Javadoc Output Example

Javadoc Output Example

A real example of what bxSitesJavadocDoc (Gradle) / bxsites:javadoc (Maven) actually produces - see Gradle Plugin or Maven Plugin for how to turn this on in your own project.

Point it at your own project's .java sources and it walks every public top-level type, emitting one page per type, its constructors/methods/fields wrapped in a small Alpine.js-driven filter toolbar - a search box plus one chip per kind actually present - so a long class stays easy to scan. Given this class:

package com.example;

/**
 * A single book in the shelf, immutable once created. Two books are
 * considered equal only by reference, not by content.
 */
public class Book {

    /** The longest title this shelf will accept. */
    public static final int MAX_TITLE_LENGTH = 200;

    /**
     * Creates a book with the given title and author.
     *
     * @param title  the book's title
     * @param author the book's author
     */
    public Book(String title, String author) {
        this.title = title;
        this.author = author;
    }

    /**
     * Returns a copy of this book with a new title.
     *
     * @param newTitle the new title
     * @return a copy of this book with the given title
     */
    public Book withTitle(String newTitle) {
        return new Book(newTitle, author);
    }
}

...the generator writes api/javadoc/com/example/Book.md with this frontmatter:

---
title: "Book"
summary: "A single book in the shelf, immutable once created."
tags: [api, javadoc]
---

And this body, reproduced below not as a code block but rendered for real - this is the generated page's actual body, live, including the filter toolbar. Try the Fields chip, or type "title" into the search box:

Book

com.example.Book

A single book in the shelf, immutable once created. Two books are considered equal only by reference, not by content.

Notice the class's first sentence became the frontmatter summary, while the page body carries the full doc comment (both sentences) - that's real DocCommentTree first-sentence extraction, not a simplification made for this example. Notice too that the toolbar/chips/search box above are real, working Alpine.js - not a screenshot - built entirely from plain HTML attributes and a small inline stylesheet driven by this theme's own --bxsites-* variables, so it looks right in whichever theme you're reading this in without bx-sites itself needing any changes.

What this doesn't cover

Deliberately scoped down for v1, not a complete Javadoc-to-Markdown converter - this is the heaviest of the three Spring Boot generators. Nested and package-private types are skipped entirely; only each member's own doc comment is used, never an inherited one; inline HTML in doc comments is stripped rather than converted to Markdown; {@link}/{@see} render as inline code with no cross-page hyperlink resolution; no index/nav page is generated. A record also picks up its compiler-generated accessors/toString/equals/hashCode, matching the standard javadoc tool's own behavior - Book above is a plain class so this example stays focused on the common case. See the Gradle/Maven guides linked above for the full list.