import { Rule } from './rule-interface';
/**
* MD051: Link fragments should be valid
*
* This rule is triggered when a link fragment doesn't match any of the
* anchor IDs generated from the headings in the document.
*
* For this rule, our fix will strip invalid fragments from links when they can't be resolved.
*/
export const name = 'MD051';
export const description = 'Link fragments should be valid';
/**
* Fix invalid link fragments by removing them
* @param lines Array of string lines to fix
* @returns Fixed lines array with valid link fragments
*/
export function fix(lines: string[]): string[] {
// First, extract all headings to get valid fragment IDs
const headingIds = extractHeadingIds(lines);
// Then fix links with invalid fragments
return lines.map(line => {
// Match links with fragments: [text](#fragment)
return line.replace(/\[([^\]]+)\]\(([^)]+)(#[^)]+)\)/g, (match, text, url, fragment) => {
const fragmentId = fragment.substring(1); // Remove the # character
// Check if the fragment is valid
if (headingIds.includes(fragmentId)) {
return match; // Fragment is valid, keep as is
}
// Fragment is invalid, remove it
return `[${text}](${url})`;
});
});
}
/**
* Extract heading IDs from the document
* These are usually lowercase versions of the heading text with spaces replaced by hyphens
*/
function extractHeadingIds(lines: string[]): string[] {
const headingIds: string[] = [];
// Regular expression to match ATX headings (# Heading)
const atxHeadingRegex = /^#{1,6}\s+(.+?)(?:\s+#{1,6})?$/;
for (const line of lines) {
const match = line.match(atxHeadingRegex);
if (match) {
const headingText = match[1].trim();
const headingId = headingText
.toLowerCase()
.replace(/[^\w\s-]/g, '') // Remove special characters
.replace(/\s+/g, '-'); // Replace spaces with hyphens
headingIds.push(headingId);
}
}
return headingIds;
}
/**
* Rule implementation for MD051
*/
export const rule: Rule = {
name,
description,
fix
};
export default rule;
