refactor(format/html): overhaul HTML formatter by dyc3 · Pull Request #8806 · biomejs/biome

dyc3 · 2026-01-19T13:19:25Z

Summary

This completely overhauls the HTML formatter to address whitespace sensitivity and other various hard-to-debug mismatches.

There are 2 major aspects:

It no longer uses best_fitting like the JsxChildList formatter does, which is more closely aligned to Prettier's IR.

There is now a concept of "internal" and "external" whitespace sensitivity. Some elements are sensitive to siblings, but not to their children.

AI was used pretty heavily to analyze prettier's IR output, and diagnose where biome's output differs.

There are some regressions, but mostly for snippets that use legacy/deprecated elements like <marquee>.

TODO:

changeset

rebase this to next

fix html full support

Test Plan

Prettier snapshots show meaningful improvements to matching prettier's output.
Some of our snapshots have changed too, but only to match what prettier would output.

Docs

ematipico, Conaclos, and siketyan reacted with hooray emoji

changeset-bot · 2026-01-19T13:19:30Z

🦋 Changeset detected

Latest commit: a8a7cdf

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 14 packages

Name	Type
@biomejs/biome	Patch
@biomejs/cli-win32-x64	Patch
@biomejs/cli-win32-arm64	Patch
@biomejs/cli-darwin-x64	Patch
@biomejs/cli-darwin-arm64	Patch
@biomejs/cli-linux-x64	Patch
@biomejs/cli-linux-arm64	Patch
@biomejs/cli-linux-x64-musl	Patch
@biomejs/cli-linux-arm64-musl	Patch
@biomejs/wasm-web	Patch
@biomejs/wasm-bundler	Patch
@biomejs/wasm-nodejs	Patch
@biomejs/backend-jsonrpc	Patch
@biomejs/prettier-compare	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

dyc3 · 2026-01-19T13:19:40Z

refactor(format/html): overhaul HTML formatter #8806 👈 (View in Graphite)
feat(dx): add prettier-compare tool #8805
feat(html): add various helpers to html syntax nodes #8804
main

This stack of pull requests is managed by Graphite. Learn more about stacking.

codspeed · 2026-01-20T11:52:05Z

CodSpeed Performance Report

Merging this PR will not alter performance

_{Comparing dyc3/html-big-refactor (a8a7cdf) with next (62dc6d7)}

Summary

✅ 58 untouched benchmarks
⏩ 95 skipped benchmarks¹

95 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

coderabbitai · 2026-01-20T11:56:57Z

Walkthrough

This PR overhauls the HTML formatter: adds a CSS-display module and whitespace-sensitivity APIs; introduces token-borrowing for adjacent inline elements (sharing/suppressing the closing >), threads borrowed state through opening/closing element formatters and element lists; rewrites element-list layout to BestFitting/Multiline with ChildrenMeta and borrowing-aware child handling; refactors comment placement and verbatim leading-comment formatting (suppression-aware); and updates Svelte auxiliary formatters to use the new element-list API and inline child formatting.

Possibly related PRs

biomejs/biome PR 8804 — Adds HTML syntax helpers and token accessors (tag_name, closing-angle tokens, AnyHtmlElement) used by the borrowing and closing-angle logic.
biomejs/biome PR 8595 — Modifies html children splitting and trailing-whitespace handling that overlap the new html_split_children and whitespace-preservation changes.
biomejs/biome PR 7467 — Introduces embedded-content formatting and propagate_expand plumbing that interacts with element-list and embedded formatting paths.

🚥 Pre-merge checks | ✅ 2

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'refactor(format/html): overhaul HTML formatter' directly and clearly summarises the main change: a comprehensive refactoring of the HTML formatter.
Description check	✅ Passed	The description explains the motivation (whitespace sensitivity issues), the two major aspects of the overhaul, AI assistance used, and the test plan with Prettier snapshots. It's fully related to the changeset.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

📝 Generate docstrings

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

coderabbitai

Actionable comments posted: 2

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

crates/biome_html_formatter/src/comments.rs (1)

83-112: Guard preceding_node before unwrap()—line 97 will panic. When a comment appears immediately before an unmatched closing tag, preceding_node() can be None, causing a panic. The inner guard at line 87 only checks whether the node is an HtmlOpeningElement, not whether preceding_node itself exists.

