| Age | Commit message (Collapse) | Author |
|---|
|
This is a normative change in the Temporal spec.
See: https://github.com/tc39/proposal-temporal/commit/6fa5b9d
|
|
When opening canvas-rotate.html on my host machine, I noticed that
the rotation was going the other way.. :^)
|
|
These are editorial changes in the Intl spec. See:
https://github.com/tc39/ecma402/commit/7d0326c
https://github.com/tc39/ecma402/commit/05a299b
https://github.com/tc39/ecma402/commit/8c24ea7
https://github.com/tc39/ecma402/commit/fd8dea9
|
|
This FIXME was addressed in 0975eba7241ff44fcfc5f3be074a09b353ebae5f.
|
|
|
|
Instead of downloading nearly 20 files individually, we can download a
single .zip file similar to how we download a single CLDR .zip. This is
to reduce the number of connections/downloads to/from unicode.org.
|
|
Having bogus values here when we just initialize the thread state with a
process can lead to all sorts of bad things down the line, like infinite
draws.
|
|
In the process model we check the thread with tid=pid to figure out the
main thread of a process. This is used to construct the process view
tree with non-main threads listed as children of the process row.
However, there are sometimes circumstances where there is no main
thread, even though the process should have been removed from the
internal list by then. As a safe fallback, let's default to an invalid
model index if we can't figure out what the main thread of a process is.
|
|
BrowsingContext shouldn't be scrolling itself, instead it has to update
the layout (to ensure that we have current document metrics, and then
ask the PageClient nicely to scroll it.
This fixes an issue where BrowsingContext sometimes believed itself to
be scrolled, but OOPWV had a different idea.
|
|
This was a mixin class that allowed sharing a set of hooks between
InProcessWebView and OutOfProcessWebView. Now that there is only OOPWV,
we don't need the mixin.
|
|
Let's simplify our offering by only having the OutOfProcessWebView.
|
|
This was built on Web::InProcessWebView which is going to be removed.
Since this feature wasn't really used or maintained, let's just remove
it for now, and it can be resurrected on top of OutOfProcessWebView if
someone finds it useful enough to do that work.
|
|
This was not used or maintained, and relied on InProcessWebView which we
need to get rid of.
|
|
|
|
Use our existing helper function for parsing a HTML string and opening
it as the main content of the attached browsing context.
|
|
Previously, the paste action was always enabled and always assumed that
anything was selected, which led to a crash by clicking the paste action
right after the application startup.
This patch will automatically enable/disable the paste action depending
on whether a selection exists (it usually does, except on the app launch
and after adding a new tab) and if the clipboard mime type is a text/
group.
So no, you can't paste an image into the app anymore, even though this
mostly froze the app before...
|
|
With this we can run following script with no errors:
```sh
for /usr/lib/*.so {
disasm "$it" > /dev/zero
}
```
|
|
This is a huge FIXME right now, and should either be delegated to
SoftVPU or handled in these instructions.
|
|
This allows disassembly of binaries with SSE2 instructions in them.
SSE2 also extends all MMX instructions without affecting the mnemonic,
therefore these are just directed to the same function for now.
The UserspaceEmulator does not know this as of
this commit.
|
|
SSE2 expands on the same opcodes as MMX, so we have to mutate on prefix
here.
|
|
|
|
This was annotated the wrong way around.
|
|
When we lock a mutex, eventually `Thread::block` is invoked which could
in turn invoke `Process::big_lock().restore_exclusive_lock()`. This
would then try to add the current thread to a different blocked thread
list then the one in use for the original mutex being locked, and
because it's an intrusive list, the thread is removed from its original
list during the `.append()`. When the original mutex eventually
unblocks, we no longer have the thread in the intrusive blocked threads
list and we panic.
Solve this by making the big lock mutex special and giving it its own
blocked thread list. Because the process big lock is temporary and is
being actively removed from e.g. syscalls, it's a matter of time before
we can also remove the fix introduced by this commit.
Fixes issue #9401.
|
|
Change "compute" to "calculate" to make clearer that this is unrelated
to the CSS "computed height" concept.
Change "intrinsic" to "auto" to make clearer that this is not the same
as the intrinsic min-content and max-content sizing calculations.
|
|
|
|
This is the name that is used for every other collection type so let's
be consistent.
|
|
This prevents a crash when the inspected process closes the socket (i.e.
when the app is exited). This crash was caused by the InspectableProcess
removing itself from the global process table within a callback Function
that is stored as part of the InspectableProcess.
|
|
This is a hack until persistent model indices work.
|
|
This shows all non-main threads as children of the process they belong
to. We also show the TID as that is important to distinguish the
different threads in one process.
Fixes #65
:skeleyak:
|
|
This would previously cause silly things like [GUI::Icon] to appear if a
non-textlike column was used as the tree column (like, in this example,
an icon column). Let's just not write anything instead.
|
|
This will cause trouble later when the row is not enough to identify a
selection.
|
|
This will look much nicer once we enable that column.
|
|
This has safer fallbacks than toggle_index, because we want to be able
to call it on indices that don't have children.
|
|
This was using internal_data beforehand, which relies on the internal
data to be distinct for different model indices. That's not the case for
example for SortingProxyModel. Using the model index directly makes tree
expansion work properly when using a tree table widget with a
SortingProxyModel.
|
|
|
|
|
|
This was never properly delegated to the source model, so the tree
column would just always be zero.
|
|
|
|
When we want to use the find_first_index that base Vector provides, we
need to provide an element of the real contained type. That's impossible
for OwnPtr, however, and even with RefPtr there might be instances where
we have a raw reference to the object we want to find, but no smart
pointer. Therefore, overloading this function (with an identical body,
the magic is done by the find_index templatization) with `T const&` as a
parameter allows there use cases.
|
|
This was used before by SystemMonitor but not actually registered.
|
|
This was declared but not defined. :yakfusion:
|
|
When running the min-content and max-content sizing algorithms and the
target box creates a flex formatting context, we don't need to measure
its children.
FFC has already assigned the content_width and content_height values,
so we just need to pick those up from the container's formatting state.
|
|
|
|
This adds the ability to hide certain options from certain help texts.
`--complete` is always hidden, whereas `--help` and `--version` are
hidden from Markdown help text only.
Note that in all cases these three options are hidden from the short
usage line.
|
|
|
|
|
|
This loads libsoftgpu.so during GLContext creation and instantiates the
device class which is then passed into the GLContext constructor.
|
|
This adds a virtual base class for GPU devices located in LibGPU.
The OpenGL context now only talks to this device agnostic interface.
Currently the device interface is simply a copy of the existing SoftGPU
interface to get things going :^)
|
|
This introduces a new device independent base class for Images in LibGPU
that also keeps track of the device from which it was created in order
to prevent assigning images across devices.
|
|
|
