mirrors/ladybird
0
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2025-04-30 00:09:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

LibWeb: Batch processing of successfully downloaded images

Browse source 
Before this change, we would process each image as it finished
downloading. This often led to a situation where we'd decode 1 image,
schedule a layout, do the layout, then decode another image, schedule
a layout, do the layout, etc. Basically decoding and layouts would get
interleaved even though we had multiple images fetched and ready for
decoding.

This patch adds a simple BatchingDispatcher thingy that HTMLImageElement
uses to batch the handling of successful fetches.

With this, the number of layouts while loading https://shopify.com/ goes
from 48 to 6, and the page loads noticeably faster. :^)
This commit is contained in:
Andreas Kling 2023-07-27 15:05:42 +02:00
parent fedd087546
commit a9aecbbd6f
Notes: sideshowbarker 2024-07-17 07:25:39 +09:00
1 changed files with 66 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

96
Userland/Libraries/LibWeb/HTML/HTMLImageElement.cpp
View file

@ -262,6 +262,40 @@ bool HTMLImageElement::uses_srcset_or_picture() const
return has_attribute(HTML::AttributeNames::srcset) || (parent() && is<HTMLPictureElement>(*parent()));
}
// We batch handling of successfully fetched images to avoid interleaving 1 image, 1 layout, 1 image, 1 layout, etc.
// The processing timer is 1ms instead of 0ms, since layout is driven by a 0ms timer, and if we use 0ms here,
// the event loop will process them in insertion order. This is a bit of a hack, but it works.
struct BatchingDispatcher {
public:
BatchingDispatcher()
: m_timer(Core::Timer::create_single_shot(1, [this] { process(); }).release_value_but_fixme_should_propagate_errors())
{
}
void enqueue(JS::SafeFunction<void()> callback)
{
m_queue.append(move(callback));
m_timer->restart();
}
private:
void process()
{
auto queue = move(m_queue);
for (auto& callback : queue)
callback();
}
NonnullRefPtr<Core::Timer> m_timer;
Vector<JS::SafeFunction<void()>> m_queue;
};
static BatchingDispatcher& batching_dispatcher()
{
static BatchingDispatcher dispatcher;
return dispatcher;
}
// https://html.spec.whatwg.org/multipage/images.html#update-the-image-data
ErrorOr<void> HTMLImageElement::update_the_image_data(bool restart_animations, bool maybe_omit_events)
{
@ -473,44 +507,46 @@ after_step_7:
image_request->add_callbacks(
[this, image_request, maybe_omit_events, url_string, previous_url]() {
VERIFY(image_request->shared_image_request());
auto image_data = image_request->shared_image_request()->image_data();
image_request->set_image_data(image_data);
batching_dispatcher().enqueue([this, image_request, maybe_omit_events, url_string, previous_url] {
VERIFY(image_request->shared_image_request());
auto image_data = image_request->shared_image_request()->image_data();
image_request->set_image_data(image_data);
ListOfAvailableImages::Key key;
key.url = url_string;
key.mode = m_cors_setting;
key.origin = document().origin();
ListOfAvailableImages::Key key;
key.url = url_string;
key.mode = m_cors_setting;
key.origin = document().origin();
// 1. If image request is the pending request, abort the image request for the current request,
// upgrade the pending request to the current request
// and prepare image request for presentation given the img element.
if (image_request == m_pending_request) {
abort_the_image_request(realm(), m_current_request);
upgrade_pending_request_to_current_request();
image_request->prepare_for_presentation(*this);
}
// 1. If image request is the pending request, abort the image request for the current request,
// upgrade the pending request to the current request
// and prepare image request for presentation given the img element.
if (image_request == m_pending_request) {
abort_the_image_request(realm(), m_current_request);
upgrade_pending_request_to_current_request();
image_request->prepare_for_presentation(*this);
}
// 2. Set image request to the completely available state.
image_request->set_state(ImageRequest::State::CompletelyAvailable);
// 2. Set image request to the completely available state.
image_request->set_state(ImageRequest::State::CompletelyAvailable);
// 3. Add the image to the list of available images using the key key, with the ignore higher-layer caching flag set.
document().list_of_available_images().add(key, *image_data, true).release_value_but_fixme_should_propagate_errors();
// 3. Add the image to the list of available images using the key key, with the ignore higher-layer caching flag set.
document().list_of_available_images().add(key, *image_data, true).release_value_but_fixme_should_propagate_errors();
// 4. If maybe omit events is not set or previousURL is not equal to urlString, then fire an event named load at the img element.
if (!maybe_omit_events || previous_url != url_string)
dispatch_event(DOM::Event::create(realm(), HTML::EventNames::load).release_value_but_fixme_should_propagate_errors());
// 4. If maybe omit events is not set or previousURL is not equal to urlString, then fire an event named load at the img element.
if (!maybe_omit_events || previous_url != url_string)
dispatch_event(DOM::Event::create(realm(), HTML::EventNames::load).release_value_but_fixme_should_propagate_errors());
set_needs_style_update(true);
document().set_needs_layout();
set_needs_style_update(true);
document().set_needs_layout();
if (image_data->is_animated() && image_data->frame_count() > 1) {
m_current_frame_index = 0;
m_animation_timer->set_interval(image_data->frame_duration(0));
m_animation_timer->start();
}
if (image_data->is_animated() && image_data->frame_count() > 1) {
m_current_frame_index = 0;
m_animation_timer->set_interval(image_data->frame_duration(0));
m_animation_timer->start();
}
m_load_event_delayer.clear();
m_load_event_delayer.clear();
});
},
[this, image_request, maybe_omit_events, url_string, previous_url, selected_source]() {
// The image data is not in a supported file format;