C++20 can automatically synthesize `operator!=` from `operator==`, so
there is no point in writing such functions by hand if all they do is
call through to `operator==`.
This fixes a compile error with compilers that implement P2468 (Clang
16 currently). This paper restores the C++17 behavior that if both
`T::operator==(U)` and `T::operator!=(U)` exist, `U == T` won't be
rewritten in reverse to call `T::operator==(U)`. Removing `!=` operators
makes the rewriting possible again.
See https://reviews.llvm.org/D134529#3853062
When parsing relative URLs, we have to check the first <base href> in
tree order (if one is available). This was getting *very* costly on
large DOMs with many relative urls.
This patch avoids all that repeated traversal by letting Document cache
the first <base href> and invalidating the cache whenever a <base>
element is added/removed/edited in the DOM.
The browser was stuck doing this for a *very* long time when loading
the ECMA-262 spec, and this removes that problem entirely.
HTML template elements don't affect rendering, so invalidating the
entire document's layout after poking into a <template> was a huge waste
of work on template-heavy pages.
There's no need to force a synchronous relayout after the viewport has
been resized. By making it lazy, we might be able to coalesce it with
other layout work.
Like for attribute changes, we now only invalidate the insertion parent
and all of its descendants. Again, this is very aggressive, but also
way less than doing the entire document.
Once we implement the CSS :has() selector, we'll need to become more
sophisticated about invalidation.
We now only invalidate the style of the context element and all of its
descendants. It's still very aggressive, but much less than before.
Note that this will need to become a lot smarter once we implement the
CSS :has() selector.
Following another abort signal basically means to make an abort signal
abort when another abort signal is aborted, unless the following signal
is already aborted.
It's potentially unsafe to access `m_root` in the destructor since it
may have been swept, so move unregistration of the NodeIterator into a
GC finalizer instead.
Now that the layout tree is also GC-allocated, we can't be messing with
it from the DOM::Node destructor. Move everything to a GC finalizer
so we know it runs before the GC sweep phase.
Since SafeFunction strongly protects all of its captures, we can't
capture `this` when activating an event handler since that creates a
reference cycle and we end up leaking the entire world.
This removes a set of complex reference cycles between DOM, layout tree
and browsing context.
It also makes lifetimes much easier to reason about, as the DOM and
layout trees are now free to keep each other alive.
(And BrowsingContextGroup had to come along for the ride as well.)
This solves a number of nasty reference cycles between browsing
contexts, history items, and their documents.
This prevents a reference cycle between a HTMLParser opened via
document.open() and the document. It was one of many things keeping
some documents alive indefinitely.
When a new document becomes the active document of a browsing context,
we now notify the old document, allowing it to tear down its layout
tree. In the future, there might be more cleanups we'd like to do here.
Print exceptions passed to `HTML::report_exception` in the JS console
Refactored `ExceptionReporter`: in order to report exception now
you need to pass the relevant realm in it. For passed `JS::Value`
we now create `JS::Error` object to print value as the error message.
The intent is to use these to autogenerate prototype declarations for
Window and WorkerGlobalScope classes.
And the spec links are just nice to have :^)
We parse the arguments that come in, but since we don't yet track
scrollable overflow, we can't do the full "scroll an element into view"
algorithm. For now, we just call out to the PageClient and ask it to
bring the nearest principal box into the visible viewport.
I've left a FIXME here about populating the events with mouse
coordinates, button states, etc. We also need to verify that the
dispatch order either doesn't matter or at least match other engines.
`.split_view(Infra::ASCII_WHITESPACE)` tries to split the string view on
the string "\t\n\f\r " (not any of the individual characters of that
string).
The correct way to split this string views here is
`.split_view_if(Infra::is_ascii_whitespace)`, this is a little
inconsistent with String, so probably should be addressed.
This is a big and messy change, and here's the gist:
- AvaliableSpace is now 2x AvailableSize (width and height)
- Layout algorithms are redesigned around the idea of available space
- When doing layout across nested formatting contexts, the parent
context tells the child context how much space is available for the
child's root box in both axes.
- "Available space" replaces "containing block width" in most places.
- The width and height in a box's UsedValues are considered to be
definite after they're assigned to. Marking something as having
definite size is no longer a separate step,
This probably introduces various regressions, but the big win here is
that our layout system now works with available space, just like the
specs are written. Fixing issues will be much easier going forward,
since you don't need to do nearly as much conversion from "spec logic"
to "LibWeb logic" as you previously did.
