On macOS, CMake incorrectly tries to add and/or remove rpaths from files
that it has already processed when it performs installation. Setting the
rpaths during the build process ensures that they are only set once, and
as a bonus, makes installation slightly more performant.
Fixes#10055.
Currently, we only look at the relative path `./{helper}/{helper}`,
which fails if the working directory is not the same as the directory
where the ladybird binary lives.
This will make it a lot easier to understand what went wrong, especially
when the failure occurs on CI but not at home.
And of course, use LibDiff to generate the diff! :^)
Instead of starting a new headless-browser for every layout & text test,
headless-browser now gets a mode where it runs all the tests in a single
process.
This is massively faster on my machine, taking a full LibWeb test run
from 14 seconds to less than 1 second. Hopefully it will be a similarly
awesome improvement on CI where it has been soaking up more and more
time lately. :^)
This looks a lot more "at home" than usual pixel art logo on
non-SerenityOS systems. :^)
Also, stop using site favicons as the app icon as that made it
annoyingly hard to find Ladybird in task switchers sometimes.
This allows us to create "text tests" in addition to "layout tests".
Text tests work the same as layout tests, but dump the document content
as text and exit upon receiving the window "load" event.
Previously, calling `.right()` on a `Gfx::Rect` would return the last
column's coordinate still inside the rectangle, or `left + width - 1`.
This is called 'endpoint inclusive' and does not make a lot of sense for
`Gfx::Rect<float>` where a rectangle of width 5 at position (0, 0) would
return 4 as its right side. This same problem exists for `.bottom()`.
This changes `Gfx::Rect` to be endpoint exclusive, which gives us the
nice property that `width = right - left` and `height = bottom - top`.
It enables us to treat `Gfx::Rect<int>` and `Gfx::Rect<float>` exactly
the same.
All users of `Gfx::Rect` have been updated accordingly.
The goal here is to reduce the amount of WebContent client APIs that are
duplicated across every ViewImplementation. Across our three browsers,
we currently:
Ladybird - Mix some AK::Function callbacks and Qt signals to notify
tabs of WebContent events.
Browser - Use only AK::Function callbacks.
headless-browser - Drop most events on the floor.
Instead, let's only use AK::Function callbacks across all three browsers
to propagate events to tabs. This allows us to invoke those callbacks
directly from LibWebView instead of all three browsers needing to define
a trivial `if (callback) callback();` override of a LibWebView virtual
function. For headless-browser, we can simply not set these callbacks.
As a first pass, this only converts WebContent events that are trivial
to this approach. That is, events that were simply passed onto the tab
or handled without much fuss.
This is to match Browser, where ownership of all "subwidgets" is placed
on the tab as well. This further lets us align the web view callbacks to
match Browser's OOPWV as well, which will later let us move them into
the base LibWebView class.
Note that the real implementations of these functions are:
notify_server_did_output_js_console_message
notify_server_did_get_js_console_messages
Which have the same method bodies as these unused variants.
The implementations of handle_web_content_process_crash and
take_screenshot are exactly the same across Browser and Ladybird. Let's
reduce some code duplication and move them to LibWebView.
Previously, we were doing mapToGlobal() via the Tab widget, but the
widget position was actually relative to the WebContentView. This
meant context menus appeared slightly vertically offset from where
you clicked.
This just sets up the IPC to notify the browser process of context menu
requests on video elements. The IPC contains a few pieces of information
about the state of the video element.
While resizing, we now pad the shared bitmap allocations with 256 pixels
extra in both axes. This avoids churning through huge allocations for
every single resize step.
We also don't reallocate at all when making the window smaller.
3 seconds after the user stops resizing, we resize the backing stores
again so they fit the window perfectly.
This fixes an unpleasant visual glitch when resizing the window.
When the user makes our QAbstractScrollArea larger, the scroll bars can
end up with negative values, which we were happily forwarding to the
WebContent process and asking it to paint the whole page at an offset.
