Interpreting Keyboard Data
This topic covers the identification of keys for which data is reported by the IDirectInputDevice8::GetDeviceState and IDirectInputDevice8::GetDeviceData methods. For more information about interpreting the data from IDirectInputDevice8::GetDeviceData, see Time Stamps and Sequence Numbers.
In one important respect, DirectInput applications read the keyboard differently from the way Windows does. For DirectInput applications, keyboard data refers not to virtual keys but to the actual physical keys - that is, the scan codes. DIK_ENTER, for example, refers only to the ENTER key on the main keyboard, not to the ENTER key on the numerical keypad.
Additionally, differences among keyboards can affect the way DirectInput applications interpret keyboard data. For example, DirectInput defines a constant for each key on the enhanced keyboard, as well as the additional keys on international keyboards. Because NEC keyboards support different scan codes than the PC-enhanced keyboards, DirectInput translates NEC key-scan codes into PC-enhanced scan codes where possible.
Other keyboard differences to consider are:
Not all PC-enhanced keyboards have the Windows logo keys (DIK_LWIN, DIK_RWIN, and DIK_APPS), there is no way to determine whether the keys are physically available.
Laptops and other small computers often do not implement a full set of keys. Instead, some keys (typically numeric keypad keys) are multiplexed with other keys, which are selected by an auxiliary mode key that does not generate a separate scan code.
If the keyboard subtype indicates a PC XT or PC AT keyboard, the following keys are not available: DIK_F11, DIK_F12, and all the extended keys (DIK_* values greater than 0x7F). Furthermore, the PC XT keyboard lacks DIK_SYSRQ.
Japanese keyboards, particularly the NEC PC-98 keyboards, contain a substantially different set of keys than U.S. keyboards. For more information, see DirectInput and Japanese Keyboards.