[wittjeff]

Fixes #745.

What changed

Markdown / plain text. MarkdownMetaSerializer._serialize_meta_field had no branch for LanguageMetaField, so it fell through to str(field_val) and the export gained a paragraph like confidence=0.99 created_by='langdetect' code=<HumanLanguageLabel.DE: 'de'>. A language code is metadata, not content: it now renders only when mark_meta=True ([Language] de) and is skipped otherwise. Plain text inherits the same serializer.

HTML.

• HTMLParams.html_lang was declared but never used; the root element was the literal "<html>". It is now <html lang="…"> from html_lang (default "en", unchanged) — WCAG 3.1.1 Language of Page.
• Text items whose meta.language differs from html_lang get a lang attribute on their <p> / <hN> element via the existing attrs parameter of get_html_tag_with_text_direction — WCAG 3.1.2 Language of Parts. Items in the document language get no attribute, so output is unchanged for documents without language meta.
• The language row in the meta <details> block is left as is (covered by test_semantic_base_meta_fields_roundtrip_and_html_rendering).

Tests

• test_md_and_text_language_meta_is_not_rendered_as_prose and test_html_declares_document_and_element_languages in test/test_metadata.py.
• Groundtruth regenerated with DOCLING_GEN_TEST_DATA=1: the 23 HTML files change only on the <html> line (PNG fixtures that regenerated with byte-level churn were reverted).
• pre-commit run --all-files and the full suite pass locally (701 passed, 6 skipped).

Context

Downstream, yra-doc-pipeline populates meta.language from a deterministic detector and maps it to PDF/UA /Lang; with this change the same data reaches the HTML export instead of leaking into text exports.

🤖 Generated with Claude Code

[github-actions]

✅ DCO Check Passed

Thanks @wittjeff, all your commits are properly signed off. 🎉

[mergify]

Merge Protections

🔴 1 of 2 protections blocking · waiting on 👀 reviews

	Protection	Waiting on
🔴	Require two reviewer for test updates	👀 reviews
🟢	Enforce conventional commit	—

🔴 Require two reviewer for test updates

Waiting for

• #approved-reviews-by >= 2

This rule is failing.

When test data is updated, we require two reviewers

• #approved-reviews-by >= 2

---

Show 1 satisfied protection

🟢 Enforce conventional commit

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[github-actions]

AI triage (advisory)

Checked commit b46f4e88f9. The analysis read the diff only. It did not run code from this pull request.

The PR stops MarkdownMetaSerializer from printing meta.language as a text paragraph. It shows the language only when mark_meta=True. In HTML, it writes html_lang into the <html lang> root tag and adds lang attributes to <p> and <hN> elements whose language differs.

Topic labels added: bug, serialization.

CI safety: no concern found. A maintainer must still approve CI.

Possible duplicates

PR	State	Signal	Assessment
#869	open	also references issue #745	duplicate: The same author opened it for the same issue #745. It hides meta.language in Markdown and adds <html lang> and lang attributes in HTML. It edits the same code and tests, so only one PR can merge.

2 related pull request(s) in the same code

PR	State	Signal	Assessment
#701	open	changes the same code	related: It changes list item marker styles in html.py. It does not touch language handling, and the two changes can merge together.
#933	open	changes the same code	related: It changes how headings with inline children render in html.py and markdown.py. It does not touch language metadata.

Reference data changes

Change	Files
text	23

• text tests/data/doc/2206.01062.yaml.html: +1 / -1 lines
• text tests/data/doc/2408.09869_p1_split.gt.html: +1 / -1 lines
• text tests/data/doc/2408.09869v3_enriched_p1_include_annotations_false.gt.html: +1 / -1 lines
• text tests/data/doc/2408.09869v3_enriched_p1_include_annotations_true.gt.html: +1 / -1 lines
• text tests/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html: +1 / -1 lines
• text tests/data/doc/2408.09869v3_enriched_split.gt.html: +1 / -1 lines
• text tests/data/doc/2408.09869v3_enriched_split_p2.gt.html: +1 / -1 lines
• text tests/data/doc/activities.gt.html: +1 / -1 lines
• text tests/data/doc/activities_p1.gt.html: +1 / -1 lines
• text tests/data/doc/activities_p2.gt.html: +1 / -1 lines
• text tests/data/doc/barchart.gt.html: +1 / -1 lines
• text tests/data/doc/concatenated.html: +1 / -1 lines
• text tests/data/doc/constructed_doc.embedded.html.gt: +1 / -1 lines
• text tests/data/doc/constructed_doc.placeholder.html.gt: +1 / -1 lines
• text tests/data/doc/constructed_doc.referenced.html.gt: +1 / -1 lines
• text tests/data/doc/constructed_document.yaml.html: +1 / -1 lines
• text tests/data/doc/constructed_orig_false.gt.html: +1 / -1 lines
• text tests/data/doc/constructed_orig_true.gt.html: +1 / -1 lines
• text tests/data/doc/dummy_doc.yaml.html: +1 / -1 lines
• text tests/data/doc/inline_and_formatting.gt.html: +1 / -1 lines
• text tests/data/doc/polymers.gt.html: +1 / -1 lines
• text tests/data/doc/rich_table.gt.html: +1 / -1 lines
• text tests/data/docling_document/export/formula_mathml.html: +1 / -1 lines

Assessment: expected. Each of the 23 HTML files changes by one line (+1/-1). This matches the change of the root tag from <html> to <html lang="en"> in HTMLDocSerializer. The summary shows no other changes.

_{The ai:* labels are advisory. They do not approve CI and they do not add tests:full. A new push runs this check again.}

[github-actions]

AI first review (advisory): no blocker found

This review covers the full diff at b46f4e88f9. It read the code and did not run it.

No merge-deciding defect found. I checked the Markdown and plain-text language skip, the HTML <html lang> and element lang attributes, the attribute escaping in get_html_tag_with_text_direction, the subclasses of MarkdownMetaSerializer, and the new tests. The <html> line is the only change in the HTML reference data, as the PR describes.

_{A maintainer makes the merge decision. This review never approves the pull request.}