json/no-unnormalized-keys

Unicode has more than one way to write some characters. é is one character, U+00E9, or an e followed by the combining accent U+0301. The two look the same but are different keys, so a lookup with one doesn't find the other.

This rule reports keys that aren't in the chosen Unicode normalization form, which is NFC by default. --fix rewrites the key, unless it's written with escape sequences.

Configure#

Off by default. It needs an ESLint config that says which files are in the language:

import json from "@eslint/json";

export default [
  {
    files: ["**/*.json"],
    plugins: { json },
    language: "json/json",
    rules: { "json/no-unnormalized-keys": "error" },
  },
];

It also works with json/jsonc and json/json5.

Examples#

The é in this key is an e and a combining accent. Reported:

{
  "café": 1
}

Here it's the single character. Not reported:

{
  "café": 1
}

Options#

export default [
  {
    rules: { "json/no-unnormalized-keys": ["error", { form: "NFD" }] },
  },
];

With NFD, it's the single character that's reported:

{
  "café": 1
}