Contentful Hugo Extractor

This tool extracts all content from your Contentful space and makes it easily consumable by Hugo. You can run it locally or as part of a CI server like Travis.

Install

Go Install Method

Assuming Go (1.10 +) is installed as well as dep

go get -u "github.com/friends-of-hugo/contentful-export"
cd "$GOPATH/src/github.com/friends-of-hugo/contentful-export"
dep ensure
go install

Usage

contentful-export [Flags]

Flags:
 -space-id=value      "Id of the contentful space from which to extract content. If not present will default to an environment variable named `$CONTENTFUL_API_SPACE`"
 -api-key=value       "API Key used to authenticate with contentful for content delivery. If not present will default to an environment variable named `$CONTENTFUL_API_KEY`. The preview API key should be provided if -p is used."
 -config-file=value   "Path to the config TOML file to load. Defauls to `./extract-config.tml`"
 -p                   "If present, the contentful preview API will be used so that draft content will be included as part of the export."

The tool requires two parameters to work, a contentful space id and API key. These can be provided as command line flags or as environment variables

As environment vars...

export CONTENTFUL_API_KEY=YOUR-CONTENT-DELIVERY-API-ACCESS-TOKEN-HERE
export CONTENTFUL_API_SPACE=YOUR-SPACE-ID-HERE

contentful-export

As flags...

contentful-export -space-id=[YOUR-ID-HERE] -api-key=[YOUR-ACCESS-KEY-HERE] -config-file="./export-conf.toml"

Expected output

Contentful Hugo Extractor stores all content under the /content directory. For each content type, it makes a subdirectory. For each item, it creates a markdown file with the all properties in TOML format.

Special cases:

• Items of type Homepage are stored as /content/_index
  • Note that there can only be one such item
• Fields named mainContent are used for the main content of the markdown file
• File names are based on the ID of the item to make it easily referencable from related items (for the machine, not humans)

Configuration

Use the --config-file command line flag to provide the location of a TOML configuration to laod or ensure that there is a extract-config.toml file in the work directory of contentful-hugo

Configure YAML output

While the default output is in TOML format, it is also possible to output content in YAML format. Use the following key in your config file:

encoding = "yaml"

Configure Hugo Page Bundles

contentful-export will export each content type in contentful into its own content directory ./content/ and, since hugo treats each rootlevel content directory as a Section, you will end up having a hugo section for each contentful content type. Hugo allows you to provide Section level configuration for its Page Bundles by dropping a file named _index.md in the section's content directory. It is likely that you'll want to provide such configuration for some sections.

For example, let's say you need to make a section headless. Pretend that you have a contentful content type with the id question and you have some questions in your contentful content model which you intend to reference in a seperate FAQ page. After a contentful-hugo export, you might the following directory structure:

./
|   content
|   |   _index.md
|   |   question
|   |   |   12h3jk213n.md   //question 1
|   |   |   sdfer343sn.md   //question 2
|   |   page
|   |   |   sdf234dd32.md   //FAQ page - refs questions in its frontmatter
|   layouts
|   |   _default
|   |   |   single.html
|   |   page
|   |   |   single.html //question refs are loaded via .Site.GetPage

Without any further confuguration, hugo would generate a HTML file for the page using the ./layouts/page/single.html layout template but it would aslo generate HTML files for the questions using the ./layouts/_default/single.html layout template. To prevent this from happening you would create the following file under the path ./content/question/index.md:

+++
headless = true
+++

If you need this kind of configuration, the contentful-hugo export process can generate this index.md file for you. Simply provide the TOML to use in your config file:

encoding = "toml"
[section]
[section.question]
 headless = true

You can nest as many tables as you need under the [sections] and if the nested table name matches a contentful content type id than the configuration provided will be propagated to the section's index.md frontmatter.