Vainu pylinter

The purpose of this module is to allow easy import to the vainu repositories using python.

The repository contains

• custom_runner.py : Lints each file separately. If any of the files fail, the runner will return exit code 1.
• vainupylinter : console script

Current implementation is python2 and python3 compatible. This can change in the future.

Installation

Install the module with pip:

pip install git+https://github.com/vainuio/vainupylinter.git

Add the module to requirements.txt using "git+https://github.com/vainuio/vainupylinter.git".

Requirements

pylint

And that's it!

How to use

To check files with custom pylinter, use

vainupylinter <FNAME1> <FNAME2> ...

DEVELOPING

Make sure that you have enabled commit hooks in .githooks:

vainupylinter <FNAME1> <FNAME2> ...

Inputs

Argument	Type	Default	Required	Description
	str		Yes	File paths to be checked. Separated by space
--thresh, -t	float	9.2	No	Threshold for files to be considered OK
--rcfile, -r	str	""	No	Custom path for .pylintrc file. If not given, defaults to pylint logic.
--allow-errors, -e	bool	False	No	Return zero exit-code even though pylint emits an error-level message
--keep-results, -r	bool	False	No	Do not clean up runner in between runs.
--custom_path, -cp	str	""	No	Use custom rules and scoring. See Custom settings below.

Custom settings

Vainupylinter supports following features:

1. Custom rules:
   • Allows users to define their own logic whether file should be accepted. For example, one can implement logic that does not allow pdb.set_trace() to be used
   • The function MUST BE named custom_rules. Input is expected to be a 1) stats from linter result and file path and output has to be tuple containing 2 floats: did the file pass and should the result override the standard check.
2. Custom scoring:
   • Allows users to define custom scoring function. For example, if one wants to treat certain type of warning as fatal message.
   • The function MUST BE named custom_score. stats containing the pylint result statistics is given as input and output is expected to be boolean
3. Custom thresholding
   • Allows more complex threshold checks. For example, files in specified subdirectories can have different threshold than in main.
   • The function MUST BE named custom_thresholding. The function takes score, default threshold and fname as input, and returns bool (passed or not)

These functions should be defined in the same python file. The --custom_path argument needs to be module path (so rules.custom_rules instead of rules/custom_rules.py). You may have to add __init__.py for the import to work. The file must be in the directory or subdirectory of the directory vainupylinter is called.

In the following file structure, vainupylinter can find rules from project, folder1 and subfolder1 when called at project-level. If vainupylinter is called in subfolder1, rules cannot be found from folder1or its parent directory.

project
│   README.md
│   __init__.py  
│
└───folder1
│   │   __init__.py
│   │  code1.py
│   │
│   └───subfolder1
│       │   __init__.py
│       │   
│       │   ...
│   
└───folder2
    │   code2.py

Remember, that if you want the option to silence the defined custom warnings, you have to add logic for it in custom_rules function.