Move html characeter converting mechanism to regexp-handling

[tompng]

Text#to_html_characters was a postprocess that converts ascii quotes/marks to multibyte characters. Postprocessing HTML to do that is not a good idea. Convert plain text node with regexp-handling is better.

[tompng]

I think html character is ©, not ©, but it's another issue.

[matzbot]

🚀 Preview deployment available at: https://0733bf1b.rdoc-6cd.pages.dev (commit: d8f23d9)

[st0012]

Could TO_HTML_CHARACTERS[quote.encoding] ever be nil and causes NoMethodError?

[tompng]

TO_HTML_CHARACTERS = Hash.new do |h, encoding| ... end will generate a new hash if there is no key.

[Copilot]

Pull request overview

This PR moves “smart punctuation” (quotes/dashes/ellipsis/(c)/(r) conversions) from RDoc::Text’s HTML post-processing into RDoc::Markup::ToHtml’s inline regexp-handling so conversions apply only to plain-text nodes (and avoid rewriting already-generated HTML).

Changes:

• Removed RDoc::Text#to_html_characters (and related helpers) and stopped post-processing rendered HTML.
• Added regexp-based handling in RDoc::Markup::ToHtml for HTML-character aliases and quote conversion, scoped to inline/plain-text processing.
• Updated snippet and crossref behaviors/tests to reflect the new conversion location and boundaries.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
lib/rdoc/text.rb	Removes the old HTML post-processing conversion mechanism from RDoc::Text.
lib/rdoc/markup/to_html.rb	Introduces encoding-aware character mapping + regexp-handling and quote conversion in the HTML formatter.
lib/rdoc/markup/to_html_snippet.rb	Aligns snippet truncation ellipsis and HTML conversion with the new formatter behavior.
test/rdoc/rdoc_text_test.rb	Removes tests that were specific to the deleted RDoc::Text conversion API.
test/rdoc/markup/to_html_test.rb	Adds coverage for the new formatter-level conversion behavior and encoding fallback.
test/rdoc/markup/to_html_snippet_test.rb	Updates expected truncation output to use the new ellipsis behavior.
test/rdoc/markup/to_html_crossref_test.rb	Updates expectations where generated crossref text is no longer post-processed into smart quotes.

Comments suppressed due to low confidence (1)

lib/rdoc/text.rb:172

• RDoc::Text#to_html / #to_html_characters were removed here. If any external callers still depend on these helpers (they previously handled quote/dash/ellipsis substitutions post-render), this is a breaking API change. Consider leaving a compatibility method that forwards to the new regexp-handling implementation (or raising a clearer error/deprecation) to avoid silent behavioral differences across versions.

  ##
  # Wraps +txt+ to +line_len+

  def wrap(txt, line_len = 76)
    res = []

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.