Add eslint rule to prevent bracket syntax for property access

[alexdima]

No description provided.

[Copilot]

Pull Request Overview

This PR adds a new ESLint rule code-no-bracket-notation-for-identifiers to discourage the use of bracket notation for accessing properties that are valid JavaScript identifiers. The rule aims to prevent bypassing TypeScript's type checking and access modifiers by using bracket notation (e.g., obj['_privateMember'] instead of obj._privateMember).

Key changes:

• New ESLint rule implementation with auto-fix capability
• Rule is enabled globally with 'warn' level in the ESLint configuration

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
eslint.config.js	Enables the new rule code-no-bracket-notation-for-identifiers at the 'warn' level
.eslint-plugin-local/code-no-bracket-notation-for-identifiers.ts	Implements the ESLint rule to detect and auto-fix bracket notation usage for valid identifiers

[Copilot]

The fix implementation replaces the entire MemberExpression node with reconstructed text. This approach fails to preserve whitespace, comments, and complex expressions properly. It should use fixer.replaceTextRange targeting only the bracket portion (from the opening bracket to the closing bracket) and replacing it with .${propertyName}. For example, replace the range from the start of '[' to the end of ']' instead of the entire node.

Suggested change

	// We need to replace the ['property'] part with .property
	const sourceCode = context.getSourceCode();
	const objectText = sourceCode.getText(memberExpr.object as unknown as eslint.Rule.Node);
	const dotNotation = `${objectText}.${propertyName}`;
	return fixer.replaceText(node, dotNotation);
	// Only replace the ['property'] part with .property, preserving whitespace/comments
	const sourceCode = context.getSourceCode();
	const leftBracket = sourceCode.getTokenBefore(memberExpr.property, token => token.value === '[');
	const rightBracket = sourceCode.getTokenAfter(memberExpr.property, token => token.value === ']');
	if (!leftBracket || !rightBracket) {
	return null;
	}
	return fixer.replaceTextRange(
	[leftBracket.range[0], rightBracket.range[1]],
	'.' + propertyName
	);

[Copilot]

The regular expression doesn't account for reserved JavaScript keywords. Properties like 'class', 'const', 'function', etc. are valid in bracket notation but cannot be used with dot notation. Add a check to exclude reserved keywords: const reservedWords = new Set(['break', 'case', 'catch', 'class', 'const', 'continue', 'debugger', 'default', 'delete', 'do', 'else', 'export', 'extends', 'finally', 'for', 'function', 'if', 'import', 'in', 'instanceof', 'new', 'return', 'super', 'switch', 'this', 'throw', 'try', 'typeof', 'var', 'void', 'while', 'with', 'yield', 'let', 'static', 'enum', 'await', 'implements', 'interface', 'package', 'private', 'protected', 'public']); return /^[a-zA-Z_$][a-zA-Z0-9_$]*$/.test(str) && !reservedWords.has(str);

Suggested change

	// Check if it's a valid JavaScript identifier
	// Must start with letter, underscore, or dollar sign
	// Can contain letters, digits, underscores, or dollar signs
	return /^[a-zA-Z_$][a-zA-Z0-9_$]*$/.test(str);
	// Check if it's a valid JavaScript identifier and not a reserved word
	// Must start with letter, underscore, or dollar sign
	// Can contain letters, digits, underscores, or dollar signs
	const reservedWords = new Set([
	'break', 'case', 'catch', 'class', 'const', 'continue', 'debugger', 'default', 'delete', 'do', 'else',
	'export', 'extends', 'finally', 'for', 'function', 'if', 'import', 'in', 'instanceof', 'new', 'return',
	'super', 'switch', 'this', 'throw', 'try', 'typeof', 'var', 'void', 'while', 'with', 'yield', 'let',
	'static', 'enum', 'await', 'implements', 'interface', 'package', 'private', 'protected', 'public'
	]);
	return /^[a-zA-Z_$][a-zA-Z0-9_$]*$/.test(str) && !reservedWords.has(str);