Try to tell Macs with two graphics cards that it's ok to
use the low-performance one for the GC bitmap (when
GL is used for that, which is OS X 10.11 and later).
Provided by Eli Barzilay.
This approach is better than a separate process, because creating
too many processes can overwhelm the OS. Also, MCI supports more
sound formats.
Add more agressive re-enabling of screen updates and explicit `update`
calls to avoid partially refreshed frames and never-updated titlebars
on El Capitan.
Also, use `close` instead of `orderOut` to hide a frame. That fixes
problems with closing windows that are in fullscreen mode.
Closesracket/drracket#33
The GC blit implementation used on Mc OS X 10.11 assumed that
the no-GC bitmap is blank. Make it use the given no-GC bitmap.
Also, repair the left-to-right flipping(!) of the GC bitmap,
and repair a backing-scale mismatch that could leave a thin
border around a GC blit.
A relatively late correction to refresh handling for El Capitan
(commit 9bf18505d5) causes a canvas to be refreshed too often
in some cases. Delay the refresh to restore the old timing
without the old bug.
Opening a shared object in global mode risks conflicts with
other shared objects, but opening only one library that way
will hopefully not create conflicts.
The new system control font is (intentially) not associated with a
font family, so it's inaccessible via Pango. A patch to Pango lets
us synthesize and register new families, so the control font can
be made accessible.
The `fullscreen` method implied `show` on Mac OS X and Windows, but
not on Gtk, and a frae shown that was was not properly registered
as shown (e.g., the application could exit). For consistency, adjust
`fullscreen` to not imply `show`. Also, shift to a remaining frame
on Mac OS X when a fullscreen frame is closed.
Previous versions of OS X seem to have automatically disabled
update for a window when flushing is disabled, but now we have to
do it ourselves, otherwise text editors can flicker a lot. (The
system change is probably related to the removal of the
`isFlushWindowDisabled` method of `NSWindow` (which was
deprecated in 10.10).
Closes PR 15121
A better solution would be to implement the dialog, but given
how long it's been broken, maybe no one cares enough for that
to be worthwhile.
The old strategy of switching a transparent window to solid and back
doesn't work on 10.11; it appears that queued messages must be handled
for the window to become visible, but that's not allowed during a GC.
The strategy for 10.11 and up create an OpenGL canvas, which acts as a
direct-to-screen drawing area that a GC callback can affect without
Racket-level allocation.
It seems that the right answer was produced in GTK+ 2 due to a
client-location callback, but change the code to use the intended
approach of tracking the menu height.
To enable mouse events to a panel, go back to using a
GtkEventBox for a panel, but make it non-visible so that
it doesn't interfere with the parent's background color.
The extra layers added in the previous commit are still
important to avoid turning that into a native window
(which can completely breaks background drawing). Finally,
add an extra layer just around tab panels so that events
get delivered to the tabs. (It's possible that the extra
layer will make the background wrong just around the tabs,
if the tabs don't fill the area above the panel; too bad.)
A GC-icon display is implemented with a child window on OS X. In some
OS X version prior to 10.10 (not sure which ones), screen
reconfigurations did not move child windows properly relative to
parent windows, so all child windows were reset after a screen change.
In 10.10, meanwhile, Mission Control can cause screen-change
notifications, and adding a child window to a Mission
Control-minimized fullscreened window will goes terribly wrong.
Fortunately, 10.10 seems to update child-window locations correctly on
screen reconfigurations, in which case the old workaround that hits
the new bug can be skipped.
