| Age | Commit message (Collapse) | Author |
|---|
|
- objectMap responds to mouse move (but respond script not run well ?)
loadStrings add some special count check
- so all other LUT based resource should implement this technique
svn-id: r16594
|
|
svn-id: r16593
|
|
- remove ActionMap.h & ActionMap.cpp
- ObjectMap names move to Scene::_sceneStrings as in original engine
- fix wrong StringsTable::stringsCount calculation
svn-id: r16592
|
|
svn-id: r16591
|
|
Things which are missing:
(a) mouse support due to incomplete interface implementation
(b) arrows do not pop up by same reason mentioned above
(c) scrolling does not work
(d) kReplyOnce flag is missing due to wrong threads memory implementation
svn-id: r16589
|
|
'scene_change 33' in debugger.
svn-id: r16588
|
|
difference
o Fix regression in Mac titles where scene LUT wasn't conversed
o Fix sound types in many games. Old Win demo is still wrong.
svn-id: r16586
|
|
svn-id: r16585
|
|
already, let it continue playing rather than restarting it. (I've been
meaning to make this change for some time, but now it appeared on the TODO
list as well. :-)
svn-id: r16583
|
|
the reason the ITE intro crashed for me.) I still get read warnings in
Sprite::decodeRLEBuffer() though.
Also added a couple of hacks to keep the IHNM intro alive. Once we support
IHNM actors, we'll be able to remove them.
svn-id: r16579
|
|
-pathfind speedup
svn-id: r16576
|
|
svn-id: r16575
|
|
test: in ITE type "scene_change 130" at console
svn-id: r16574
|
|
The first IHNM script does not use random branching after all.)
svn-id: r16573
|
|
implementation is only right for ITE, disable it for IHNM for now.
svn-id: r16572
|
|
svn-id: r16571
|
|
supported)
svn-id: r16568
|
|
- many structers and fields renamed to proper names
- added missing functions
svn-id: r16562
|
|
interpreter. Supposedly these colours will stay fairly constant throughout
the game, in which case we no longer need to repaint it with the "best
white" colour every time the palette changes.
svn-id: r16561
|
|
NULL it'll use the English string instead. It seems unnecessary to store
the same string twice, even if it's possible that the compiler is smart
enough to eliminate the duplicates.
Either way it doesn't make much difference now, but we may want to add the
intro strings to the getTextString() mechanism eventually. In that case
most of the credits would be duplicates, for instance.
svn-id: r16560
|
|
svn-id: r16559
|
|
IHNM script seems to use it, but doesn't get that far because it terminates
on an unknown script function.
svn-id: r16558
|
|
svn-id: r16557
|
|
hardcoded, but I'll save that for later :)
o Eliminated all leftovers of previous language loader. If you'll spot another
one, just remove it silently.
svn-id: r16556
|
|
corresponding string array)
- first step in verb implementation: proper button draw, keyboard respond 'w','l' &etc
- added comments and some fields to GameDisplayInfo
svn-id: r16554
|
|
svn-id: r16553
|
|
* Added speechCoords to the SpeechData structure so that talking actors
don't drag their speech subtitles with them just because they're moving
and talking at the same time. (Could this also be useful for non-actor
speech?) If the actor has multiple strings, the coordinates are updated
for each new string.
* Made speechColor and outlineCoor arrays so that simultaneous speech (i.e.
where several actors are talking at the same time) multi-coloured. This
is completely untested, though.
* Used getBlack() to get the black colour for the text outline. The
original uses a constant, but we could always make getBlack() return
that constant, if we want to.
svn-id: r16550
|
|
svn-id: r16549
|
|
it changes to the link. I'm guessing that the last frame in the animation
is identical to the first frame in the link.
The original calls SetAlarm(&deltaAlarm[i], delta->timer) between each
frame, but amends itself by calling SetAlarm(&deltaAlarm[i], 0) when the
link happens. I'm trying to simulate that effect by using 0 as frame time
instead of anim->frame_time. I think that's the right thing to do.
svn-id: r16548
|
|
svn-id: r16546
|
|
- all display information of current game goes to SagaEngine
- remove GameFontIds
- hide SagaEngine _gameId,_gameType & etc with methods getGameId(), getGameType()
svn-id: r16545
|
|
used in the intro. That way, if the IHNM demo plays an intro we'll use the
first voice file both for the demo and the full game.
Eventually we'll need to implement switching between different voice files,
of course, but that can wait.
svn-id: r16542
|
|
MIN() and MAX(). I then removed util.h from a bunch of files which I don't
think need it any more. (Please let me know if I got too blood-thirsty!)
This reverts some of the changes I made this morning.
svn-id: r16541
|
|
svn-id: r16540
|
|
o Implemented and tested converse drawing. Still some features like
arrows and hardcoded values are present, and it is not used in scripts
svn-id: r16536
|
|
in ITE_interface structure.
svn-id: r16533
|
|
svn-id: r16530
|
|
reduced this (total dependencies on system.h went down from 193 to 85 files)
svn-id: r16527
|
|
svn-id: r16519
|
|
svn-id: r16516
|
|
svn-id: r16515
|
|
- structures renamed
- SagaEngine class gives all current game descriptions
regression : "verb" is broken cause work in progress
svn-id: r16511
|
|
- actor barrier rects
- unreachable code
svn-id: r16510
|
|
svn-id: r16509
|
|
o Now we correctly show Wyrmkeep logo in Mac demos
o Applied patch #1098971 ITE credits
svn-id: r16507
|
|
of the engine maintainers can look into using it
svn-id: r16503
|
|
svn-id: r16498
|
|
- sprites decoded on first load
- some speed optimization
- actors uses kExtended flag to append spriteList
svn-id: r16497
|
|
svn-id: r16492
|
|
introduced getObjectName
svn-id: r16491
|