Suggested fix

         if let Some(_closing_tag) = comment
             .following_node()
             .and_then(|node| node.clone().cast::<HtmlClosingElement>())
         {
-            if let Some(_preceding_opening_tag) = comment
-                .preceding_node()
-                .and_then(|node| node.clone().cast::<HtmlOpeningElement>())
-            {
-                return CommentPlacement::dangling(
-                    comment.preceding_node().unwrap().clone(),
-                    comment,
-                );
-            } else {
-                return CommentPlacement::trailing(
-                    comment.preceding_node().unwrap().clone(),
-                    comment,
-                );
-            }
+            if let Some(preceding) = comment.preceding_node() {
+                if preceding.clone().cast::<HtmlOpeningElement>().is_some() {
+                    return CommentPlacement::dangling(preceding, comment);
+                }
+                return CommentPlacement::trailing(preceding, comment);
+            }
         }

🤖 Fix all issues with AI agents

In `@crates/biome_html_formatter/src/utils/children.rs`:
- Around line 401-430: Fix the typo in the explanatory comment inside
children.rs: change "attacked" to "attached" in the sentence explaining that
trivia adjacent to the closing tag is attached to the closing tag's leading
trivia; this is purely a comment change near the block that checks
closing_element and l_angle_token (the code that inspects leading_trivia and
then uses builder.entry(HtmlChild::EmptyLine/Newline/Whitespace)).

In `@crates/biome_html_formatter/src/utils/metadata.rs`:
- Around line 701-760: The SVG filter tag lookup fails because SVG_ALL_TAGS
contains mixed/camelCase entries (e.g., feBlend) while the lookup in
css_display.rs lowercases the tag before binary_search; fix by making
SVG_ALL_TAGS entries all-lowercase (e.g., feBlend -> feblend), ensure the array
remains sorted for binary_search, and update any related comments/tests if they
assume original casing; alternatively, change the lookup code to compare against
the original non-lowercased tag name instead of lowercasing — reference
SVG_ALL_TAGS and the binary_search-based SVG/MathML lookup in css_display.rs.

🧹 Nitpick comments (7)

crates/biome_html_formatter/src/utils/css_display.rs (1)
231-239: Binary search assumes sorted tag lists. If SVG_ALL_TAGS or MATHML_ALL_TAGS drift out of order, membership checks silently fail. Consider contains to remove the ordering dependency.
♻️ Possible simplification
-            if SVG_ALL_TAGS.binary_search(&other).is_ok()
-                || MATHML_ALL_TAGS.binary_search(&other).is_ok()
+            if SVG_ALL_TAGS.contains(&other) || MATHML_ALL_TAGS.contains(&other)
             {
                 CssDisplay::Block
             } else {
                 CssDisplay::Inline
             }
crates/biome_html_formatter/src/utils/children.rs (1)

62-66: Consider removing dead code or adding a TODO.

The text() method is marked with #[expect(dead_code)]. If it's intended for future use, a brief comment explaining the plan would be helpful. Otherwise, consider removing it until needed.
crates/biome_html_formatter/src/verbatim.rs (2)
241-241: Unused peekable() call.

The iterator is converted to Peekable but peek() is never called. This can be simplified.
♻️ Simplify iterator
-    let leading_comments_iter = leading_comments.iter().peekable();
-    for comment in leading_comments_iter {
+    for comment in leading_comments {
276-289: TODO for line comment logic review.

The TODO at line 277 indicates this logic needs review. The current implementation handles spacing based on lines_before and lines_after, but the interaction may need further verification against Prettier's behaviour.

Would you like me to help investigate how Prettier handles line comments in HTML to verify this logic?
crates/biome_html_formatter/src/html/auxiliary/element.rs (2)
123-130: Add a comment explaining the nth(2) magic number.

A past review requested clarification. The inline comment explains what nth(2) does, but a brief note on why index 2 specifically would help future readers.
📝 Enhance comment
         let is_root_element_list = node
             .syntax()
             .ancestors()
-            // get the third element (index 2)
-            // because the first one is Self (`HtmlElement`), 2nd one is the `HtmlElementList`
-            // third one is either `HtmlRoot` or another `HtmlElement`
+            // Get the grandparent (index 2 in ancestors):
+            // - Index 0: Self (HtmlElement)
+            // - Index 1: Parent HtmlElementList
+            // - Index 2: Grandparent (HtmlRoot if at root level, otherwise another HtmlElement)
             .nth(2)
             .is_some_and(|ancestor| HtmlRoot::can_cast(ancestor.kind()));
303-313: Consider clarifying operator precedence in condition.

