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. :^)
The crash happens very rarely and is hard to reproduce so it is
hard to know for certain, but I am confident this fixes it.
I previously delayed the start of the game-over animation by one
frame, but neglected to check m_start_game_over_animation_next_frame
wasn't set. This means multiple calls to start_game_over_animation()
on the same frame (or rather, before the first timer_event) would
each call Object::start_timer(). Now that we do check the flag,
that should no longer be possible.
Fixes#8122.
This fixes#7792.
The visual glitches with card corners, and the screen-clear when
opening a menu, were both caused by the widgets having
`fill_with_background_color` set. We need that set most of the time,
so we just disable it for the duration of the game-over animation.
Also resove a FIXME that no longer applies. :^)
Previously, the timer started if you clicked within the game area,
whether that was on a card or not. Now, we only start when you click
on a card or otherwise attempt a move.
As a bonus, we now immediately update the status bar time indicator
on game start, instead of having to wait until 1 second has elapsed.
Previously, if a tab-move moved all the remaining cards to the
foundations, the game would not notice until you moved a card
away and then back again.
I had to add a 1-frame delay to starting the animation, so that it
would have time to re-paint the foundation in that case.
Keeping this as a separate commit as I'm not certain whether this
is a good change or not. The repeated if-else for each Foundation
stack bothered me a bit, though more so before I reduced the code
in the {}. But maybe the ifs are clearer than the loop?
Doing that also meant I could inline the move_card() code instead
of needing to make it a lambda. Again, maybe it would be better as
a lambda? I'm still figuring out the style Serenity uses, and I
know Andreas is big on expressiveness, and move_card() is more
expressive than just having the code in the loop.
The two paths did the same thing, in two different ways. Now they
are the same. :^)
Can't quite put all of the logic into
attempt_to_move_card_to_foundations() because the double-click has
to check that you clicked on the top card.
Now that the cards have rounded corners, draw the stack box behind the
cards with rounded corners as well. This way, the corner of the stack
box doesn't peek out from behind the cards.
The caveat here is that the "play" card stack now needs to hold a
reference to the "waste" stack beneath it so it knows when not to draw
its background on top of the waste stack. To faciliate that, the array
of card stacks is now a NonnullRefPtrVector so the play stack can store
a smart pointer to the waste stack (instead of a raw pointer or
reference).
Also shifts logic of starting game length timer into function
`start_timer_if_necessary`, so it can be called from original
mouse event handler and new `auto_move_eligible_cards_to_stacks`
Currently, the player loses 100 points each time the waste stack is
recycled. In three-card draw mode, it's standard to only lose 20 points
after the third recycle event.
This invocation will exit immediately. There's also no reason to invoke
stop_game_over_animation here because that's the first thing that will
happen in the call to setup.
Currently, it is possible for the player to drag an entire stack
of cards to the foundation stack, provided the top card of the stack
(i.e the "root" card) can be dropped onto the foundation stack.
This causes an invalid state where, e.g, red cards end up in a
black foundation stack, or vice versa.
