y (#98)
effecting multiple methods in a class:
TheClassName: add support for very fancy feature (#99)
fixing a typo or obvious mistake on a previous commit:
AndorAtmcd: pass index to super (fixup a16bef042a41)
documentation only:
doc: add example for multiple cameras with hardware triggering
Test suite¶
Most of Python Microscope is about controlling very specific hardware
and there are no realist mocks of such hardware. Still, there are
some tests written. They can be run with tox. The repository has the required
configuration, so simply run tox
at the root of the repository.
All test units, as well as other tools for testing purposes, are part
of the microscope.testsuite
package.
If your changes do not actually change a specific device, please include a test unit.
Copyright¶
We do not request that copyright is assigned to us, you can remain the copyright holder of any contribution made. However, please ensure that you are the copyright holder. Depending on your contract, even if you are a student, the copyright holder is likely to be your employer or university. Ask your employer or PhD supervisor if you are not sure.