Antora Xref Extension

The Antora Xref Extension augments Antora to improve xref macro handling.

Warning

This is ALPHA software! That means the extension is experimental and likely to change at any time without notice. You’re welcome to test it and give feedback, but take caution when relying on it in a production site.

Overview

The Antora Xref Extension augments Antora’s support for xref macros with the following features:

Validate that the target file contains an anchor that matches the fragment.
Use default link text from the title or reftext of the target.
Stub links that don’t yet exist
Produce warnings if link text is unnecessary

Note	This extension will slightly increase build times since `.adoc` files are parsed twice.

Prerequisites

In order to use this extension, you must be using Node.js 16 and Antora 3.1.0. The following instructions assume you’ve already set up an Antora playbook file (i.e., antora-playbook.yml) to build your site.

To learn about using extensions with Antora, see the Antora extension documentation.

Install

Use the following command to install the extension into your playbook project:

$ npm i @springio/antora-xref-extension

Register

Open your Antora playbook file and register the extension as an entry in the antora.extensions key. If this key doesn’t yet exist in your playbook, first create it.

antora-playbook.yml

antora:
  extensions:
  - require: '@springio/antora-xref-extension'

Configure

By default no configuration is required to use the extension. Once the extension has been registered macros of the form xref:version@component:module:file.adoc#fragment[] will:

Be validated to ensure the target file contains an anchor that matches the fragment.
Use default link text from the title or reftext of the target.

Additional configuration is only required for the additional features listed below.

Stubbed References

In some situations you may want to build an Antora content source that includes reference to sources that won’t be built until later.

For example, you may publish a complete docs site that aggregates several sources, but also want to be able to build those individual sources independently in order to catch errors early. In these situations it can be useful to defer errors and use stubbed links so that an individual build can run.

In order to stub xrefs you can provide regular expression patterns in the stub extension configuration. Any xref that matches the pattern will be replaced replaced with a link to the source document that defines it.

Note	If the xref doesn’t include link text then the full reference will be used as a placeholder.

The following example shows how any xref patterns that reference missingmodule can be stubbed:

antora-playbook.yml

antora:
  extensions:
  - require: '@springio/antora-xref-extension'
    stub:
    - 'missingmodule:.*'

Xref Text Warnings

If you have an existing document that uses fragment based xrefs, it’s likely that the link text will also be specified. This is necessary with a stock Antora installation because without the link text the URL of the published target page is displayed.

Since this extension can deduce the text to use for fragment links, you may want to simplify your markup.

For example, with the following files:

welcome.adoc

= Welcome

Hello

[[getting-started]]
== Getting Started
...

details.adoc

= Details
For details see xref:welcome.adoc#getting-started[Getting Started] for details.

The xref:welcome.adoc#getting-started[Getting Started] link can be simplified to xref:welcome.adoc#getting-started[].

To help catch links that can be simplified, you can configure the extension to log warnings when it finds unnecessary link text:

antora-playbook.yml

antora:
  extensions:
  - require: '@springio/antora-xref-extension'
    log-unnecessary-link-text-warnings: true

philwebb / antora-xref-extension Goto Github PK