Giters

sindresorhus / eslint-plugin-unicorn

More than 100 powerful ESLint rules

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

`prevent-abbreviations` - non-ASCII characters ignored in filenames

Kristinita opened this issue · comments

commented

1. Summary

If I use eslint-plugin-html and check JavaScript inside HTML by using Unicorn, I get errors like this:

8:3  error  The filename `Мiръ.html` should be named `Мindexръ.html`. A more descriptive name will do too  unicorn/prevent-abbreviations

✖ 1 problem (1 error, 0 warnings)

2. MCVE

2.1. Files

.eslintrc.yml:

extends:
- plugin:unicorn/all

plugins:
- html
- unicorn

Мiръ.html (“мiръ” meaning “world”):

<!DOCTYPE html>
<html lang="en">
<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Document</title>
	<script>
		console.log("Kira Goddess!");
	</script>
</head>
<body>
	<p>Kira Goddess!</p>
</body>
</html>

2.2. Steps to reproduce

npm install --save-dev eslint eslint-plugin-html eslint-plugin-unicorn

npx eslint Мiръ.html

2.3. Behavior

2.3.1. Desired

No errors.

2.3.2. Current
8:3  error  The filename `Мiръ.html` should be named `Мindexръ.html`. A more descriptive name will do too  unicorn/prevent-abbreviations

✖ 1 problem (1 error, 0 warnings)

3. Possible implementations of the desired behavior

  1. Make it possible to ignore HTML files via the ignore option or by introducing a new option such as checkFilenamesExtensions. For example:

    unicorn/prevent-abbreviations:
- error
- checkFilenamesExtensions:
  - ".js"

    In this case, Unicorn will solely check names of files with the .js extension, but not with other extensions like .html or .md (if the user is using eslint-plugin-markdown).

  2. Make it possible to skip filenames like Мiръ containing non-ASCII symbols. For example, via the option checkNonASCIIFilenames: false.

4. Not helped

  1. I couldn’t find on Google and in this repository how I can solve the problem.

  2. I can’t ignore all .html files with the ignore option. prevent-abbreviations documentation:

    When checking filenames, only the basename is tested. For example, with a file named foo.e2e.js, ignore: [/\.e2e$/] would pass and ignore: [/\.e2e\.js/] would fail.

5. Don’t offer

5.1. “Just use checkFilenames: false”

I need to ignore solely HTML files, I want to use checkFilenames for JavaScript files. checkFilenames: false also ignores checking of .js files.

5.2. “Just ignore specific filenames”

Like this:

rules:
  unicorn/prevent-abbreviations:
  - error
  - ignore:
    - "Мiръ"

It would be nice if it were possible to ignore all .html files instead of ignoring files individually. I solely use ASCII symbols for my JavaScript filenames, but I don’t think HTML filenames should have the same restrictions. I haven’t found reasons why I shouldn’t use filenames like Мiръ for HTML files.

6. Environment

  1. Microsoft Windows 11 [Version 10.0.22621.3085]
  2. ESLint 8.57.0
  3. eslint-plugin-html 8.0.0
  4. eslint-plugin-unicorn 51.0.1

Thanks.

commented

It looks like the bug is that it's not counting non-ASCII characters, so it sees "mi" instead of the full filename.

This is a bug and can be fixed.

If you want to disable specific rules on specific files, you can already do so via eslint overrides.