Fix the name of the extra word lists we load

[peternewman]

I'm a bit unclear why this fix didn't trip it up TBH:
https://github.com/codespell-project/codespell/tree/peternewman-test-filename-checking

Ideas welcome, and would be sort of good to solve before we just merge this...

[DimitriPapadopoulos]

Indeed:

$ cd codespell_lib/tests/data
$ 
$ ls
en_GB-additionnal.wordlist
en_US-additionnal.wordlist
$ 
$ codespell -f
$ 

[DimitriPapadopoulos]

It's the dash, and I think it's expected:

$ cat > additionnal.txt
additionnal
$ 
$ cat > list-additionnal.txt
list-additionnal    
$ 
$ cat > list_additionnal.txt
list_additionnal
$ 
$ codespell -f
./additionnal.txt: additionnal ==> additional
./additionnal.txt:1: additionnal ==> additional
$ 

[peternewman]

It's the dash, and I think it's expected:

Yeah I'd expect the hyphen to, for e.g. semi-skimmed, but I tried to change it to another underscore hear and it still didn't catch it:
https://github.com/codespell-project/codespell/tree/peternewman-test-filename-checking

Which might be an outstanding issue around snake_case checking...

[DimitriPapadopoulos]

Indeed, - is part of the default word definition, but not _:

codespell/codespell_lib/_codespell.py

Line 31 in 47064e0

word_regex_def = "[\\w\\-'’`]+"

It looks like _ is part of \w:

>>> import re
>>> 
>>> word_regex_def = "[\\w\\-'’`]+"
>>> word_regex = re.compile(word_regex_def)
>>> 
>>> word_regex.findall("list-additional")
['list-additional']
>>> 
>>> word_regex.findall("list_additional")
['list_additional']
>>> 
>>> word_regex.findall("list additional")
['list', 'additional']
>>> 
>>> word_regex.findall("list.additional")
['list', 'additional']
>>> 

From https://docs.python.org/3/library/re.html#regular-expression-syntax:

\w

For Unicode (str) patterns:
Matches Unicode word characters; this includes alphanumeric characters (as defined by str.isalnum()) as well as the underscore (_). If the ASCII flag is used, only [a-zA-Z0-9_] is matched.

For 8-bit (bytes) patterns:
Matches characters considered alphanumeric in the ASCII character set; this is equivalent to [a-zA-Z0-9_]. If the LOCALE flag is used, matches characters considered alphanumeric in the current locale and the underscore.

[DimitriPapadopoulos]

Is there an easy way to get \w except _ in the non-ASCII case?

[DimitriPapadopoulos]

See #2979 for snake_case checking.