The condition spans multiple lines with && and || operators. While correct, the precedence could be clearer with explicit parentheses or restructuring.
♻️ Clarify precedence
-    if !node.children().is_empty() && tag_name.eq_ignore_ascii_case("body")
-        || node.children().iter().any(|child| {
+    let is_non_empty_body =
+        !node.children().is_empty() && tag_name.eq_ignore_ascii_case("body");
+    let grandchildren_have_elements = node.children().iter().any(|child| {
             if let Some(element) = child.as_html_element() {
                 has_non_text_child(&element.children())
             } else {
                 false
             }
-        })
-    {
+        });
+
+    if is_non_empty_body || grandchildren_have_elements {
         return true;
     }
crates/biome_html_formatter/src/html/lists/element_list.rs (1)
391-394: Commented-out code should be removed or documented.

The // || children_meta.has_block_element appears to be leftover from development. If it's intentionally disabled, please add a TODO comment explaining the rationale. Otherwise, remove it to keep the codebase clean.
Suggested fix
             let mut force_multiline = matches!(self.layout, HtmlChildListLayout::Multiline)
-                // || children_meta.has_block_element
                 || children_meta.multiple_block_elements;

ematipico

Unfortunately there's a lot of code that I can hardly follow, so take my review with a pinch of salt. Looking at the snapshots though, I think we made some great improvements!

Changeset is missing, maybe a patch with some like this

**BREAKING CHANGES the HTML formatter only.**

Refactor the internal logic of the HTML formatter to be closer to Prettier's formatting.... etc.

ematipico · 2026-01-19T14:38:12Z

-//! This implementation is very heavily inspired by the JSX formatter implementation for JsxChildList.
-
-use std::cell::RefCell;
+//! # Caution


I love this comment

coderabbitai

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@crates/biome_html_formatter/src/html/lists/element_list.rs`:
- Around line 1-182: Fix the minor spelling/grammar mistakes in the top-of-file
documentation in element_list.rs: update misspellings such as "acomplishes" to
"accomplishes", "wierd" to "weird", and any other small typos or punctuation
issues in the HtmlElementList/HTML formatter doc comments so the documentation
reads cleanly and consistently.

🧹 Nitpick comments (1)

.changeset/ready-taxes-build.md (1)
5-9: Add a tiny before/after diff example.
For formatter overhauls, a short diff helps users gauge impact.
Proposed addition
-<div><span>foo</span><span>bar</span></div>
+<div><span>foo</span> <span>bar</span></div>
Based on learnings, please include a diff snippet for formatter changes.

coderabbitai

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

crates/biome_html_formatter/src/comments.rs (1)

95-112: Avoid unwrap() here to prevent a formatter crash on malformed HTML.
If a comment precedes a stray closing tag, preceding_node() can be None, and this will panic.

🛠️ Proposed guard

-            if let Some(_closing_tag) = comment
+            if let Some(closing_tag) = comment
                 .following_node()
                 .and_then(|node| node.clone().cast::<HtmlClosingElement>())
             {
-                if let Some(_preceding_opening_tag) = comment
-                    .preceding_node()
-                    .and_then(|node| node.clone().cast::<HtmlOpeningElement>())
-                {
-                    return CommentPlacement::dangling(
-                        comment.preceding_node().unwrap().clone(),
-                        comment,
-                    );
-                } else {
-                    return CommentPlacement::trailing(
-                        comment.preceding_node().unwrap().clone(),
-                        comment,
-                    );
-                }
+                if let Some(preceding) = comment.preceding_node() {
+                    if preceding.clone().cast::<HtmlOpeningElement>().is_some() {
+                        return CommentPlacement::dangling(preceding.clone(), comment);
+                    }
+                    return CommentPlacement::trailing(preceding.clone(), comment);
+                }
+                return CommentPlacement::leading(closing_tag.syntax().clone(), comment);
             }

add get_css_whitespace mostly AI slop but cleaned up a little progress? i think? fix breaking when attributes break fix spacing between comments and text fix something

…rowing

… container is not internally sensitive

specifically for html/bracket-same-line/void-elements.html

coderabbitai

Actionable comments posted: 1

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (2)

crates/biome_html_formatter/src/comments.rs (1)

95-113: Potential panic when preceding_node() is None.

