When inserting a new utf-16 surrogate next to an existing surrogate
with replaceData, the surrogates would not get merged correctly into a
single code point. This is because internally the text data is stored
as utf-8, and the two surrogates would be converted seperately. This
has now been fixed by first recreating the whole string in utf-16 and
then converting it back to utf-8.
It's not the most efficient solution, but this fixes at least 6 WPT
subtests.
This isn't directly in the spec, but since replaceChild is implemented
in terms of remove + insert, the removal step may cause arbitrary code
to execute, and so we have to verify that the replaceChild inputs still
make sense afterwards, before doing the insertion.
This roughly matches what WebKit does, and makes a bunch of HTML parsing
tests in WPT stop asserting.
In particular, input character lookahead now knows how to stop at the
insertion point marker if needed.
This makes it possible to do amazing things like having document.write()
insert doctypes one character at a time.
Elements with transforms were tested on their pre-transformed
positions, causing incorrect hits.
Copy the position transformation done in `StackingContext::hit_test`
to ensure that hit tests are done on the _actual_ position.
If we reach the insertion point at the same time as we switch to another
tokenizer state, we have to bail immediately without proceeding with the
next code point. Otherwise we'd fetch the next token, get an EOF marker,
and then proceed as if we're at the end of the input stream, even though
more data may be coming (with more calls to document.write()..)
The repository being in static storage is a bit of a hodgepodge, but in
line with how our current storage partitioning is done. We should
eventually move this, along with other across browsing context APIs to a
proper location at a later stage. But for now, this makes progress on
the meat of the BroadcastChannel API.
In the case where we had a preferred aspect ratio and a natural height
but no natural width, we'd get into ping-ponging infinite recursion by
trying to find the width to resolve the height to resolve the width to
resolve the height...
This one is particularly weird as there's a priority order, and we even
have to look at attributes from the container element if we're inside a
subframe.
This change adds support for computing accessible names for SVG
elements, per the https://w3c.github.io/svg-aam/#mapping_additional_nd
spec requirements. Otherwise, without this change, accessible names for
SVG elements don’t get exposed as expected.
Also removing a FIXME about not covering all of the event names as it is
not exactly clear when such a FIXME would be addressed, especially as
these come from multiple specifications.
Now, along with the mouse events we also dispatch pointerup, pointerdown
and pointermove.
With this change shape painting works on https://excalidraw.com/
Previously, the list was copied when constructing the FormData object,
then the original list was passed to the event, meaning any changes to
the list that happened within the event would not be reflected outside
of it.
This commit introduces proper handling of three intrinsic size keywords
when used for CSS heights:
- min-content
- max-content
- fit-content
This necessitated a few plumbing changes, since we can't resolve these
values without having access to containing block widths.
This fixes some visual glitches on https://www.supabase.com/ as well
as a number of WPT tests. It also improves the appearance of dialogs.
For example, in the following HTML:
```html
<label>
<input type="radio" name="fruit" value="apple" id="radio1">
<span class="box"></span>
</label>
```
When any descendant of a <label> element is clicked, a "click" event
must be dispatched on the <input> element nested within the <label>, in
addition to the "click" event dispatched on the clicked descendant.
Previously, this behavior was implemented only for text node descendants
by "overriding" the mouse event target using `mouse_event_target()` in
the TextPaintable. This approach was incorrect because it was limited to
text nodes, whereas the behavior should apply to any box. Moreover, the
"click" event for the input control must be dispatched *in addition* to
the event on the clicked element, rather than redirecting it.
The clientX and clientY values are, as per the spec, the offset from
the viewport.
This makes them actually be that and also fixes up the calculations
for offsetX, offsetY, pageX and pageY.
I assume all of these got messed up in some sort of refactor in the
past.
The spec comment from the now-removed
compute_mouse_event_client_offset() function sadly has no convenient
place to be anymore so, for now, it is just gone as well.
Personally, I think it'd make sense to refactor a lot of this file so
that not every mouse event repeats a large chunk of (almost) identical
code. That way there'd be a nice place to put the comment without
repeating it all over the file.
But that is out of the scope of this PR.
Also: I know, offsetX and Y are not fully fixed yet, they still
don't ignore the element's CSS transforms but I am working on that
in a new PR.
This change fixes unhoverable toolbar on https://excalidraw.com/
The problem was that React.js uses setProperty() to add style properties
specified in the "style" attribute in the virtual DOM, and we were
failing to add the CSS variable used to set the "pointer-events" value
to "all".
In line with the ShadowRealm proposal changes in the WebIDL spec:
webidl#1437 and supporting changes in HTML spec.
This is required for ShadowRealms as they have no relevant settings
object on the shadow realm, so fixes a crash in the QueueingStrategy
test in this commit.
These interfaces are exposed on *, meaning it should work for workers
and our newly added shadow realm global object by being stored on the
universal global scope mixin.
