- background properties
- box-shadow
- cursor
- SVG fill/stroke properties
- image-rendering
- outline properties
- pointer-events
- user-select
This should be basically all of them. I skipped `opacity` and
`transform` since establishing a stacking context feels like a
layout-affecting thing, but I could be very wrong on that!
The previous implementation used relative X offsets for both left and
right-side floats. This made right-side floats super awkward, since we
could only determine their X position once the width of the BFC root was
known, and for BFC roots with automatic width, this was not even working
at all most of the time.
This patch changes the way we deal with floats so that BFC keeps track
of the offset-from-edge for each float. The offset is the distance from
the BFC root edge (left or right, depending on float direction) to the
"innermost" margin edge of the floating box.
Floating box are now laid out in two passes: while going through the
normal flow layout, we put floats in their *static* position (i.e the
position they would have occupied if they weren't floating) and then
update the Y position value to the final one.
The second pass occurs later on, when the BFC root has had its width
assigned by the parent context. Once we know the root width, we can
set the X position value of floating boxes. (Because the X position of
right-side floats is relative to the right edge of the BFC root.)
Using the intrinsic size cache means we only perform the nested layout
to determine intrinsic size *once* per root layout pass.
Furthermore, by using a throwaway FormattingState, details of the nested
layout can't leak into and mutate the outer layout.
Instead of caching them with the current state, we can cache them at the
root of the state tree. Since intrinsic sizes are immutable during the
same layout, this allows layout to take advantage of intrinsic sizes
discovered during nested layout (and avoids a *lot* of duplicate work.)
FormattingStates can have parents, in case we're performing nested
layouts to determine something's intrinsic size. In those cases, it will
soon be useful to find the outermost (root) state.
No need to call the expensive establishes_stacking_context() here, as
we've already built the stacking context tree and can simply test for
the presence of existing stacking contexts.
This is preparation for allowing blocks with their own internal BFC to
flow around floating boxes in the parent BFC.
Note that IFC still has the available_space_for_line() API, which
returns space available within the IFC's own containing block, while the
BFC available_space_for_line() returns space available within its root.
In Selectors level 4, `:nth-child()` and `:nth-last-child()` can both
optionally take a selector-list argument. This selector-list acts as a
filter, so that only elements matching the list are counted. For
example, this means that the following are equivalent:
```css
:nth-child(2n+1 of p) {}
p:nth-of-type(2n+1) {}
```
This fixes the specificity for :not(), :is() and :where(). Also, we now
clamp the specificity numbers instead of letting them overflow, and I
sprinkled in some spec comments for good measure.
`<forgiving-selector-list>` and `<forgiving-relative-selector-list>` are
the same as regular selector-lists, except that an invalid selector
does not make the whole list invalid. The former is used by the `:is()`
pseudo-class.
For example:
```css
/* This entire selector-list is invalid */
.foo, .bar, !?invalid { }
/* This is valid, but the "!?invalid" selector is removed */
:is(.foo, .bar, !?invalid) { }
```
Also as part of this, I've removed the `parse_a_selector(TokenStream)`
and `parse_a_relative_selector(TokenStream)` methods as they don't add
anything useful.
It makes no sense to require passing a global object and doing a stack
space check in some cases where running out of stack is highly unlikely,
we can't recover from errors, and currently ignore the result anyway.
This is most commonly in constructors and when setting things up, rather
than regular function calls.
Right now the only functionality supported is getting/setting via JS
and resetting when browsing cross origin.
The HTML Specification (7.11 Browsing the web) also specifies how the
name should be restored from history entries, but we don't have those
yet.
As noted, this is not 100% to the spec, but effectively the same -
`no-preference` is only allowed to appear in features that evaluate it
as false in a boolean context. This is also the only identifier besides
`none` that evaluates to false. If other identifiers gain this property
in the future, we can make it more robust then.
