An alternative way of loading content in Astro


An alternative way of using the fetchContent loading in Astro

16 Mar, 2022 ยท 4 min read

Astro has a super cool function called fetchContent it can be used to fetch content from your pages.

A little example, if I want to retrieve all my posts in Astro I simply use the following command:

const allPosts = Astro.fetchContent('../pages/posts/*.md');

Which will return an object like this:

    // Fronmatter parts for example:
    "title": "The title",
    "date": "2022-02-22",
    "astro": {
      "headers": [],
      "source": "",
      "html": ""
    "url": ""

As you can see, pretty slick way to retrieve all our content with all the data quickly.

And this method is perfect since it returns everything we need.

The downside and a solution

I wanted to render related posts on each article, which sounded like a great idea. I started by using fetchContent, but quickly came across a downside to using this function.

It could cause infinite loops. This is because it technically renders the HTML, so if you refer to other posts (like recommended posts), it will cause an endless loop in your script.

The solution is to use Astroโ€™s underlying method, limiting the response.

Officially they used globEager, but I used glob from the start. Not sure what the difference is between the two at the time of writing.

To load all the posts as before we can use the following code:

const allPosts = await import.meta.glob('../pages/posts/*.md');

However, instead of an excellent formatted array, this will return promise based array like this:

	  '../pages/posts/': [Function: ../pages/posts/],
	  '../pages/posts/': [Function: ../pages/posts/],

The first part (the key) is the file name, and the value is lazy loaded import (powered by Vite).

If you need to access the front matter part of your posts, weโ€™ll have to loop over them, as until now, all we have is the filename.

Iโ€™m going to create another array called mappedPosts in which weโ€™ll loop over all the keys, which will look like this:

const mappedPosts = await Promise.all(
  Object.keys(allPosts).map((key) => {
    // Todo

Since this is a lazy promise, we need to retrieve the function and do wait for the return.

But first, Iโ€™ll store the current post and convert the file name into a slug.

const mappedPosts = await Promise.all(
  Object.keys(allPosts).map((key) => {
    const post = allPosts[key];
    const url = key.replace('../pages/', '/').replace('.md', '/');

We simply replace our prefix, and the extension, this will result in the following examples:

// ../pages/posts/
// /posts/why-tailwind-jit-compiler-is-amazing

The last thing we want to do is return the promised result and format the necessary elements.

Remember from the intro, which elements are available? You can use any of those. I choose only to return the front matter part and the url.

const mappedPosts = await Promise.all(
  Object.keys(allPosts).map((key) => {
    const post = allPosts[key];
    const url = key.replace('../pages/', '/').replace('.md', '/');
    return post().then((p) => {
      return { ...p.frontmatter, url };

And now we have a neat array of posts with frontmatter parts that donโ€™t cause an infinite loop.

You can now do all the filtering you want on this array.

This way is a bit more work, but it does make things more dynamic for you.

Thank you for reading, and letโ€™s connect!

Thank you for reading my blog. Feel free to subscribe to my email newsletter and connect on Facebook or Twitter

Spread the knowledge with fellow developers on Twitter
Tweet this tip
Powered by Webmentions - Learn more

Read next ๐Ÿ“–

Computed Nano Stores

6 Sep, 2022 ยท 2 min read

Computed Nano Stores

Astro Nano Stores maps

5 Sep, 2022 ยท 3 min read

Astro Nano Stores maps

Join 2099 devs and subscribe to my newsletter