- because the main event loop is not called if initialization fails, it is presently impossible for MsgBoxError() to work if UI initialization fails; this basically means we cannot allow initializiation to fail on Mac OS X if we want to be able to report UI init failures to the user with one (which would be desirable, maybe (would violate Windows HIG?))
- NSComboBox scans the entered text to see if it matches one of the items and returns the index of that item if it does; find out how to suppress this so that it returns -1 unless the item was chosen from the list (like the other platforms)
- make sure scrollbars in Listbox work identically on all platforms (specifically the existence and autohiding of both horizontal and vertical scrollbars)
- listboxes spanning the vertical height of the window don't always align with the bottom border of the edit control attached to the bottom of the window...
- make sure mouse events don't trigger if the control size is larger than the Area size and the mouse event happens outside the Area range on all platforms
- formalize what happens if Modifiers by themselves are held
- OS X: find out if multiple DIFFERENT modifiers released at once produces multiple events
- in general, figure out what to do on multiple events, period
- OS X: handle Insert/Help key change in a sane and deterministic way
- will need old and new Mac keyboards...
- should pressing modifier+key in the keyboard test mark the key alone as pressed as well? I'm leaning toward no, in which case make sure this behavior exists on all platforms
- 64-bit crashes in malloc in wine with heap corruption warnings aplenty during DLL loading; in windows 7 it works fine
- 32-bit: it works, but if I save the class name converted to UTF-16 beforehand, wine indicates that the class name is replaced with the window title, so something there is wrong...
- https://developer.apple.com/library/mac/documentation/cocoa/Conceptual/NSScrollViewGuide/Articles/Creating.html#//apple_ref/doc/uid/TP40003226-SW4 the warning about pixel alignment may or may not be heeded, not sure
- frame sizes are a bit of a hack: the preferred size of a NSScrollView is the preferred size of its document view; the frameSize method described on the above link might be better but a real solution is optimal
- when resizing a GTK+ window smaller than a certain size, the controls inside will start clipping in bizarre ways (progress bars/entry lines will just cut off; editable comboboxes will stretch slightly longer than noneditable ones; the horizontal scrollbar in Area will disappear smoothly; etc.)
- for our two custom window classes, we should allocate extra space in the window class's info structure and then use SetWindowLongPtrW() during WM_CREATE to store the sysData and not have to make a new window class each time; this might also fix the s != nil && s.hwnd != 0 special cases in the Area WndProc if done right
- Area redraw on Windows is still a bit flaky, especially after changing the Area size to something larger than the window size and then resizing the window(???)
- despite us explicitly clearing the clip area on Windows, Area still doesn't seem to draw alpha bits correctly... it appears as if we are drawing over the existing image each time
- pressing global keycodes (including kwin's zoom in/out) when running the keyboard test in wine causes the Area to lose keyboard focus; this doesn't happen on the GTK+ version (fix the Windows version to behave like the GTK+ version)
- make passing of parameters and type conversions of parameters to uitask on Windows consistent: explicit _WPARAM(xxx)/_LPARAM(xxx)/uintptr(xxx), for example
- do this for type signatures in exported functions: (err error) or just error?
- on windows 7, progress bars seem to animate from 0 -> pos when you turn off marquee mode and set pos; see if that's documented or if I'm doing something wrong
- clean up windows struct field names (holdover from when the intent was to make a wrapper lib first and then use it rather than using the windows API directly)
