fpf 0.1.4

filename_filter

file_path_filter can be a simple function:

def filter(filepath):
   return filepath.lower().endswith('.py'):

# Or, for those that like to use lambda:
lambda x: x.lower().endswith('.py)

Applying a filter like like this on a directory should return only python file filepaths. Easy enough.

But there are advanced usecases. Consider our (simplified) project directory:

.
├── build
│   └── lib
│       └── fpf
│           ├── filters.py
│           ├── __init__.py
│           ├── logger.py
│           └── mixins.py
├── fpf
│   ├── filters.py
│   ├── __init__.py
│   ├── logger.py
│   └── mixins.py
├── LICENSE
├── README.md
├── setup.cfg
├── setup.py
├── tests
│  ├── __init__.py
│  └── test_fpf.py
├── .gitignore
└── .travis.yml

Applying the filter, would give us the following result:

from fpf import file_path_filter

for path in file_path_filter(root_dir='.', path_filter=filter ):
   print(path)

build/lib/fpf/__init__.py
build/lib/fpf/filters.py
...
fpf/__init__.py
fpf/filters.py
...
tests/__init__.py
setup.py

• Q: But what if I am not interested in build artifacts or tests?

• A: You can add more conditions to filter(filepath)

• Q: Does that scale?

• A: No

• Q: Is a list of exceptions easy to maintain?

• A: No

• Q: Is creating a library to deal with these usecases overkill?

• A: Maybe. But is too late now.

Introducing ignore files

The Git project and their users had the same issue. They solved this with the .gitignore(pathspec) file.

This library offers some helper classes and functions to apply this to your project.