Corresponds to 03ab71775b
I've also split the `Document::has_focus()` method for clarity. Actually
implementing the "has focus steps" turns out to be quite involved so
I've left it for now.
We achieve this by keeping track of all checked pseudo class selectors
in the SelectorEngine code. We also give StyleComputer per-pseudo-class
rule caches.
These callbacks are evaluated synchronously via JS::Call. We do not need
to construct an expensive RootVector container just to immediately
invoke the callbacks.
Stylistically, this also helps indicate where the actual arguments start
at the call sites, by wrapping the arguments in braces.
Check if box has associated layout node is not mentioned in the spec,
but it is required to match behavior of other browsers that do not
invoke intersection observer steps for boxes without layout node.
With this change we save a copy of of scroll state at the time of
recording a display list, instead of actual ScrollState pointer that
could be modifed by the main thread while display list is beings
rasterized on the rendering thread, which leads to a frame painted with
inconsistent scroll state.
Fixes https://github.com/LadybirdBrowser/ladybird/issues/4288
There's a bit of a UTF-8 assumption with this change. But nearly every
caller of these methods were immediately creating a String from the
resulting ByteString anyways.
This fixes the frame-ancestors WPT tests from crashing when an iframe
is blocked from loading. This is because it would get an undefined
location.href from the cross-origin iframe, which causes a crash as it
expects it to be there.
Allows us to avoid DOM node lookup whenever we need to query natural
size of a box during layout.
Makes 3-4% of `Box::preferred_aspect_ratio()` go away from profiles on
www.nyan.cat
With this change we maintain a data structure that maps ids to
corresponding elements. This allows us to avoid tree traversal in
getElementById() in all cases except ones when lookup happens for
unconnected elements.
All abspos boxes are expected to be blockified, so we are certain that
we can ignore non-box elements when collecting abspos nodes for layout.
Fixes a crash caused by an attempt to cast a BreakNode to a Box while
performing abspos layout.
The upcoming generated types will match those for pseudo-classes: A
PseudoElementSelector type, that then holds a PseudoElement enum
defining what it is. That enum will be at the top level in the Web::CSS
namespace.
In order to keep the diffs clearer, this commit renames and moves the
types, and then a following one will replace the handwritten enum with
a generated one.
Instead of bothering the GC heap with a bunch of DOMRect allocations,
we can just pass around CSSPixelRect internally in many cases.
Before this change, we were generating so much DOMRect garbage that
we had to do a garbage collection *every frame* on the Immich demo.
This was due to the large number of intersection observers checked.
We still need to relax way more when idle, but for comparison, before
this change, when doing nothing for 10 seconds on Immich, we'd spend
2.5 seconds updating intersection observers. After this change, we now
spend 600 ms.
This allows us to parse the Content-Security-Policy header and
Referrer-Policy header from navigation responses and actually allow
them to start having an effect.
This isn't actually necessary, since we already invalidate style for the
entire document, and the subsequent style update will discover any
additional layout invalidation needed as well.
12c6ac78e2 with fixed mistake when cache
slot is copied instead of being referenced:
```cpp
auto cache =
box.cached_intrinsic_sizes().min_content_height.ensure(width);
```
while it should've been:
```cpp
auto& cache =
box.cached_intrinsic_sizes().min_content_height.ensure(width);
```
This change moves intrinsic sizes cache from
LayoutState, which is local to current layout run,
to layout nodes, so it could be reused between
layout runs. This optimization is possible because
we can guarantee that these measurements will
remain unchanged unless the style of the element
or any of its descendants changes.
For now, invalidation is implemented simply by
resetting cache on whole ancestors chain once we
figured that element needs layout update.
The case when layout is invalidated by DOM's
structural changes is covered by layout tree
invalidation that drops intrinsic sizes cache
along with layout nodes.
I measured improvement on couple websites:
- Mail list on GMail 28ms -> 6ms
- GitHub large code page 47ms -> 36ms
- Discord chat history 15ms -> 8ms
(Time does not include `commit()`)
This allows us to avoid a full layout tree rebuild after change of
"display" property, which happens frequently in practice. It also
allows us to avoid a full rebuild after DOM node insertion, since
previously, computing styles for newly inserted nodes would trigger a
complete layout tree rebuild.
