Python f Codec

Wrap lonesome f-strings in print().

How to install

> python3 -m pip install f-codec

Why?

# -*- coding: f -*-

def dictionary(inst, depth=0):
    ಠ_ಠ = '    ' * depth

    f''' {{
'''

    for name, val in inst.items():
        if isinstance(val, dict):
            f'''
    {name} =
''' > ಠ_ಠ
            dictionary(val, depth + 1)
        else:
            f'''
    {name} = {val},
''' > ಠ_ಠ

    f'''
}}
''' > ಠ_ಠ

decl = {
    'a': 1,
    'b': 2,
    'c': { 'a': 3, 'b': 4 },
    'd': { 'a': 5, 'b': { 'a': 6, 'b': 7 }, 'c': 8 },
    'e': 9
}

f'''
decl =
'''
dictionary(decl)

> python3 test.py

decl = {
    a = 1,
    b = 2,
    c = {
        a = 3,
        b = 4,
    }
    d = {
        a = 5,
        b = {
            a = 6,
            b = 7,
        }
        c = 8,
    }
    e = 9,
}

Development

> python3 -m unittest test_fcodec -v  # run the regression suite
> uv build                            # build sdist and wheel into dist/

The test suite compares decoding against golden files in tests/golden/, which lock in the exact transformation of known-good sources.

TODO

• true streaming support for IncrementalDecoder (currently it buffers the whole input and decodes at once)
• skip f-string if the previous line ends with "" and it isn't a comment
• skip f-string inside another string
• support for ''' > indent on a single-line f-string (only whitespace or a comment may follow the closing quotes)
• support for triple-quoted strings inside an ''' > indent expression
• preserve CRLF line endings (currently normalized to LF)