[arturobernalg]

Fixes #807.

HtmlToPlainTextConverter.FormattingVisitor appended "\n * " for every <li> regardless of nesting depth, and its cleanup step collated any run of spaces -- including indentation meant to show nesting -- down to one. A nested <ul>/<li> structure in a Mojo's or parameter's javadoc always rendered as one flat bullet level in the plain-text output used by mvn <plugin>:help, losing the nesting the javadoc author intended.

FormattingVisitor now tracks <ul>/<ol> nesting depth and indents each <li> accordingly, using a placeholder character for the indentation spaces so the existing space-collation cleanup doesn't erase it -- the placeholder is only expanded to real spaces as the last step of toString(). Added tests for a single-level list (unchanged output), a nested list, and a doubly-nested list. Full module suite, checkstyle and spotless pass locally.

To make clear that you license your contribution under the Apache License Version 2.0, January 2004, check one of the following:

• I hereby declare this contribution to be licensed under the Apache License Version 2.0, January 2004
• I hereby declare that I have filed an Apache Individual Contributor License Agreement.

[gnodet-bot]

Clean, minimal fix for a real bug — LGTM.

What the fix does: FormattingVisitor now tracks <ul>/<ol> nesting depth and indents each <li> with (depth - 1) spaces. A U+0001 placeholder is used for indentation spaces so the existing space-collation cleanup in toString() doesn't erase them; the placeholder is only expanded to real spaces as the last step.

Correctness verified:

• The toString() replacement ordering is essential and correct: replaceAll(" +", " ") collapses real spaces → replace("\n ", "\n") strips leading spaces after newlines → replace(INDENT_MARKER, " ") expands markers last. Any other order would corrupt the output.
• U+0001 is a C0 control character that cannot appear in Java identifiers or normal javadoc — collision risk is effectively zero.
• Top-level list output (depth=1) is identical to the old code: old code appended "\n * " then toString() stripped the leading space via replace("\n ", "\n") → "\n* "; new code appends "\n* " directly.
• Math.max(0, listDepth - 1) on decrement is reasonable defensive coding for malformed HTML.

This review was generated by an AI agent, Hermès on behalf of @gnodet.

[kwin]

@arturobernalg Thanks for the fix. LGTM, however can you edit this PRs description? The formatting looks a bit off.

[arturobernalg]

@kwin Fixed — the HTML tags weren't escaped. Should render correctly now.