This commit adds the common actions you'd expect to the Ladybird context
menu, arranged like so:
┌──────────────────────────────┐
│ Go Back Alt+Left │
│ Go Forward Alt+Right │
│ Reload Ctrl+R │
│ ──────────────────────────── │
│ Copy Ctrl+C │
│ Select All Ctrl+A │
│ ──────────────────────────── │
│ View Source Ctrl+U │
│ Inspect Element │
└──────────────────────────────┘
The asynchronous query execution keeps causing bugs with document.cookie
so let's make the SQL database backend default off until we can trust it
to do what we need.
This adds "Inspect Element" (currently the only entry) to the context
menu for the page, which will do what you expect (most of the time),
and bring up the Inspector with hovered element selected.
We should only rely on LibGfx to decode images for us, if LibGfx
can't decode an image that should be motivation to improve LibGfx,
not hidden by Qt picking up the slack :^)
While it's nice to see <img src="foo.svg"> suddenly work in Ladybird
after linking with the Qt SVG module, this is cheating.
We should implement SVG-as-image ourselves instead of relying on 3rd
party code to do it. :^)
For some reason, this was causing incomplete HTTP loads in some cases.
As an example, we would only load half of the "Ahem" CSS font from the
wpt.live server when running Acid3.
I only enabled pipelining in the first place because I assumed it would
be a performance boost, but it appears to do more than that.
I suppose there's a reason it's off by default (and most Qt API users
don't bother enabling it.)
We now load SVG icons (via the Qt resource system) and render them into
a QIcon (with normal and disabled variants) using system colors.
We also re-render them if the system color theme changes.
This instantly makes Ladybird look less foreign on my Linux box.
I drew the icons myself, and they could definitely be more optimized,
but this was my first time using Inkscape. :^)
The WebContent process behaves a bit awkwardly on macOS. When we launch
the process, we have to create a QGuiApplication to access system fonts.
But on macOS, doing so creates an entry in the Dock, and also causes the
WebContent to be focused. So if you enter cmd+Q without first focusing
the Ladybird GUI, WebContent is closed, while the Ladybird process keeps
running.
This shouldn't have been moved to EventLoopManager, as the manager is
global and one-per-process, and the implementation is one-per-loop.
This makes cross-thread event posting work again, and unbreaks
SoundPlayer (and probably other things as well.)
We currently query Qt for system fonts using QFont::setStyleHint(). The
docs from Qt have the following note regarding this API on X11:
Qt does not support style hints on X11 since this information is not
provided by the window system.
This prevents any monospace font from working on X11 systems. For now,
work around this by specifying the font-family for fonts which Qt has
listed as mapping to a CSS generic font-family.
Things such as timers and notifiers aren't specific to one instance of
Core::EventLoop, so let's not tie them down to EventLoopImplementation.
Instead, move those APIs + signals & a few other things to a new
EventLoopManager interface. EventLoopManager also knows how to create a
new EventLoopImplementation object.
Using QEventLoop works for everything but it breaks *one* little feature
that we care about: automatically quitting the app when all windows have
been closed.
That only works if you drive the outermost main event loop with a
QCoreApplication instead of a QEventLoop. This is unfortunate, as it
complicates our API a little bit, but I'm sure we can think of a way to
make this nicer someday.
In order for QCoreApplication::exec() to process our own
ThreadEventQueue, we now have a zero-timer that we kick whenever new
events are posted to the thread queue.
Now that the Core::EventLoop is driven by a QEventLoop in Ladybird,
we don't need to patch LibWeb with Web::Platform plugins.
This patch removes EventLoopPluginQt and TimerQt.
Note that we can't just replace the Web::Platform abstractions with
LibCore stuff immediately, since the Web::Platform APIs use
JS::SafeFunction for callbacks.
This patch adds EventLoopImplementationQt which is a full replacement
for the Core::EventLoopImplementationUnix that uses Qt's event loop
as a backend instead.
This means that Core::Timer, Core::Notifier, and Core::Event delivery
are all driven by Qt primitives in the Ladybird UI and WC processes.
Not a single client of this API actually used the event mask feature to
listen for readability AND writability.
Let's simplify the API and have only one hook: on_activation.
This aligns the Ladybird console implementation with the Browser console
a bit more, which uses OutOfProcessWebView for rendering console output.
This allows us to style the console output to try and match the system
theme.
