Bug? ignore-words-list seems to have no effect

[joshgoebel]

Am I doing something wrong? How can I get it to stop flagging MSDOS as an error? ignore-words-list seems to have no effect

> codespell
./src/keyszer/models/key.py:161: MSDOS ==> MS-DOS

My config:

[codespell]
ignore-words-list = MSDOS
skip = .venv,.git,build,keyszer.egg-info

[peternewman]

Like so:

[codespell]
ignore-words-list = msdos
skip = .venv,.git,build,keyszer.egg-info

Which admittedly doesn't match our docs:
https://github.com/codespell-project/codespell/#usage

Based on:

codespell/codespell_lib/data/dictionary.txt

Line 20494 in 978a223

MSDOS->MS-DOS

[joshgoebel]

That's it - fixes the issue, but yeah the way i read the docs was that I should try to match case...

Perhaps we need a docfix?

[peternewman]

codespell/codespell_lib/_codespell.py

Lines 456 to 462 in 1d9dee3

	[key, data] = line.split('->')
	# TODO for now, convert both to lower. Someday we can maybe add
	# support for fixing caps.
	key = key.lower()
	data = data.lower()
	if key in ignore_words:
	continue

I think the issue may be that MSDOS is one of the few entries which is upper case in the dictionary, hence flagging this issue.

[mwestphal]

Looks like there is a case issue:

nNumber ==> number

ignore-words-list=nNumber

Not excluded

ignore-words-list=nnumber

excluded successfully.

[peternewman]

Looks like there is a case issue:

Yeah our comparison to match the dictionary from the ignore list is happening with different case requirements to what we claim in the docs.

[yarikoptic]

FWIW just ran into this issue and was about to complain as well -- would be great to have it fixed, as to conform to the promise of being case sensitive. My use case:

❯ codespell --help | grep -A3 -e ' -L'
  -L WORDS, --ignore-words-list WORDS
                        comma separated list of words to be ignored by
                        codespell. Words are case sensitive based on how they
                        are written in the dictionary file
❯ cat spellme.txt
Ingress for Microsoft Azure Kubernetes Service (AKS)
❯ codespell -L AKS spellme.txt
spellme.txt:1: AKS ==> ASK
❯ codespell --version
2.2.2

so I would dislike completely avoiding fixing aks into ask ;)

[iamibi]

I am facing a similar issue mentioned here. When using ignore-words-list as part of pyproject.toml to ignore "OT" as a words, it is not respecting it.
full configuration for codespell

[tool.codespell]
quiet-level = 0
skip = "poetry.lock"
ignore-words-list = "OT"

Any suggestions?

[yarikoptic]

I some times add those into ignore-regex instead with end of the word boundary, like ignore-regex = "\bOT\b" if I want to retain regular to to be a typo to be detected.

[DimitriPapadopoulos]

@iamibi As already explained:

ignore-words-list = "ot"

[yarikoptic]

that would also skip lower case ot and I thought that the goal here is to just skip upper cased OT only.

[perillo]

I'm having a similar problem when checking Zig source files.
codespell refuses to ignore ACCES and WRONLY.

[DimitriPapadopoulos]

@perillo What's your question ?

[perillo]

@perillo What's your question ?

Looking at the comments, I assumed that using lowercase for ignore words was a temporary fix. However https://github.com/codespell-project/codespell/#ignoring-words documents that this is the expected behavior.

Thanks.