Hadooken.

Hadooken.

Introduction

While migrating zerotohero.dev to its new design, I needed to find a quick and dirty way to redirect incoming requests to certain websites to their corresponding pages on Zero to Hero.

This nugget gives a neat trick to achieve this with the least amount of effort.

<meta> Refresh to the Rescue

Since all of those websites were static sites backed by an S3 bucket, updating the error page (404.html) and the index page (index.html) was the quickest way to achieve this:

<!DOCTYPE html>
<!--
404.html and index.html are the same.
S3 bucket is configured to serve 404.html for errors,
and index.html is configured to be the default page.
-->
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" 
   content="width=device-width, initial-scale=1.0">
  
  <title>Redirecting...</title>
  
  <meta http-equiv="refresh" content="0; 
    url=https://zerotohero.dev/about/volkan/">
  <link rel="canonical" href="https://zerotohero.dev/about/volkan/">
  <meta name="robots" content="noindex, follow">
  
  <script type="text/javascript">
    setTimeout(function() {
      window.location.href = "https://zerotohero.dev/about/volkan/";
    }, 0);
  </script>
</head>
<body>
<p>If you are not redirected automatically,
  follow this link:
  <a href="https://zerotohero.dev/about/volkan/"
    >https://zerotohero.dev/about/volkan/</a>.</p>
</body>
</html>

This way, I won’t have to worry about configuring the S3 bucket to redirect incoming requests to the new URL. And as a bonus, I can still keep certain (hidden) pages on the old website without having to worry about them being indexed by search engines.

In a Nutshell

Here’s a breakdown of the code:

  • The meta tag with the http-equiv attribute is used to refresh the page after 0 seconds to the new URL.
  • The script tag is used as a fallback in case the meta tag is not supported by the browser (very unlikely).
    • The setTimeout function is used to redirect the user to the new URL after 0 seconds (i.e. immediately).

For search engines:

  • With <link rel="canonical" ... />, we are telling the search engines that the alternative url is the preferred one.
  • The <meta name="robots" content="noindex, follow"> tag is used to tell the search engines not to index the page but to follow the links on the page.
  • Which is good enough for search engines to understand that the page has been moved to a new location.

How About 3xx Redirects?

The only thing I am not doing for the search engines is to set the 301 status code in the HTTP response.

That is doable with some CloudFront and AWS Lambda at Edge trickery, but I feel it’s an overkill.

Why?

  • Firstly, search engines are a thing of the past.
  • And, secondly, we already inform the search engines about the change anyway.

Conclusion

Hope you liked this little nugget. It’s the fastest to way to achieve the task without having to worry about the underlying infrastructure.

Until next time… May the source be with you 🦄.


Section Contents