All necessary invalidations are issued while invalidating animated
style. There is no need to drop display list simply because there are
some animations that might need an update.
We set the page's focused navigable upon mouse-down events from the UI.
However, we neglected to ever clear that focused navigable upon events
such as subsequent page navigations. This left the page with a stale
reference to a no-longer-active navigable. The effect was that any key
events from the UI would not be sent to the new page until either the
reference was collected by GC, or another mouse-down event occurred.
In the test added here, without this fix, the text sent to the input
element would not be received, and the change event would not fire.
This commit implements the main "render blocking" behavior for link
elements, drastically reducing the amount of FOUC (flash of unstyled
content) we subject our users to.
The document will now block rendering until linked style sheets
referenced by parser-created link elements have loaded (or failed).
Note that we don't yet extend the blocking period until "critical
subresources" such as imported style sheets have been downloaded
as well.
This change fixes a bug that can be reproduced with the following steps:
```js
const iframe = document.createElement("iframe");
document.body.appendChild(iframe);
iframe.contentWindow.location.href = ("http://localhost:8080/demo.html");
```
These steps are executed in the following order:
1. Create iframe and schedule session history traversal task that adds
session history entry for the iframe.
2. Generate navigation id for scheduled navigation to
`http://localhost:8080/demo.html`.
3. Execute the scheduled session history traversal task, which adds
session history entry for the iframe.
4. Ooops, navigation to `http://localhost:8080/demo.html` is aborted
because addings SHE for the iframe resets the navigation id.
This change fixes this by delaying all navigations until SHE for a
navigable is created.
Previously, they would stay open for the entire WebContent lifetime,
or until the server closed the connection. This was particularly
noticeable on collaborative websites/games such as
https://jigsawpuzzles.io/, where the user using Ladybird would stick
around even after they had navigated away.
The spec changes seem to mostly be about introducing a TrustedHTML type
which we do not yet support, so we have a couple of FIXMEs.
TrustedTypes::InjectionSink is an attempt at matching the spec, but it's
not entirely clear to me how it should work. I'm sure it'll get
revisited once we start implementing trusted types.
Our own Inspector differs from most other DevTools implementations with
regard to highlighting DOM nodes as you hover elements in the inspected
DOM tree. In other implementations, as you change the hovered node, the
browser will render a box model overlay onto the page for that node. We
currently don't do this; we wait until you click the node, at which
point we both paint the overlay and inspect the node's properties.
This patch does not change that behavior, but separates the IPCs and
internal tracking of inspected nodes to support the standard DevTools
behavior. So the DOM document now stores an inspected node and a
highlighted node. The former is used for features such as "$0" in the
JavaScript console, and the latter is used for the box model overlay.
Our Inspector continues to set these to the same node.
This reduces the number of `.cpp` files that need to be recompiled when
one of the below header files changes as follows:
CSS/ComputedProperties.h: 1113 -> 49
CSS/ComputedValues.h: 1120 -> 209
Analysis of selectors on modern websites shows that the `:hover`
pseudo-class is mostly used in the subject position within relatively
simple selectors like `.a:hover`. This suggests that we could greatly
benefit from segregating them by id/class/tag name, this way reducing
number of selectors tested during hover style invalidation.
With this change, hover invalidation on Discord goes down from 70ms to
3ms on my machine. I also tested GMail and GitHub where this change
shows nice 2x-3x speedup.
This is required to store Content Security Policies, as their
Directives are implemented as subclasses with overridden virtual
functions. Thus, they cannot be stored as generic Directive classes, as
it'll lose the ability to call overridden functions when they are
copied.
If selector does not have any descendant combinators then we know for
sure it won't be filtered out by ancestor filter, which means there is
no need to check for it.
This change makes hover style invalidation go faster on Discord where
with this change we spend 4-5% in `should_reject_with_ancestor_filter()`
instead of 20%.
Before this change, we did the following:
1. Created a bitmap with the matching state for each rule containing
`:hover`.
2. Changed the actively hovered element in the document.
3. Created another bitmap with the matching state for each rule
containing `:hover`.
