• Copyright (c) 2012-2024 SATOH Fumiyasu @ OSSTech Corp., Japan
• Copyright (c) 2005 Matsumoto, Tadashi (original author)
• License: BSD
• Development home: https://github.com/fumiyas/python-nkf
• Author's home: https://fumiyas.github.io/

Install from PyPI:

$ sudo pip install nkf

Install from source tree:

$ make
$ sudo make install

Run tests:

$ make test

## flag is same as the flags of nkf itself
import nkf
output = nkf.nkf(flag, input)

## For example, to convert from euc-jp to utf-8
output = nkf.nkf('-Ew', 'some euc-jp string')

## Options can also be passed as a list or tuple
output = nkf.nkf(['-E', '-w'], 'some euc-jp string')
output = nkf.nkf(['--ic=euc-jp', '--oc=utf-8n'], 'some euc-jp string')

## Guess character encoding
input_encoding = nkf.guess('some string')

## Guess character encoding and newline type
encoding, newline = nkf.guess_detail('some string')

guess() function guesses an input string encoding and returns one of next strings:

• BINARY
• ASCII
• Shift_JIS
• CP932
• EUC-JP
• EUCJP-MS
• CP51932
• ISO-2022-JP
• CP50221
• CP50220
• UTF-8
• UTF-16
• UTF-32

guess_detail() function returns a tuple of (encoding, newline). encoding is same as guess(), and newline is one of:

• None (no newline detected)
• LF
• CR
• CRLF
• MIXED