Templating guide

Following gives you the idea of how the templating works. This is necessary if you want to modify or replace the templates.

Template engines

Job of applying extracted data to templates and generating final documentation is dedicated to template engines. This gives the flexibility of generating a wide range of output types.

The template engine interface defined by this package is dictated by com.mohamnag.doclavax.TemplateEngine interface and has absolutely minimalistic implications. This means anybody can potentially plug his own implementation into the process (this plugging is not yet in place see roadmap).

Each template engine is also passed some parameters like template input and output directories (not yet in place, see roadmap).

Handlebars

Currently there only exists one engine implementation which is based on Handlebars templates using the handlebars.java engine.

You can see the guides on how to modify handlebars templates.

Templating concepts

There are three categories of pages that should be handled by templating engines:

• Class pages for each root class (in contrast to inner classes)
• Package pages for each package
• Meta pages, can contain any kind of additional data

The templating engine will handle the first two types specially, each class template will be rendered multiple times, each time with the context of a dedicated class. Similarly a package template will be rendered multiple times each time with context of a single package. While compiling each class or package page, additional info is also provided in the context which contains at least the full package and root class list. However each template engine may add more data into these meta information.

All additional templates fall into the category of meta pages and are rendered ONLY once with context of full list of packages and root classes.