Line 109 calls unwrap() on comment.preceding_node() in the else branch, but the outer condition only checks that following_node() is a HtmlClosingElement. If there's no preceding node (e.g., comment is the first child), this will panic.

🐛 Proposed fix

             if let Some(_preceding_opening_tag) = comment
                 .preceding_node()
                 .and_then(|node| node.clone().cast::<HtmlOpeningElement>())
             {
                 return CommentPlacement::dangling(
                     comment.preceding_node().unwrap().clone(),
                     comment,
                 );
-            } else {
+            } else if let Some(preceding) = comment.preceding_node() {
                 return CommentPlacement::trailing(
-                    comment.preceding_node().unwrap().clone(),
+                    preceding.clone(),
                     comment,
                 );
             }

crates/biome_html_formatter/src/html/auxiliary/element.rs (1)

196-207: Borrowing still enabled for force‑break elements.

The comment says force‑break elements should not borrow tokens, but the guards don’t check forces_break_children. If that’s unintended, add the guard so the implementation matches the rule.

💡 Suggested patch

-        let should_borrow_opening_r_angle = is_element_internally_whitespace_sensitive
+        let should_borrow_opening_r_angle = is_element_internally_whitespace_sensitive
+            && !forces_break_children
             && !children.is_empty()
             && !content_has_leading_whitespace
             && !should_be_verbatim
             && !should_format_embedded_nodes;
-        let should_borrow_closing_tag = is_element_internally_whitespace_sensitive
+        let should_borrow_closing_tag = is_element_internally_whitespace_sensitive
+            && !forces_break_children
             && !children.is_empty()
             && !content_has_trailing_whitespace
             && !should_be_verbatim
             && !should_format_embedded_nodes;

🤖 Fix all issues with AI agents

In `@crates/biome_html_formatter/src/html/lists/element_list.rs`:
- Around line 375-413: The code computes force_multiline for children but
ignores the stored opening_tag_group option; update the multiline decision to
consult opening_tag_group (or remove the unused field). Specifically, in the
block that initializes force_multiline (currently using matches!(self.layout,
HtmlChildListLayout::Multiline) || children_meta.multiple_block_elements), also
OR in a check of self.opening_tag_group indicating a broken/multiline opening
tag (e.g., self.opening_tag_group.is_multiline() or
matches!(self.opening_tag_group, OpeningTagGroup::Multiline)). Ensure you
reference the same boolean (force_multiline) so that if the opening tag broke
onto multiple lines the children are forced multiline; alternatively, if you
decide not to use it, remove the opening_tag_group field and any code that sets
it to avoid dead state.

🧹 Nitpick comments (1)

crates/biome_html_formatter/src/verbatim.rs (1)

276-289: TODO comment needs attention.

The CommentKind::Line branch has a TODO indicating the logic needs review. Since HTML comments are now classified as Block (per get_comment_kind), this branch may be unreachable for standard HTML. However, if this code path is exercised by other syntaxes or edge cases, the lines_after == 0 case doing nothing could be intentional or a gap.

Would you like me to verify whether this code path is reachable, or help clarify the intended behaviour?

coderabbitai · 2026-01-21T17:23:34Z

+        let formatted_children = format_with(|f| {
+            // Print borrowed opening `>` token
+            if let Some(ref r_angle_token) = self.borrowed_tokens.borrowed_opening_r_angle {
+                write!(f, [r_angle_token.format()])?;
+            }

-                        None => None,
-                    };
+            // Determine if we should force multiline
+            // Force multiline if:
+            // - Layout is explicitly set to Multiline
+            // - Children contain block elements
+            // - There are multiple block elements
+            let mut force_multiline = matches!(self.layout, HtmlChildListLayout::Multiline)
+                || children_meta.multiple_block_elements;
+            let mut children_iter = HtmlChildrenIterator::new(children.iter());
+
+            // Trim leading new lines, but track if there were any
+            let mut had_leading_newline = false;
+            if !self.is_container_whitespace_sensitive {
+                while matches!(
+                    children_iter.peek(),
+                    Some(HtmlChild::Newline | HtmlChild::EmptyLine)
+                ) {
+                    had_leading_newline = true;
+                    children_iter.next();
+                }
+            }