With this change, we iterate rule by rule and compare the matching
state. This allows us to break early once we find the first rule whose
matching state changes after the hovered element update. Additionally,
this removes the need to allocate a bitmap.
Instead of using `has_pseudo_elements()` that iterates over all pseudo
elements, only check if `::before` or `::after` are present.
Before this change, `has_pseudo_elements()` was 10% of profiles on
Discord while now it's 1-2%.
MediaQueryList will now remember if a state change occurred when
evaluating its match state. This memory can then be used by the document
later on when it's updating all queries, to ensure that we don't forget
to fire at least one change event.
This also required plumbing the system visibility state to initial
about:blank documents, since otherwise they would be stuck in "hidden"
state indefinitely and never evaluate their media queries.
Instead of checking all elements in a document for containment in
`:has()` invalidation set, we could narrow this down to ancestors and
ancestor siblings, like we already do for subject `:has()` invalidation.
This change brings great improvement on GitHub that has selectors with
non-subject `:has()` and sibling combinators (e.g., `.a:has(.b) ~ .c`)
which prior to this change meant style invalidation for whole document.
This commit changes the strategy for updating inherited styles. Instead
of marking all potentially affected nodes during style invalidation, the
decision is now made on-the-fly during style recalculation. Child nodes
will only have their inherited styles recalculated if their parent's
properties have changed.
On Discord this allows to 1000x reduce number of nodes with recalculated
inherited style.
`invalidate_style()` already tries to avoid scheduling invalidation for
`:has()` by checking result of `may_have_has_selectors()`, but it might
still result in unnecessary work because `may_have_has_selectors()`
does not force building of rules cache. This change adds
`have_has_selectors()` that forces building of rules cache and is
invoked in `update_style()` to double-check whether we actually need to
process scheduled `:has()` invalidations.
This allows to skip ~100000 ancestor traversals on this WPT test:
https://wpt.live/html/select/options-length-too-large.html
The current implementation of `:has()` style invalidation is divided
into two cases:
- When used in subject position (e.g., `.a:has(.b)`).
- When in a non-subject position (e.g., `.a > .b:has(.c)`).
This change focuses on improving the first case. For non-subject usage,
we still perform a full tree traversal and invalidate all elements
affected by the `:has()` pseudo-class invalidation set.
We already optimize subject `:has()` invalidations by limiting
invalidated elements to ones that were tested against `has()` selectors
during selector matching. However, selectors like `div:has(.a)`
currently cause every div element in the document to be invalidated.
By modifying the invalidation traversal to consider only ancestor nodes
(and, optionally, their siblings), we can drastically reduce the number
of invalidated elements for broad selectors like the example above.
On Discord, when scrolling through message history, this change allows
to reduce number of invalidated elements from ~1k to ~5.
Before this change, tasks associated with a destroyed document would get
stuck in the task queue forever, since document-associated tasks are not
allowed to run when their document isn't fully active (and destroyed
documents never become fully active again). This caused everything
captured by task callbacks to leak.
We now treat tasks for destroyed documents as runnable immediately,
which gets them out of the queue.
This fixes another massive GC leak on Speedometer.
This is not really a context, but more of a set of parameters for
creating a Parser. So, treat it as such: Rename it to ParsingParams,
and store its values and methods directly in the Parser instead of
keeping the ParsingContext around.
This has a nice side-effect of not including DOM/Document.h everywhere
that needs a Parser.
This ad-hoc code informs the client of a potentially changed page title.
But because we always update the title element (either the SVG or HTML
title) the client was already informed, causing the code to run twice.
Before this change, checking if fast selector matching could be used was
only enabled in style recalculation and hover invalidation. With this
change it's enabled for all callers of SelectorEngine::matches() by
default. This way APIs like `Element.matches()` and `querySelector()`
could take advantage of this optimization.
This is "update document for history step application" but that's too
long for the commit title. :^)
No code changes, just adding more FIXME comments for the new steps.
(And indented step 7's substeps for clarity.)
Corresponds to https://github.com/whatwg/html/pull/10910
Before this change, `m_needs_repaint` was reset in
`Document::record_display_list()` only when the cached display list was
absent. This meant that if the last triggered repaint used the cached
display list, we would keep repainting indefinitely until the display
list was invalidated (We schedule a task that checks if repainting is
required 60/s).
