WordPress で書かれた記事を Markdown 変換する（物理）

convert_html_to_markdown.php

<?php

$dir = './content/archives/';
$filenames = [];
if ($handle = opendir($dir)) {
  while (false !== ($filename = readdir($handle))) {
    if ($filename != "." && $filename != ".." && strtolower(substr($filename, strrpos($filename, '.') + 1)) == 'md') {
        $filenames[] = $filename;
    }
  }
  closedir($handle);
}

foreach($filenames as $index => $filename) {
  echo $filename . "\n";

  $path = $dir . $filename;
  $contents = file_get_contents($dir . $filename);

  $pattern = '/<div class="amazlet-box" style="margin-bottom:0px;"><div class="amazlet-image" style="float:left;margin:0px 12px 1px 0px;"><a href="http:\/\/www\.amazon\.co\.jp\/exec\/obidos\/ASIN\/(.+?)\/rakuishi-22\/ref=nosim\/" name="amazletlink" target="_blank"><img src="(.+?)" alt="(.+?)" style="border: none;" \/><\/a><\/div><div class="amazlet-info" style="line-height:120%; margin-bottom: 10px"><div class="amazlet-name" style="margin-bottom:10px;line-height:120%"><a href="http:\/\/www.amazon.co.jp\/exec\/obidos\/ASIN\/.+?\/rakuishi-22\/ref=nosim\/" name="amazletlink" target="_blank">.+?<\/a><div class="amazlet-powered-date" style="font-size:80%;margin-top:5px;line-height:120%">posted with <a.+?>amazlet<\/a>.+?<\/div><\/div><div class="amazlet-detail">.+?<br \/><\/div><div class="amazlet-sub-info" style="float: left;"><div class="amazlet-link" style="margin-top: 5px"><a href="http:\/\/www.amazon.co.jp\/exec\/obidos\/ASIN\/.+?\/rakuishi-22\/ref=nosim\/" name="amazletlink" target="_blank">Amazon.co.jpで詳細を見る<\/a><\/div><\/div><\/div><div class="amazlet-footer" style="clear: left"><\/div><\/div>/i';
  $replacement = '{{<amazon id="$1" title="$3" src="$2">}}';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<a href="https:\/\/itunes\.apple\.com\/jp\/app\/id(\d+)\?mt=.+?" target="_blank" rel="nofollow"><img width="100" class="alignleft.*?" align="left" src="(.+?)".*?><\/a><strong>.*?<a.*?>(.+?)<\/a><\/strong>.*?<br style="clear: both;"><br \/>/i';
  $replacement = '{{<app id="$1" title="$3" src="$2">}}';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<a href="https:\/\/itunes\.apple\.com\/jp\/app\/id(\d+)\?mt=.+?" target="_blank" rel="nofollow"><img width="100" class="alignleft.*?" align="left" src="(.+?)".*?><\/a><strong>.*?(.+?)<\/strong>.*?<br style="clear: both;"><br \/>/i';
  $replacement = '{{<app id="$1" title="$3" src="$2">}}';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<a href="https:\/\/itunes.apple.com\/jp\/app\/id(\d+)\?mt=.+?" target="_blank" rel="nofollow"><img width="100" class="alignleft.*?" align="left" src="(.+?)".*?><\/a><strong>.*?<a.*?>(.+?)<\/a><\/strong>.*?<br style="clear: both;">/i';
  $replacement = '{{<app id="$1" title="$3" src="$2">}}';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<a href="https:\/\/itunes.apple.com\/jp\/app\/id(\d+)\?mt=.+?" target="_blank" rel="nofollow"><img width="100" class="alignleft.*?" align="left" src="(.+?)".*?><\/a><strong>.*?(.+?)<\/strong>.*?<br style="clear: both;">/i';
  $replacement = '{{<app id="$1" title="$3" src="$2">}}';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<h2>(.+?)<\/h2>/';
  $replacement = '## $1';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<h3>(.+?)<\/h3>/';
  $replacement = '## $1';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<a.*?href="(.+?)".*?>(.+?)<\/a>/';
  $replacement = '[$2]($1)';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<img .*?src="(.+?)".*?>/';
  $replacement = '![]($1)';
  $contents = preg_replace($pattern, $replacement, $contents);

  $pattern = '/<pre><code>(.+?)<\/code><\/pre>/is';
  preg_match_all($pattern, $contents, $matches);
  if (count($matches) && count($matches[0])) {
    for ($i = 0; $i < count($matches[0]); $i++) {
      $replacement = $matches[1][$i];
      $replacement = str_replace('&lt;', '<', $replacement);
      $replacement = str_replace('&gt;', '>', $replacement);
      $replacement = str_replace('&quot;', '"', $replacement);
      $replacement = '```' . "\n" . $replacement . "\n" . '```';
      $replacement = str_replace("\n\n```", "\n```", $replacement);
      $contents = str_replace($matches[0][$i], $replacement, $contents);
    }
  }

  $pattern = '/<ul>(.+?)<\/ul>/is';
  preg_match_all($pattern, $contents, $matches);
  if (count($matches) && count($matches[0])) {
    for ($i = 0; $i < count($matches[0]); $i++) {
      $replacement = $matches[1][$i];
      $replacement = str_replace(array('<ul>', '</ul>', '</li>'), '', $replacement);
      $replacement = str_replace('<li>', '* ', $replacement);
      $contents = str_replace($matches[0][$i], $replacement, $contents);
    }
  }

  $pattern = '/<ol>(.+?)<\/ol>/is';
  preg_match_all($pattern, $contents, $matches);
  if (count($matches) && count($matches[0])) {
    for ($i = 0; $i < count($matches[0]); $i++) {
      $replacement = $matches[1][$i];
      $replacement = str_replace(array('<ol>', '</ol>', '</li>'), '', $replacement);
      $replacement = str_replace('<li>', '1. ', $replacement);
      $contents = str_replace($matches[0][$i], $replacement, $contents);
    }
  }

  $pattern = '/<code>(.+?)<\/code>/';
  $replacement = '`$1`';
  $contents = preg_replace($pattern, $replacement, $contents);

  $contents = str_replace(' title=" ', ' title="', $contents);
  $contents = str_replace('<br style="clear:both;">', '', $contents);

  file_put_contents($path, $contents);
}