diff options
author | Max Horn | 2007-06-30 12:01:27 +0000 |
---|---|---|
committer | Max Horn | 2007-06-30 12:01:27 +0000 |
commit | fc6da0b436ed1ebf4585e3519f2d79b2379d342c (patch) | |
tree | 613b18b45873c58598afea6ee05f0739a0b2ef44 /engines | |
parent | 252d18329168bfdfd141a0dbaa07175f2cb2b1c8 (diff) | |
download | scummvm-rg350-fc6da0b436ed1ebf4585e3519f2d79b2379d342c.tar.gz scummvm-rg350-fc6da0b436ed1ebf4585e3519f2d79b2379d342c.tar.bz2 scummvm-rg350-fc6da0b436ed1ebf4585e3519f2d79b2379d342c.zip |
Removed/clarified various FIXME comments
svn-id: r27784
Diffstat (limited to 'engines')
-rw-r--r-- | engines/scumm/input.cpp | 35 |
1 files changed, 12 insertions, 23 deletions
diff --git a/engines/scumm/input.cpp b/engines/scumm/input.cpp index 05a9a8115a..6e75aba812 100644 --- a/engines/scumm/input.cpp +++ b/engines/scumm/input.cpp @@ -114,19 +114,14 @@ void ScummEngine::parseEvents() { VAR(VAR_KEY_STATE) = keyState; } - // FIXME: There is a discrepancy between EVENT_KEYDOWN and EVENT_KEYUP here: - // For EVENT_KEYDOWN, we use _keyPressed.keycode, which has potentially been - // modified, while for EVENT_KEYUP we use the unfiltered event.kbd.keycode. - // This could lead problems (like a key becoming 'stuck'). - - // FIXME #2: We are mixing ascii and keycode values here. We probably should - // be using keycodes, but at least INSANE checks for "Shift-V" by looking for - // the 'V' key being pressed. It would be easy to solve that by also storing the - // the modifier flags. However, since getKeyState() is also called by scripts, - // we have to be very careful with semantic changes. - // Nevertheless, it's bad to rely on "ascii" holdoing keycode values for special - // keys (like the function keys), so this should be fixed. - + // FIXME: We are using ASCII values to index the _keyDownMap here, + // yet later one code which checks _keyDownMap will use KEYCODEs + // to do so. That is, we are mixing ascii and keycode values here, + // which is bad. We probably should be only using keycodes, but at + // least INSANE checks for "Shift-V" by looking for the 'V' key + // being pressed. It would be easy to solve that by also storing + // the modifier flags. However, since getKeyState() is also called + // by scripts, we have to be careful with semantic changes. if (_keyPressed.ascii >= 512) debugC(DEBUG_GENERAL, "_keyPressed > 512 (%d)", _keyPressed.ascii); else @@ -134,11 +129,9 @@ void ScummEngine::parseEvents() { break; case Common::EVENT_KEYUP: - // FIXME: for some reason Common::KBD_ALT is set sometimes - // possible to a bug in sdl-common.cpp - if (event.kbd.ascii >= 512) + if (event.kbd.ascii >= 512) { debugC(DEBUG_GENERAL, "keyPressed > 512 (%d)", event.kbd.ascii); - else { + } else { _keyDownMap[event.kbd.ascii] = false; // Due to some weird bug with capslock key pressed @@ -148,6 +141,8 @@ void ScummEngine::parseEvents() { // both upper and lower letters are unpressed on keyup event // // Fixes bug #1709430: "FT: CAPSLOCK + V enables cheating for all fights" + // + // Fingolfin remarks: This wouldn't be a problem if we used keycodes. _keyDownMap[toupper(event.kbd.ascii)] = false; } break; @@ -473,12 +468,6 @@ void ScummEngine::processKeyboard(Common::KeyState lastKeyHit) { if (_game.id == GID_CMI) mainmenuKeyEnabled = true; -/* - FIXME: We also used to force-enable F5 in Sam&Max and HE >= 72 games -- why? - if ((_game.version <= 3) || (_game.id == GID_SAMNMAX) || (_game.id == GID_CMI) || (_game.heversion >= 72)) - mainmenuKeyEnabled = true; -*/ - if (mainmenuKeyEnabled && (lastKeyHit.keycode == Common::KEYCODE_F5 && lastKeyHit.flags == 0)) { if (VAR_SAVELOAD_SCRIPT != 0xFF && _currentRoom != 0) runScript(VAR(VAR_SAVELOAD_SCRIPT), 0, 0, 0); |