ktg-plugin-marketplace/plugins/config-audit/scanners/lib/yaml-parser.mjs

/**
 * Regex-based YAML frontmatter parser for Claude Code .md files.
 * Handles YAML frontmatter (--- delimited) and basic YAML parsing.
 * Zero external dependencies.
 */

/**
 * Parse YAML frontmatter from markdown content.
 * @param {string} content
 * @returns {{ frontmatter: object | null, body: string, bodyStartLine: number }}
 */
export function parseFrontmatter(content) {
  const match = content.match(/^---\r?\n([\s\S]*?)(?:\r?\n)?---(?:\r?\n|$)/);
  if (!match) {
    return { frontmatter: null, body: content, bodyStartLine: 1 };
  }

  const raw = match[1];
  const bodyStartLine = raw.split('\n').length + 3; // 2 for --- lines + 1-based
  const body = content.slice(match[0].length);
  const frontmatter = parseSimpleYaml(raw);

  return { frontmatter, body, bodyStartLine };
}

/**
 * Parse simple YAML key-value pairs (no nesting beyond arrays).
 * @param {string} yaml
 * @returns {object}
 */
export function parseSimpleYaml(yaml) {
  const result = {};
  const lines = yaml.split('\n');
  let currentKey = null;
  let multiLineValue = '';
  let inMultiLine = false;

  for (const line of lines) {
    // Skip comments and empty lines
    if (line.trim().startsWith('#') || line.trim() === '') {
      if (inMultiLine) multiLineValue += '\n';
      continue;
    }

    // Key-value pair
    const kvMatch = line.match(/^(\w[\w-]*):\s*(.*)/);
    if (kvMatch && !inMultiLine) {
      if (currentKey && multiLineValue) {
        result[normalizeKey(currentKey)] = multiLineValue.trim();
      }

      currentKey = kvMatch[1];
      const value = kvMatch[2].trim();

      if (value === '|' || value === '>') {
        inMultiLine = true;
        multiLineValue = '';
        continue;
      }

      result[normalizeKey(currentKey)] = parseValue(value);
      currentKey = null;
      continue;
    }

    // Multi-line continuation
    if (inMultiLine) {
      if (line.match(/^\s+/)) {
        multiLineValue += (multiLineValue ? '\n' : '') + line.trim();
      } else {
        result[normalizeKey(currentKey)] = multiLineValue.trim();
        inMultiLine = false;
        multiLineValue = '';
        // Re-process this line as a new key
        const reMatch = line.match(/^(\w[\w-]*):\s*(.*)/);
        if (reMatch) {
          currentKey = reMatch[1];
          result[normalizeKey(currentKey)] = parseValue(reMatch[2].trim());
          currentKey = null;
        }
      }
    }
  }

  // Flush remaining multi-line
  if (inMultiLine && currentKey) {
    result[normalizeKey(currentKey)] = multiLineValue.trim();
  }

  // Normalize arrays for known list fields
  for (const field of ['allowed_tools', 'tools', 'paths', 'globs']) {
    if (typeof result[field] === 'string') {
      result[field] = result[field].split(',').map(s => s.trim()).filter(Boolean);
    }
  }

  return result;
}

/**
 * Parse a YAML value string.
 */
function parseValue(str) {
  if (str === '' || str === '~' || str === 'null') return null;
  if (str === 'true') return true;
  if (str === 'false') return false;
  if (/^\d+$/.test(str)) return parseInt(str, 10);
  if (/^\d+\.\d+$/.test(str)) return parseFloat(str);

  // Inline array: [a, b, c]
  if (str.startsWith('[') && str.endsWith(']')) {
    return str.slice(1, -1).split(',').map(s => {
      const v = s.trim();
      return v.replace(/^["']|["']$/g, '');
    }).filter(Boolean);
  }

  // Quoted string
  if ((str.startsWith('"') && str.endsWith('"')) || (str.startsWith("'") && str.endsWith("'"))) {
    return str.slice(1, -1);
  }

  return str;
}

/**
 * Normalize key: hyphens to underscores.
 */
function normalizeKey(key) {
  return key.replace(/-/g, '_');
}

/**
 * Parse a JSON file content. Returns null on error.
 * @param {string} content
 * @returns {object | null}
 */
export function parseJson(content) {
  try {
    return JSON.parse(content);
  } catch {
    return null;
  }
}

/**
 * Find @import references in CLAUDE.md content.
 * @param {string} content
 * @returns {{ path: string, line: number }[]}
 */
export function findImports(content) {
  const imports = [];
  const lines = content.split('\n');
  for (let i = 0; i < lines.length; i++) {
    const match = lines[i].match(/^@(.+)$/);
    if (match) {
      imports.push({ path: match[1].trim(), line: i + 1 });
    }
  }
  return imports;
}

/**
 * Extract markdown sections (## headings) from content.
 * @param {string} content
 * @returns {{ heading: string, level: number, line: number }[]}
 */
export function extractSections(content) {
  const sections = [];
  const lines = content.split('\n');
  for (let i = 0; i < lines.length; i++) {
    const match = lines[i].match(/^(#{1,6})\s+(.+)/);
    if (match) {
      sections.push({
        heading: match[2].trim(),
        level: match[1].length,
        line: i + 1,
      });
    }
  }
  return sections;
}