-                    child_breaks = separator.is_some_and(|separator| separator.will_break());
+            // Force multiline if there was a leading newline AND there are block elements
+            // This respects the user's intent to break when they have:
+            // <div>
+            //   <div>...</div>
+            // </div>
+            if had_leading_newline && children_meta.has_block_element {
+                force_multiline = true;
+            }

-                    flat.write(&format_args![word, separator], f);
+            if force_multiline {
+                write!(f, [expand_parent()])?;
+            }


⚠️ Potential issue | 🟡 Minor

🧩 Analysis chain

🏁 Script executed:

# First, let's explore the file structure find . -name "element_list.rs" -type f

Repository: biomejs/biome

Length of output: 117

🏁 Script executed:

# Read the file to understand its context and find the relevant lines wc -l crates/biome_html_formatter/src/html/lists/element_list.rs

Repository: biomejs/biome

Length of output: 119

🏁 Script executed:

# Read around lines 375-413 to see the actual code sed -n '350,450p' crates/biome_html_formatter/src/html/lists/element_list.rs | cat -n

Repository: biomejs/biome

Length of output: 5671

🏁 Script executed:

# Search for opening_tag_group usage in this file rg "opening_tag_group" crates/biome_html_formatter/src/html/lists/element_list.rs -n

Repository: biomejs/biome

Length of output: 212

🏁 Script executed:

# Search for where opening_tag_group is defined/stored (likely in struct initialization) rg "opening_tag_group" crates/biome_html_formatter/src/html/lists/ -n -B 2 -A 2

Repository: biomejs/biome

Length of output: 1380

🏁 Script executed:

# Look for the struct definition that contains opening_tag_group rg "struct.*{" crates/biome_html_formatter/src/html/lists/element_list.rs -A 20 | head -60

Repository: biomejs/biome

Length of output: 138

🏁 Script executed:

# Search for all uses of opening_tag_group - specifically where it's READ (not just assigned) rg "self\.opening_tag_group" crates/biome_html_formatter/src/html/lists/element_list.rs -n

Repository: biomejs/biome

Length of output: 120

🏁 Script executed:

# Also check for any usage of opening_tag_group without self rg "opening_tag_group" crates/biome_html_formatter/src/html/lists/element_list.rs -n

Repository: biomejs/biome

Length of output: 212

🏁 Script executed:

# Let's look at the whole fmt_children method and surrounding context sed -n '200,420p' crates/biome_html_formatter/src/html/lists/element_list.rs | cat -n

Repository: biomejs/biome

Length of output: 10333

Wire opening_tag_group into the multiline decision or remove it.

opening_tag_group is stored via options but never consulted when determining if children should be multiline. This means children can stay flat even when the opening tag's attributes break onto multiple lines. Wire it into the decision logic if you want them in sync, otherwise drop the unused field.

🤖 Prompt for AI Agents

In `@crates/biome_html_formatter/src/html/lists/element_list.rs` around lines 375 - 413, The code computes force_multiline for children but ignores the stored opening_tag_group option; update the multiline decision to consult opening_tag_group (or remove the unused field). Specifically, in the block that initializes force_multiline (currently using matches!(self.layout, HtmlChildListLayout::Multiline) || children_meta.multiple_block_elements), also OR in a check of self.opening_tag_group indicating a broken/multiline opening tag (e.g., self.opening_tag_group.is_multiline() or matches!(self.opening_tag_group, OpeningTagGroup::Multiline)). Ensure you reference the same boolean (force_multiline) so that if the opening tag broke onto multiple lines the children are forced multiline; alternatively, if you decide not to use it, remove the opening_tag_group field and any code that sets it to avoid dead state.