This change also moves `m_needs_repaint` from Document to
TraversableNavigable as we only ever need to repaint a document that
belongs to traversable.
Several interfaces that return a high resolution time require that
time to be coarsened, in order to prevent timing attacks. This
implementation simply reduces the resolution of the returned timestamp
to the minimum values given in the specification. Further work may be
needed to make our implementation more robust to the kind of attacks
that this mechanism is designed to prevent.
...until Document::update_style(). This allows to avoid doing full
document DOM tree traversal on each Node::invalidate_style() call.
Fixes performance regression on wpt.fyi
...in hover style invalidation. Instead, pass down a flag that indicates
all subsequent nodes in tree traversal have to be marked for inherited
style update.
By using ancestor filters some selectors could be early rejected
skipping selector engine invocation. According to my measurements it's
30-80% hover selectors depending on the website.
Instead of allocating 3 vectors with size equal to the number of
elements potentially affected by hover:
- for the elements themselves
- for selector match state of each element before hovered node change
- for selector match state of each element after hovered node change
now we allocate none of them, but mark element for style recalculation
as we traverse the tree.
This currently uses a non spec-compliant property on the Response
object, which represents the time that the Response was created.
Setting this value allows `Performance.timeOrigin` to return a
reasonable value.
Instead of traversing the entire DOM subtrees and marking nodes for
style update, this patch adds a new mechanism where we can mark a
subtree root as "entire subtree needs style update".
A new pass in Document::update_style() then takes care of coalescing
all these invalidations in a single traversal of the DOM.
This shaves *minutes* of loading time off of https://wpt.fyi/ subpages.
Instead, change the APIs from "has :foo" to "may have :foo" and return
true if we don't have a valid rule cache at the moment.
This allows us to defer the rebuilding of the rule cache until a later
time, for the cost of a wider invalidation at the moment.
Do note that if our rule cache is invalid, the whole document has
invalid style anyway! So this is actually always less work. :^)
Knocks ~1 second of loading time off of https://wpt.fyi/
Instead of ignoring any paintable immediately when they're invisible to
hit-testing, consider every candidate and while the most specific
candidate is invisible to hit-testing, traverse up to its parent
paintable.
This more closely reflects the behavior expected when wrapping block
elements inside inline elements, where although the block element might
have `pointer-events: none`, it still becomes part of the hit-test body
of the inline parent.
This makes the following link work as expected:
<a href="https://ladybird.org">
<div style="pointer-events: none">Ladybird</div>
</a>
DOM nodes now have two additional flags:
- Needs layout tree update
- Child needs layout tree update
These work similarly to the needs-style-update flags, but instead signal
the need to rebuild the corresponding part of the layout tree.
When a specific DOM node needs a layout tree update, we try to create
a new subtree starting at that node, and then replace the subtree in the
old layout tree with the newly created subtree.
This required some refactoring in TreeBuilder so that we can skip over
entire subtrees during a tree update.
Note that no partial updates happen yet (as of this commit) since we
always invalidate the full layout tree still. That will change in the
next commit.
The use of this HashMap looks very spooky, but let's at least use
finalize when cleaning them up on destruction to make things slightly
less dangerous looking.
This makes it more convenient to use the 'relvant agent' concept,
instead of the awkward dynamic casts we needed to do for every call
site.
mutation_observers is also changed to hold a GC::Root instead of raw
GC::Ptr. Somehow this was not causing problems before, but trips up CI
after these changes.
Previously, we optimized hover style invalidation to mark for style
updates only those elements that were matched by :hover selectors in the
last style calculation.
This change takes it a step further by invalidating only the elements
where the set of selectors that use :hover changes after hovered element
is modified. The implementation is as follows:
1. Collect all elements whose styles might be affected by a change in
the hovered element.
2. Retrieve a list of all selectors that use :hover.
3. Test each selector against each element and record which selectors
match.
4. Update m_hovered_node to the newly hovered element.
5. Repeat step 3.
6. For each element, compare the previous and current sets of matched
selectors. If they differ, mark the element for style recalculation.
