Contentful Hugo
This is a simple Node.js CLI tool that pulls data from Contentful CMS and turns it into Markdown or YAML files for use with a static site generator. It can be used with any static site generator that uses Markdown with YAML frontmatter, but it has some features that are specific to Hugo.
Table of Contents
Prerequisites
Install Node.js
Installation
with NPM
npm install contentful-hugo
with Yarn
yarn add contentful-hugo
Usage
Terminal Commands
Complete configuration then run the following command(s) in the terminal
When Installed Globally
contentful-hugo
When Installed Locally
npx contentful-hugo
Example Package.json
{
"name": "my-hugo-project",
"scripts": {
"prestart": "contentful-hugo",
"start": "hugo server",
"prebuild": "contentful-hugo",
"build": "hugo --minify"
}
}
In this example when you run npm start
it will first use contentful-hugo to pull Contentful data then start hugo server. In the same way when you do the command npm run build
it will first use contentful-hugo to pull Contentful data then run hugo --minify
to build a minified version of your hugo site.
Error Messages
Trying to use this package before completing configuration will return an error in the console
Configuration
Environment Variables
Before using you must first set the following environment variables. CONTENTFUL_SPACE, and CONTENTFUL_TOKEN.
This can be done with a .env file in the root directory of your project.
CONTENTFUL_SPACE = '<your-space-id>`
CONTENTFUL_TOKEN = '<content-api-access-token>`
You can also declare the environment variables in the command line
Powershell:
$env:CONTENTFUL_SPACE="<contentful_space_id>"
$env:CONTENTFUL_TOKEN="<contentful_acessToken>"
Bash:
export CONTENTFUL_SPACE=<contentful_space_id>
export CONTENTFUL_TOKEN=<contentful_acessToken>
Config File
In order to pull the data you want you will need to create a contentful-settings.yaml file in the root of your repository.
Example contentful-settings.yaml file (see below for complete configuration options)
singleTypes:
- id: homepage
directory: /content/
fileName: _index
fileExtension: md
- id: siteSettings
directory: /data/
fileName: settings
fileExtension: yaml
repeatableTypes:
- id: posts
directory: /content/posts/
fileExtension: md
mainContent: content
- id: seoFields
isHeadless: true
directory: /content/seo-fields/
- id: reviews
directory: /content/reviews/
mainContent: reviewBody
- id: staff
isHeadless: true
directory: /content/staff/
Configuration Options
field | required | description |
---|
id | required | contentful content type ID goes here |
directory | required | directory where you want the file(s) to be generated (leading and trailing slashes required for the time being) |
fileName | required (single types only) | name of the file generated |
fileExtension | optional | can be "md", "yml", or "yaml" (defaults to "md") |
isHeadless | optional (repeatable types only) | turns all entries in a content type into headless leaf bundles (see hugo docs) |
mainContent | optional | field ID for field you want to be the main Markdown content. (Does not work with rich text fields) |
Expected Output
Files will be generated in the directory specified in the contentful-settings.yaml file. Front matter will be in YAML format. Files of single types will be named after fileName specified in the config file. Files of repeatable types will be named after their entry ID in Contenful, which makes it easy to link files together.
Default Date and Time Fields
The following fields will always appear in your frontmatter:
updated:
createdAt:
date:
Asset Information
Assets like images and videos come with some extra information that makes it easy to implement things like alt text or layouts that rely on knowing the image dimensions. The fields are as follows:
assetFieldName:
assetType:
url:
title:
description:
width:
height:
If you're using Hugo you can access the information like below:
<img
src="{{ .Params.assetFieldName.url }}"
width="{{ .Params.assetFieldName.width }}"
/>
This same information will also appear in asset arrays like a gallery:
myGallery:
- assetType: 'image/jpg'
url: '//link-to-image.jpg'
title: 'Image 1'
description: 'Image 1 Description'
width: 500
height: 500
- assetType: 'image/jpg'
url: '//link-to-image-2.jpg'
title: 'Image 2'
description: 'Image 2 Description'
width: 1920
height: 1080
Entries
Linked entries will include fields for it's id and it's content type id.
linkedEntry:
id: <contentful-entry-id>
typeId: <content-type-ID>
relatedArticles:
- id: '41UFfIhszbS1kh95bomMj7'
typeId: 'articles'
- id: '85UFfIhsacS1kh71bpqMj7'
typeId: 'articles'
All files are named after their entry id in Contentful making it easy to retrieve it using .Site.GetPage in Hugo
{{ with .Site.GetPage "<path-to-file>/<entry-id>.md" }}
{{ .Title }}
{{ end }}
Rich Text Fields
A Rich text field will produce nested arrays mirroring the JSON structure that they have in the API. Each node will need to be looped through and produce HTML depending on the nodeType field.
richTextField:
- nodeType: 'paragraph'
data: {}
content:
- data: {}
marks: []
value: 'This is a simple paragraph.'
nodeType: 'text'
- nodeType: 'paragraph'
data: {}
content:
- data: {}
marks: []
value: 'This is a paragraph with '
nodeType: 'text'
- data: {}
marks:
- type: 'italic'
value: 'italicized text.'
nodeType: 'text'
- nodeType: 'embedded-asset-block'
data:
assetType: 'image/jpeg'
url: '//images.ctfassets.net/some-image-url.jpg'
title: 'Image title will appear here'
description: 'Image description will appear here'
width: 1920
height: 1080
content: []
In addition a plaintext version of the field will be generated using the field ID appended with "_plaintext". This allows you to quickly fetch the text by itself without any of the other data. A simple use case would be using the plaintext output to automatically generate a meta description for a webpage.
richTextField_plaintext: 'This is a simple paragraph. This is a paragraph with italicized text.'
Known Issues
These are some known issues.
- Date & Time Field w/o Timezone: Date fields that include time but do not have a specified timezone will have a timezone set based on whatever machine the script is run on. So using a date field in contentful with this setting could lead to unexpected results when formatting dates. Date fields that don't include time (ex: YYYY-MM-DD) are not effected by this.
- Fetching Data Before Contentful CDN Updates: Sometimes when triggering a build from a webhook, it won't always get the latest data. This is because it sometimes takes a couple seconds for the latest data to get distrubuted across Contentful's CDN. If you run into this issue it might be worth it to create a "wait function" just to delay fetching the data by a couple seconds. You could include it in the script you use contentful-hugo by doing something like the following
"node wait.js && contentful-hugo"