Simple tool to check if Unicode text files are Unicode-normalized
pip3 install unicodecheck
unicodecheck -iv SPAM.txt
To check files in a directory recursively:
unicodecheck -ivr Ham/Eggs/
The main program can be invoked either through the unicodecheck
command or through the Python main module option python3 -m unicodecheck
.
usage: unicodecheck [-h] [-V] [-m {NFC,NFD,NFKC,NFKD}] [-d] [-u [NUMBER]] [-r] [-i] [-v]
PATH [PATH ...]
positional arguments:
PATH describe input file or directory (pass '-' to specify stdin)
options:
-h, --help show this help message and exit
-V, --version show program's version number and exit
-m {NFC,NFD,NFKC,NFKD}, --mode {NFC,NFD,NFKC,NFKD}
target Unicode normalization (default: NFC)
-d, --diff show diffs between the original and normalized (default: False)
-u [NUMBER], -U [NUMBER], --unified [NUMBER]
show unified diffs with NUMBER lines of context [NUMBER=3] (default: False)
-r, --recursive follow the directory tree rooted in each PATH argument (default: False)
-i, --include-hidden include hidden files and directories (default: False)
-b PATTERN [PATTERN ...], --blacklist PATTERN [PATTERN ...]
notify if having PATTERN (case-sensitive) (default: None)
-e, --error return non-zero exit code on detection (default: False)
-v, --verbose report non-essential logs (default: False)
The convmv
command is a good alternative to using this application.
convmv -f utf8 -t utf8 --nfc -r ./
convmv -f utf8 -t utf8 --nfd -r ./
- This tool doesn't provide auto in-place (write) file normalization because Unicode normalization doesn't guarantee content equivalence.
- The procedure for determining the binary file refers to Git's algorithm.
MIT