If you want to contribute to the project there are a number of ways to do it:
- There will be bugs in the code, some systems will break, some will give you an odd answer, be wary. If and when you do find bugs, if they are not big then you should create an issue, or check the current issues and add your comments to that. When flagging a bug please give useful things like terminal output, error messages, what input you used and if possible, highlight where in the source code you think the bug is.
- If you just want to get involved and give something back, we are in the process of developing tests for all our functions, currently the major ones are implemented, but there are lots left to go.
- If you want to learn about the codes, how they work, check where they could be improved etc., then fork the repo and scan through them - make them more readable, improve the docs, anything you like, it’s all appreciated.
- If there is something you want to see in the package, you could check to see if we have it on our list of things to develop, and if not, add it!
If you want to have a go at adding a feature (which would be great) then please fork the repo on github, and create your own feature branch. We are trying to follow the github flow branching model, so you might want to install their command-line tools.
Then when you are done, create a pull request and we can integrate it back so everyone can use it :)
We are trying to migrate to a truly pep8-friendly coding style, to that end I would recommend installing flake8. Our main two developers use the atom text editor along with linter-flake8, although I often use vim as well. Up to you.
- Don’t start and end function names with double underscores, these are reserved for python;
- Never use dashes (I hate them);
- Use descriptive names, in contrast to the pep257 complaints about overly complicated variable names, if it helps the readability then do it;
- If you want an internal or hidden function, start the function name with a single underscore;
- Don’t use CamelCase for module names, separate words with underscores.
As you will have seen from the API, docs are good. We generate the docs using sphinx which can then be compiled either locally, or by readthedocs. In this way, docstrings are including in the source code.
- We work on the principle that If it’s not documented, it doesn’t exist.
- We also try to work on the principle of just enough docs: don’t over document the functions within the script, if you think they need more explaining then the chances are you are doing it wrong. We work on this basis because the more complicated the docs are, the less likely we are to maintain them.
- If you find yourself explaining a method in the docstrings: don’t, put a link to a paper.