Using a WebContentView is simpler than trying to style the old QTextEdit
widget, as the console output is HTML with built-in "-libweb-palette-*"
colors. These will override any color we set on the QTextEdit widget.
We never clear content filters on either end of the Browser-WebContent
IPC connection. So when the filters change, we re-append all filters to
the Vector holding them. This incidentally makes it impossible to remove
a filter.
Change both sides to clear their filter lists when receiving a new set
of filters.
This adds a -P option to run Ladybird under callgrind. It starts with
instrumentation disabled. To start capturing a profile (once Ladybird
has launched) run `callgrind_control -i on` and to stop it again run
`callgrind_control -i off`.
P.s. This is pretty much stolen from Andreas (and is based on the patch
everyone [that wants a profile] have been manually applying).
We had a mismatch in the GUI Identifier property, causing warnings in
Xcode. It was also missing the Product Identifier Xcode property on
ladybird itself, causing another warning.
Copy all our helper processes to the ladybird.app bundle directory so
that they can be found by ``open ladybird.app`` and the Xcode debugger.
For the future, we should look in ../Resources for resources on macOS.
Copying resources to that directory requires more CMake-fu.
Fix the problem that `cmake --build Build/ladybird` started
failing with:
fatal error: 'WebContent/WebDriverConnection.h' file not found
after 11fe34ce0f
Generate handle UUID for top-level context that is going to
run in created WebContent process and sent it over IPC.
Co-authored-by: Timothy Flynn <trflynn89@pm.me>
Currently, on Serenity, we connect to WebDriver from the browser-side of
the WebContent connection for both Browser and headless-browser.
On Lagom, we connect from within the WebContent process itself, signaled
by a command line flag.
This patch changes Lagom browsers to connect to WebDriver the same way
that Serenity browsers do. This will ensure we can do other initializers
in the same order across all platforms and browsers.
There isn't a 1:1 equivalent for all ColorRoles between Qt and LibGfx,
but we can at least make an effort to translate the various QPalette
preferred colors.
This makes text selection look a lot more "native" in Ladybird. :^)
This patch replaces the usage of QPalette::PlaceholderText with
QPalette::Text with opacity reduced to roughly 50%. This fixes the non
highlighted spans having an extremely low contrast compared to the
background in dark mode.
LibGUI and WebDriver (read: JSON) API boundaries use DeprecatedString,
so that is as far as these changes can reach.
The one change which isn't just a DeprecatedString to String replacement
is handling the "null" prompt response. We previously checked for the
null DeprecatedString, whereas we now represent this as an empty
Optional<String>.
headless-browser currently uses its own PageClient to load web pages
in-process. Due to this, it also needs to set up a whole bunch of other
objects needed to run LibWeb, e.g. image decoders, request servers, etc.
This changes headless-browser to instead implement a WebView to launch
WebContent out-of-process. This implementation is almost entirely empty,
but can be filled in as-needed. For example, we may want to print
JavaScript console messages.
We should be able to run this locally, as long as ENABLE_LAGOM_LADYBIRD
is true, or if building ladybird from the ladybird source directory.
This removes a special case from the Lagom CI yml file.
The QNetworkReply::NetworkError enum mixes all kinds of errors into one
enum, HTTP errors, network errors, proxy errors, etc.
Instead of caring about it, we now say that HTTP requests were
successful if their response has any HTTP status code attached.
This allows LibWeb to display error pages when using Qt networking.
Instead of just calling JS::Value::to_string_without_side_effects() when
printing values to the console, have all the console clients use
the same JS::Print that the REPL does to print values.
This method leaves some things to be desired as far as OOM hardening
goes, however. We should be able to create a String in a way that
doesn't OOM on failure so hard.
Otherwise Qt would not find the wayland plugin it is instructed to use
via QT_QPA_PLATFORM, and would fall back to the second option, xcb,
which looks rather sad in a modern Wayland environment :^)
This feels like something that should be addressed upstream in nixpkgs
eventually.
This adds support for WebSocket subprotocols to WebSocket DOM
objects, with some necessary plumbing to LibWebSocket and its
clients.
See the associated pull request for how this was tested.