Instead of recalculating styles for all nodes in the common ancestor of
the new and old hovered nodes' subtrees, this change introduces the
following approach:
- While calculating ComputedProperties, a flag is saved if any rule
applied to an element is affected by the hover state during the
execution of SelectorEngine::matches().
- When the hovered element changes, styles are marked for recalculation
only if the flag saved in ComputedProperties indicates that the
element could be affected by the hover state.
Instead of storing all storage objects in static memory, we now
follow the the spec by lazily creating a unique Storage object
on each document object.
Each Storage object now holds a 'proxy' to the underlying backing
storage. For now, this proxy is simply a reference to the backing
object. In the future, it will need to be some type of interface
object that stores on a SQLite database or similar.
Session storage is now correctly stored / tracked as part of the
TraversableNavigable object.
Local storage is still stored in a static map, but eventually this
should be factored into something that is stored at the user agent
level.
This event is fired while both the previous and the current phase are
active.
This prevents this test from timing out:
- css/css-animations/animationevent-types.txt
We can now mark an element as needing an "inherited style update" rather
than a full "style update". This effectively means that the next style
update will visit the element and pull all of its inherited properties
from the relevant ancestor element.
This is now used for descendants of elements with animated style.
We currently (sometimes) copy the observer map to a vector for iteration
to ensure we are not iterating over the map if the callback happens to
remove the observer. But that list was not protected from GC.
This patch ensures we protect that list, and makes all document observer
notifiers protected from removal during iteration.
There are essentially 3 URL parsing AOs defined by the spec:
1. Parse a URL
2. Encoding parse a URL
3. Encoding parse a URL and serialize the result
Further, these are replicated between the Document and the ESO.
This patch defines these methods in accordance with the spec and updates
existing users to invoke the correct method. In places where the correct
method is ambiguous, we use the encoding parser to preserve existing ad-
hoc behavior.
The DOM spec defines what it means for an element to be an "editing
host", and the Editing spec does the same for the "editable" concept.
Replace our `Node::is_editable()` implementation with these
spec-compliant algorithms.
An editing host is an element that has the properties to make its
contents effectively editable. Editable elements are descendants of an
editing host. Concepts like the inheritable contenteditable attribute
are propagated through the editable algorithm.
These variables are all captured in queued events or other event loop
tasks, but are all guarded by event loop spins later in the function.
The IGNORE_USE_IN_ESCAPING_LAMBDA will soon be required for all locals
that are captured by ref in GC::Function as well as AK::Function.
To facilitate the implementation of "delete" and all associated
algorithms, split off this piece of `Document` into a separate
directory.
This sets up the infrastructure for arbitrary commands to be supported.
This lets us move a few Host-related functions (like serialization and
checks for what the Host is) into Host instead of having them dotted
around the codebase.
For now, the interface is still very Variant-like, to avoid having to
change quite so much in one go.
An inopportune garbage collection may cause collected `ResizeObserver`s
to unregister themselves from `m_resize_observers` while we are
iterating over it, resulting in a use-after-free.
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.
Resulting in a massive rename across almost everywhere! Alongside the
namespace change, we now have the following names:
* JS::NonnullGCPtr -> GC::Ref
* JS::GCPtr -> GC::Ptr
* JS::HeapFunction -> GC::Function
* JS::CellImpl -> GC::Cell
* JS::Handle -> GC::Root
Now that the heap has no knowledge about a JavaScript realm and is
purely for managing the memory of the heap, it does not make sense
to name this function to say that it is a non-realm variant.
The main motivation behind this is to remove JS specifics of the Realm
from the implementation of the Heap.
As a side effect of this change, this is a bit nicer to read than the
previous approach, and in my opinion, also makes it a little more clear
that this method is specific to a JavaScript Realm.
Instead of doing a forced layout synchronously whenever an element's
style is changed, use a zero-timer to do the forced relayout on next
event loop iteration.
This effectively coalesces a lot of layouts and makes many pages such
as GitHub spend way less time doing redundant layout work.
The StyleResolver can find the specified CSS values for the parent
element via the DOM. Forcing everyone to locate specified values for
their parent was completely unnecessary.
Another step towards not having to carry the full specified style with
us everywhere. This isn't the ideal final layout, since we're mixing
computed and used values a bit randomly here, but one step at a time.