This adds (or at least stubs-out) the following:
- display-mode
- dynamic-range
- environment-blending
- forced-colors
- horizontal-viewport-segments
- vertical-viewport-segments
- inverted-colors
- nav-controls
- prefers-contrast
- prefers-reduced-data
- prefers-reduced-motion
- prefers-reduced-transparency
- scripting
- video-color-gamut
- video-dynamic-range
The `@media (inverted-colors)` CSS that the spec requires we add to the
UA style sheet does not actually do anything for us yet since we don't
support `filter`, but it seemed sensible to include it now to avoid
forgetting later. :^)
This was causing us to miss layout invalidations. With this fixed, we
can remove the invalidation from Element::recompute_style() along with
the associated FIXME.
Thanks to Idan for spotting this! :^)
By the time we're painting, we've already built the stacking context
tree. So instead of asking if a box establishes a stacking context, we
can ask if its paintable *has* a stacking context.
This was taking up ~6% of the profile when mousing around on the HTML
specification. With this change, it disappears completely. :^)
Instead of invalidating style for the entire document, we now locate the
nearest common ancestor between the old and new innermost hovered node,
and only invalidate that ancestor and its descendants.
This drastically reduces the amount of style update work when mousing
around on GitHub (and any other pages, really.) It's actually really
really snappy now. Very cool! :^)
Use the new CSS::property_affects_layout() helper to figure out if we
actually need to perform a full relayout after recomputing style.
There are three tiers of required invalidation after an element receives
new style: none, repaint only, or full relayout.
This avoids the need to rebuild the layout tree (and perform layout on
it) when trivial properties like "color" etc are changed.
This patch adds CSS::property_affects_layout(PropertyID) which tells us
whether a CSS property would affect layout if it were changed.
This will be used to avoid unnecessary relayout work when something
changes that really only requires us to repaint the page.
To mark a property as not affecting layout, set "affects-layout" to
false in the corresponding Properties.json entry. Note that all
properties affect layout by default.
Use the stub implementation of HTMLOptionsCollection to expose the
`option` children of `select` elements.
This fixes a JS error on openstreetmap.org, which occured when JQuery
code tried to access `options.length`:
https://github.com/jquery/jquery/blob/main/src/attributes/val.js#L121
This is a subtype of `DOM::HTMLCollection` that only holds
`HTMLOptionElement`s. In this stub implementation only `item`,
`namedItem` and `length`, inherited from HTMLCollection, are exposed.
This is good enough for applications that only read the collection.
Relative font-sizes like "2em" were previously resolved against the
fallback value (10px) which led to incorrect layouts in many places.
Fix this by resolving relative font-sizes against the absolutized
font-size of the parent or root element as appropriate.
This doesn't have any performance benefit yet as we still do string
comparisons everytime, but it should improve once type_state() has a
better implementation.
This commit is messy due to the Paintable and Layout classes being
tangled together.
The RadioButton, CheckBox and ButtonBox classes are now subclasses of
FormAssociatedLabelableNode. This subclass separates these layout nodes
from LabelableNode, which is also the superclass of non-form associated
labelable nodes (Progress).
ButtonPaintable, CheckBoxPaintable and RadioButtonPaintable no longer
call events on DOM nodes directly from their mouse event handlers;
instead, all the functionality is now directly in EventHandler, which
dispatches the related events. handle_mousedown and related methods
return a bool indicating whether the event handling should proceed.
Paintable classes can now return an alternative DOM::Node which should
be the target of the mouse event. Labels use this to indicate that the
labeled control should be the target of the mouse events.
HTMLInputElement put its activation behavior on run_activation_behavior,
which wasn't actually called anywhere and had to be manually called by
other places. We now use activation_behavior which is used by
EventDispatcher.
This commit also brings HTMLInputElement closer to spec by removing the
did_foo functions that did ad-hoc event dispatching and unifies the
behavior under run_input_activation_behavior.
TreeBuilder wasn't taking advantage of the fact that we already have
computed style cached on each DOM::Element by the time we're
constructing a layout tree.
So instead of using the cached style, we recomputed it from scratch for
every element. This was done because invalidation was broken in many
places, but now that it's more or less trustworthy, stop recomputing
style on the fly in TreeBuilder and use what the preceding style update
pass gave us instead.
This basically cuts style computation work in half. :^)
We were not actually walking past the first ancestor when setting
child-needs-update bit upwards.
Also, let's walk all the way to the root, even if there's a
child-needs-update bit already set. This ensures that we always leave
this function with the ancestor chain in a sane state.
