To transparently support multi-frame images, all decoder plugins have
already been updated to return their only bitmap for frame(0).
This patch completes the remaining cleanup work by removing the
ImageDecoder::bitmap() API and having all clients call frame() instead.
Previously, ImageDecoder::create() would return a NonnullRefPtr and
could not "fail", although the returned decoder may be "invalid" which
you then had to check anyway.
The new interface looks like this:
static RefPtr<Gfx::ImageDecoder> try_create(ReadonlyBytes);
This simplifies ImageDecoder since it no longer has to worry about its
validity. Client code gets slightly clearer as well.
This class now contains all the fun bits about laying out text in a
rect. It will handle line wrapping at a certain width, cutting off lines
that don't fit the given rect, and handling text elision.
Painter::draw_text now internally uses this.
Future work here would be not laying out text twice (once actually
preparing the lines to be rendered and once to get the bounding box),
and possibly adding left elision if necessary.
Additionally, this commit makes the Utf32View versions of
Painter::draw_text convert to Utf8View internally. The intention is to
completely remove those versions, but they're kept at the moment to keep
the scope of this PR small.
If this is not ignored by default, then one can move the mouse fast
enough to skip hovering over a parent element and hover directly over
the child, causing the parent to never receive an Enter event.
Without this change, when the mouse starts hovering over a child widget,
the parent would receive a Leave event despite the parent widget not
losing mouse hover.
GUI applications automatically have EventLoops created for them via
GUI::Application; however before this commit it was not possible to make
the application event loop inspectable. This exposes a third optional
argument to GUI::Application which allows one to make the application
event loop inspectable.
Instead of crashing when we can't make the back buffer non-volatile,
we now transition the window into single-buffered mode instead
(assuming it was originally in double-buffered mode.)
This reduces GUI fidelity a bit (by potentially making windows flicker
during repaint) but since it's only triggered in low-memory conditions,
it seems like a reasonable thing to sacrifice in order for the system
to carry on.
This patch also stops us from allocating entirely new backing stores
after the old ones were purged. If they were purged but reallocated
just fine, there's no need to allocate new memory again. We already
have fresh zero-filled pages in the existing bitmap at this point.
Making a bitmap non-volatile after being volatile may fail to allocate
physical pages after the kernel stole the old pages in a purge.
This is different from the pages being purged, but reallocated. In that
case, they are simply replaced with zero-fill-on-demand pages as if
they were freshly allocated.
This prevents GUI::TextBox and the `Paste & Go` action in the browser
from trying to paste a bitmap. Also, the paste action is enabled
and disabled on clipboard change to reflect if the clipboard data
can be pasted.
Applications previously had to create a GUI::Menubar object, add menus
to it, and then call GUI::Window::set_menubar().
This patch introduces GUI::Window::add_menu() which creates the menubar
automatically and adds items to it. Application code becomes slightly
simpler as a result. :^)
Previously, pressing 'x' for deletion on an otherwise empty line
insinuated a crash in TextEditor because a nonexistent code point was
accessed -- likewise for visual mode.
While under insert mode, Ctrl-U deletes all characters between the first
non-blank character of the line and the cursor.
Implement delete_from_line_start_to_cursor() in TextEditor. Then, call
the method in VimEditingEngine via its pointer to an instance of
TextEditor.
In Vim, Ctrl-H is effectively equivalent to backspace in insert mode, as
it deletes the previous character.
This commit implements method delete_previous_char() to TextEditor.
delete_char() already exists in EditingEngine, but it deletes the
next character rather than the previous. delete_previous_char() is then
called from within VimEditingEngine.
In Vim's insert mode, Ctrl-W deletes the word before the cursor, like
Ctrl-Backspace. Unlike Ctrl-Backspace, if only whitespace exists between
the end of the word and the cursor, the word will be deleted with the
whitespace.
To do so, this commit introduces two methods: delete_previous_word() for
TextEditor and first_word_before() for TextDocument, where the former
depends on the latter. delete_previous_word() is then called in
VimEditingEngine.
Adds a new on_rename_error handler and renames the old on_error handler
to on_directory_change_error in FileSystemModel. The on_rename_error
handler creates a MessageDialog with the error message.
AK's version should see better inlining behaviors, than the LibM one.
We avoid mixed usage for now though.
Also clean up some stale math includes and improper floatingpoint usage.
As threads come and go, we can't simply account for how many time
slices the threads at any given point may have been using. We need to
also account for threads that have since disappeared. This means we
also need to track how many time slices we have expired globally.
However, because this doesn't account for context switches outside of
the system timer tick values may still be under-reported. To solve this
we will need to track more accurate time information on each context
switch.
This also fixes top's cpu usage calculation which was still based on
the number of context switches.
Fixes#6473
Widget::is_visible_for_timer_purposes needs to also consult with the
base implementation, which ultimately checks the owning Window's
visibility and occlusion state. Widget::is_visible merely determins
whether a widget should be visible or not, regardless of the window's
state.
Fixes#8825
This saves a few lstat lookups since otherwise '/' is indexed before
set_path() is called. It also cleans up warnings if '/' is not
unveiled when opening FilePicker, like in WidgetGallery.
In the normal editing engine keys like Home, End etc work on visual
lines, but vim operates on real ones. Eg if you have a really long line
and wrapping is on then in normal editing Home would take you to the
beginning of the wrapped line, but 'I' would put you insert mode at the
beginning of the real line in vim.
This fixes a rather frustrating issue during saving a file,
when clicking on a folder (to change the path of saved file)
caused the filename to disappear from the text box.
Before this change, parent widgets such as Buttons or Labels
were stealing drop events their parents.
I noticed it during drag-n-dropping files into visualization widgets
in Sound Player (which takes practically the entire application size
and gave impression that drop events weren't supported in the app
at all).
