Quick and dirty markdown link checker

markdown-link-checker.php

<?php


function get_markdown_files($dir = '.') {

  $directory = new RecursiveDirectoryIterator($dir);
  $recursive = new RecursiveIteratorIterator($directory);
  $regex = new RegexIterator($recursive, '/^.+\.md/i', RecursiveRegexIterator::GET_MATCH);

  $files = array();

  foreach ($regex as $file) {

    if (!empty($file[0])) {

      $files[] = $file[0];
    }
  }

  return $files;
}

function get_md_links_from_file($file) {

  $links = array();

  $contents = file_get_contents($file);
  if (preg_match_all('/\[(.+?)\].*?\((.+?)\)/', $contents, $matches, PREG_SET_ORDER)) {

    $links = array_merge($links, $matches);
  }

  return $links;
}

$files = get_markdown_files();
foreach ($files as $file) {

  $links = get_md_links_from_file($file);

  foreach ($links as $link) {

    $path = $link[2];
    if (preg_match('/^(http|#)/', $path)) {

      continue;
    }

    $path = current(explode('#', $path, 2));

    $start = rtrim(__DIR__, '/');
    if ($path[0] !== '/') {

      $start .= '/' . rtrim(dirname($file), '/');
    }

    $realpath = realpath($start . '/' . $path);

    if (empty($realpath) || (!file_exists($realpath) && !is_dir($realpath))) {

      printf("%s: Possible bad link for \"%s\" to \"%s\"\n", $file, $link[1], $link[2]);
    }
  }
}