This allows the WebDriver to take advantage of the common helper process
spawning code when launching ladybird, and to not assume a particular
directory layout.
Use a list of executables to make sure that we don't miss any of the
applications used by Ladybird and its friends like WebDriver, and make
sure to install include all executables and their runtime dependencies.
Use the new get_paths_for_helper_process method in Ladybird to query
Qt for the runtime path of the current executable as well as the build
directory paths.
Previously when there was a very long url that spanned outside of the
address bar, the text shown would be the one starting from the very end
of the url instead of from the beginning, so you would be seeing the
query parameters for example, instead of the domain.
These changes will prevent duplication of code later when there will
also be the possibility to navigate to the homepage url as defined in
the settings on startup.
Don't use _exit() - this is a forceful exit that will bypass all exit
handlers. This includes AddressSanitizer, and will prevent ASan from
exiting the app with a fatal error code.
`Stream` will be qualified as `AK::Stream` until we remove the
`Core::Stream` namespace. `IODevice` now reuses the `SeekMode` that is
defined by `SeekableStream`, since defining its own would require us to
qualify it with `AK::SeekMode` everywhere.
Note that as of this commit, there aren't any such throwers, and the
call site in Heap::allocate will drop exceptions on the floor. This
commit only serves to change the declaration of the overrides, make sure
they return an empty value, and to propagate OOM errors frm their base
initialize invocations.
This will avoid loading starting about:blank page in places when we know
exactly what we want to load.
The opening in background part might be useful for future things like
file drops and right-click open in new tab.
Because TGA images don't have magic bytes as a signature to be detected,
instead assume a sequence of ReadonlyBytes is a possible TGA image only
if we are given a path so we could check the extension of the file and
see if it's a TGA image.
When we know the path of the file being loaded, we will try to first
check its extension, and only if there's no match to a known decoder,
based on simple extension lookup, then we would probe for other formats
as usual with the normal sniffing method.
This starts moving code equally shared between the OOPWV and Ladybird
WebContentView implementations to WebView::ViewImplementation, beginning
with the client state.
On many keyboards, Ctrl++ is actually Ctrl+Shift+=, and Ctrl+= makes
more sense as it's symmetric with Ctrl+-.
Both Firefox and Chrome already support this alternate shortcut,
so let's be nice and support it in Ladybird as well. :^)
This makes vector fonts load on macOS, where /usr/share/fonts doesn't
exist and Ladybird would only load the bitmap fonts from ./res/fonts
in the SerenityOS resource root directory.
Additionally, fonts in {/usr/share/local,~/.local}/fonts are now loaded
on Linux.
This patch also stubs out notify_server_did_get_accessiblity_tree in
ladybird since ViewImplementation now has it. However, this feature
is still immature, so just stubbing out in ladybird for now. Once we
have more robust support in Serenity (namely ARIA properties/state
and accessible names and descriptions) we can port this
functionality over.
This has been broken since the switch to the multiprocess architecture
(and even before then was very limited).
This restores the previous functionally and also implements the ability
to inspect individual elements (by selecting them in the tree view).
The inspector also now correctly updates when navigating between pages.
This allows us to use standard Serenity IPC infrastructure rather than
manually creating FD-passing sockets. This also lets us use Serenity's
WebDriver Session class, removing the copy previously used in Ladybird.
This ensures any changes to Session in the future will be picked up by
Ladybird for free.
Rather than manually launching the SQLServer process, use SQLClient's
new functionality to launch the server just once for all Ladybird
instances. Quit the SQLServer process when it no longer has any
connected clients.
When Ladybird exits, SQLServer can get stuck spinning at 100% CPU after
the socket connection is closed. This changes the client to quit the
event loop when that disconnect happens to ensure that SQLServer is
properly destroyed.
This adds a SQLServer binary for Ladybird to make use of Serenity's SQL
implementation. This has to use the same IPC socket handling that was
used to make WebContent and WebDriver work out-of-process.
Unlike Serenity, Ladybird creates a new SQLServer instance for each
Ladybird instance. In the future, we should try to make sure there is
only one SQLServer instance at a time, and allow multiple Ladybird
instances to communicate with it.