dyc3 mentioned this pull request Jan 19, 2026

feat(html): add various helpers to html syntax nodes #8804

Merged

dyc3 mentioned this pull request Jan 19, 2026

feat(dx): add prettier-compare tool #8805

Merged

1 task

github-actions Bot added A-Formatter Area: formatter L-HTML Language: HTML and super languages labels Jan 19, 2026

dyc3 changed the title ~~WIP~~ refactor(format/html): HTML formatter overhaul Jan 19, 2026

dyc3 requested review from a team January 19, 2026 13:33

dyc3 marked this pull request as ready for review January 19, 2026 13:33

dyc3 changed the title ~~refactor(format/html): HTML formatter overhaul~~ refactor(format/html): overhaul HTML formatter Jan 19, 2026

ematipico reviewed Jan 19, 2026

View reviewed changes

Comment thread crates/biome_html_formatter/src/html/auxiliary/element.rs Resolved

Base automatically changed from dyc3/prettier-compare-tool to main January 20, 2026 11:14

dyc3 added this to the Biome v2.4 milestone Jan 20, 2026

dyc3 force-pushed the dyc3/html-big-refactor branch from a4a37ea to 5f36696 Compare January 20, 2026 11:49

coderabbitai Bot reviewed Jan 20, 2026

View reviewed changes

Comment thread crates/biome_html_formatter/src/utils/children.rs Resolved

Comment thread crates/biome_html_formatter/src/utils/metadata.rs Resolved

github-actions Bot added the A-Project Area: project label Jan 20, 2026

ematipico approved these changes Jan 20, 2026

View reviewed changes

dyc3 force-pushed the dyc3/html-big-refactor branch from bf4ba55 to 24321e3 Compare January 21, 2026 16:15

dyc3 changed the base branch from main to next January 21, 2026 16:15

coderabbitai Bot reviewed Jan 21, 2026

View reviewed changes

Comment thread crates/biome_html_formatter/src/html/lists/element_list.rs Resolved

coderabbitai Bot reviewed Jan 21, 2026

View reviewed changes

github-actions Bot added the A-CLI Area: CLI label Jan 21, 2026

dyc3 added 5 commits January 21, 2026 12:14

WIP

de1723d

add get_css_whitespace mostly AI slop but cleaned up a little progress? i think? fix breaking when attributes break fix spacing between comments and text fix something

docs: gather thoughts on full requirements

6e0dd7d

remove new impl, start over

5444f64

mostly ai slop draft

3529bc1

adjust whitespace sensitivity logic abstraction

ffbf115

dyc3 added 25 commits January 21, 2026 12:14

fix: better inline element whitespace preservation via more token bor…

40facca

…rowing

fix: doc tests

2c82bf2

chore: accept with-colon.html snapshot that shows improvement

48e3f81

feat: handle svg and mathml elements

bdcf9f2

chore: dead code

a8eb21e

test: add sanity test

33ed0db

fix: don't bother preserving whitespace at the end of the list if the…

b6684ee

… container is not internally sensitive

chore: remove dead code

286f119

fix: clippy lint

3644845

chore: rm more dead code

aa502cf

fix: table element formatting

ae610c6

chore: accept snapshots

0a29ba5

chore: rm commented test

054d3f2

chore: actually accept snapshot

3f28589

chore: admit defeat on a regression for inline self closing elements

ff36801

specifically for html/bracket-same-line/void-elements.html

chore: remove dead code

25e2f10

chore: add comment

c201e06

chore: typo

80fee91

fix: propogate expand for embedded content

d53779b

chore: remove some unused code

603daaa

fix: lowercase svg tags

05a9271

chore: changeset

a2b80f9

fix: botched rebase for crates/biome_html_formatter/src/comments.rs

fbd6b44

chore: misc typos

23d9de3

chore: update cli tests snapshots

a8a7cdf

dyc3 force-pushed the dyc3/html-big-refactor branch from 574976c to a8a7cdf Compare January 21, 2026 17:14

coderabbitai Bot reviewed Jan 21, 2026

View reviewed changes

dyc3 merged commit cb112ce into next Jan 21, 2026
32 checks passed

dyc3 deleted the dyc3/html-big-refactor branch January 21, 2026 17:46

github-actions Bot mentioned this pull request Feb 14, 2026

ci: release #9045

Merged

Navigation Menu

Repository navigation

refactor(format/html): overhaul HTML formatter - #8806

refactor(format/html): overhaul HTML formatter#8806
dyc3 merged 61 commits into
nextfrom
dyc3/html-big-refactor

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Footnotes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!