Empty <msqrt> is misaligned vertically relative to <msqrt><mrow></mrow></msqrt>

[Ahmad-S792]

1b116ea

Empty <msqrt> is misaligned vertically relative to <msqrt><mrow></mrow></msqrt>
https://bugs.webkit.org/show_bug.cgi?id=302662
rdar://165365714

Reviewed by Frédéric Wang Nélar.

Per MathML Core, an empty <msqrt> must render as if it had a single empty
<mrow> child (https://w3c.github.io/mathml-core/#radicals-msqrt-mroot). While
layout produces the same box, the baseline reported by the renderer does not:
RenderMathMLRow::firstLineBaseline() returns nullopt when there are no
in-flow children, so the inline-block fallback uses the box's bottom edge.
The reference case with an explicit empty <mrow> child instead reports a
baseline at radicalAscent, which is where the implicit mrow sits above the
radical's rule line. The two forms therefore sit at different vertical
positions on the line (offset by radicalDescent, which is non-zero whenever
the stretched radical operator is taller than verticalGap + ruleThickness).

Override firstLineBaseline() in RenderMathMLRoot to synthesize the baseline
of the implicit empty mrow when the root is a SquareRoot with no in-flow
child. With no base content, baseAscent is zero, so the synthesized ascent
collapses to verticalGap + ruleThickness + extraAscender (plus border and
padding before). Compute this on demand from verticalParameters() rather
than caching it in a member, and make verticalParameters() const so it can
be called from firstLineBaseline().

RootType::RootWithIndex with a missing child is !isValid() and already falls
through to plain row layout, where the default baseline is correct, so it
does not need special handling.

* LayoutTests/TestExpectations: Progression
* Source/WebCore/rendering/mathml/RenderMathMLRoot.cpp:
(WebCore::RenderMathMLRoot::verticalParameters const):
(WebCore::RenderMathMLRoot::firstLineBaseline const):
* Source/WebCore/rendering/mathml/RenderMathMLRoot.h:

Canonical link: https://commits.webkit.org/313000@main

980f4e4

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows	Apple Internal
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win	✅ 🛠 ios-apple
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	🧪 wpe-wk2	🧪 win-tests	✅ 🛠 mac-apple
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe		❌ 🛠 vision-apple
	🧪 ios-wk2-wpt	🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 vision	🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
✅ 🛠 🧪 merge	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	✅ 🛠 playstation
	✅ 🧪 vision-wk2	✅ 🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-early-warning-system]

EWS run on previous version of this PR (hash bd117e0)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows	Apple Internal
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	🛠 win	✅ 🛠 ios-apple
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	✅ 🧪 wpe-wk2	🧪 win-tests	✅ 🛠 mac-apple
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	🧪 api-mac	✅ 🧪 api-wpe		✅ 🛠 vision-apple
🧪 webkitpy	🧪 ios-wk2-wpt	🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
🧪 services	✅ 🛠 vision	✅ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	✅ 🛠 playstation
	✅ 🧪 vision-wk2	🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-early-warning-system]

EWS run on previous version of this PR (hash 3d3db47)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows	Apple Internal
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win	✅ 🛠 ios-apple
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	✅ 🧪 wpe-wk2	❌ 🧪 win-tests	✅ 🛠 mac-apple
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe		✅ 🛠 vision-apple
	✅ 🧪 ios-wk2-wpt	✅ 🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 vision	✅ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	✅ 🛠 playstation
	✅ 🧪 vision-wk2	✅ 🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[fred-wang]

It's a pity we have to introduce a new member just for the sake of handling the empty msqrt... The alternative would be to use an anonymous mrow, which I guess is worse memory-wise

Incidentally, it's possible we can compute these LayoutUnit members needed as needed (sharing code with layout) if that's not too computationally costly, so that we avoid the storage and invalidation issue. I believe that's what Chromium do. Anyway, probably not worth considering for this issue.

[Ahmad-S792]

@fred-wang - I did changes here, so it would be good, if you can have another look. Thanks!

[webkit-early-warning-system]

EWS run on current version of this PR (hash 980f4e4)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows	Apple Internal
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win	✅ 🛠 ios-apple
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	🧪 wpe-wk2	🧪 win-tests	✅ 🛠 mac-apple
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe		❌ 🛠 vision-apple
	🧪 ios-wk2-wpt	🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 vision	🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
✅ 🛠 🧪 merge	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	✅ 🛠 playstation
	✅ 🧪 vision-wk2	✅ 🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[fred-wang]

OK nice!

[webkit-commit-queue]

Committed 313000@main (1b116ea): https://commits.webkit.org/313000@main

Reviewed commits have been landed. Closing PR #64606 and removing active labels.