------------------------------------------------------------------------
r22568 | wjpalenstijn | 2006-05-21 18:18:23 -0300 (Sun, 21 May 2006) | 1 line

make printExpr, skipExpr accept both gob1 and gob2 expressions
------------------------------------------------------------------------
r22567 | eriktorbjorn | 2006-05-21 17:41:53 -0300 (Sun, 21 May 2006) | 2 lines

Palette handling simplification, suggested by LordHoto.

------------------------------------------------------------------------
r22566 | eriktorbjorn | 2006-05-21 17:25:22 -0300 (Sun, 21 May 2006) | 2 lines

If possible, sync the frame to getSoundElapsedTime().

------------------------------------------------------------------------
r22565 | eriktorbjorn | 2006-05-21 17:08:14 -0300 (Sun, 21 May 2006) | 5 lines

Grotesque hack to support the jung2.vqa movie. Either the VQA is less well
understood than I hoped, or the offset to the first frame of the movie is
completely out to lunch. Scan the file for the first VQFR chunk and use that
offset instead.

------------------------------------------------------------------------
r22564 | eriktorbjorn | 2006-05-21 16:08:10 -0300 (Sun, 21 May 2006) | 3 lines

The unsupported cutscene is called "jung", not "junk". (It just happens to look
like junk at the moment.)

------------------------------------------------------------------------
r22563 | eriktorbjorn | 2006-05-21 16:03:27 -0300 (Sun, 21 May 2006) | 7 lines

More VQA fixes:

* Use setScreenPalette() rather than calling the backend directly. (As an extra
  bonus, the VQA player now only needs to store 3 bytes per colour.)

* Hide the mouse cursor while the movie is playing.

------------------------------------------------------------------------
r22562 | eriktorbjorn | 2006-05-21 15:49:26 -0300 (Sun, 21 May 2006) | 2 lines

Don't crash if drawMouse() is called before _mouseSurface is created.

------------------------------------------------------------------------
r22561 | eriktorbjorn | 2006-05-21 15:16:34 -0300 (Sun, 21 May 2006) | 8 lines

At LordHoto's request...

* The VQA move player isn't as similar to the WSA movie player as we first
  envisioned, so the VQA player no longer inherits from Movie. It does retain
  a fairly similar calling interface, though.

* Use the Kyra engine's idea of screen dimensions, rather than the backend's.

------------------------------------------------------------------------
r22560 | eriktorbjorn | 2006-05-21 14:28:03 -0300 (Sun, 21 May 2006) | 11 lines

Some minor VQA player fixes:

* The playVQA() function now takes the name of the cutscene, rather than the
  filename. This is so that playVQA("FOO") will be able to play FOO0.VQA,
  FOO1.VQA or FOO2.VQA, depending on which movie size is selected.

* The VQA player centers movies by default, so don't set the position to 0,0.

* Ask the backend for screen dimensions (for centering the movie) rather than
  hard-coding them.

------------------------------------------------------------------------
r22559 | lordhoto | 2006-05-21 13:06:29 -0300 (Sun, 21 May 2006) | 2 lines

- allows kyra3 to exit again
- implements language selection for kyra3
------------------------------------------------------------------------
r22558 | lordhoto | 2006-05-21 12:41:32 -0300 (Sun, 21 May 2006) | 2 lines

- reworks quit handling (no _system->quit() anymore)
- fixes some deinitalization problems
------------------------------------------------------------------------
r22557 | lordhoto | 2006-05-21 12:20:12 -0300 (Sun, 21 May 2006) | 2 lines

- adds debugC calls to KyraEngine_v3 functions
- finishes KyraEngine_v3::gui_printString
------------------------------------------------------------------------
r22556 | lordhoto | 2006-05-21 12:02:12 -0300 (Sun, 21 May 2006) | 1 line

Implements kyra3 main menu drawing.
------------------------------------------------------------------------
r22554 | eriktorbjorn | 2006-05-21 10:27:18 -0300 (Sun, 21 May 2006) | 7 lines

Make it so that if the new cursor palette to be set has zero colours, the
cursor palette is disabled.

Also, when replacing a cursor palette, try to re-use the old palette buffer, as
a minor optimization. (Not that these functions should need any optimization,
but it's simple and shouldn't hurt.)

------------------------------------------------------------------------
r22553 | dreammaster | 2006-05-21 10:04:49 -0300 (Sun, 21 May 2006) | 1 line

Added two lines for the Lure engine progress
------------------------------------------------------------------------
r22552 | eriktorbjorn | 2006-05-21 09:32:39 -0300 (Sun, 21 May 2006) | 2 lines

Oops, I guess I never got around to fully implement replaceCursorPalette().

------------------------------------------------------------------------
r22551 | aquadran | 2006-05-20 12:43:27 -0300 (Sat, 20 May 2006) | 1 line

fixed compilation when DISABLE_SCALERS
------------------------------------------------------------------------
r22550 | aquadran | 2006-05-20 12:22:17 -0300 (Sat, 20 May 2006) | 1 line

fix compilation, that variable no longer exists
------------------------------------------------------------------------
r22549 | fingolfin | 2006-05-20 12:10:04 -0300 (Sat, 20 May 2006) | 1 line

Make the copy constructor / assignment operator of class File private to prevent accidental copying of File objects
------------------------------------------------------------------------
r22548 | fingolfin | 2006-05-20 12:09:15 -0300 (Sat, 20 May 2006) | 1 line

Pass values by-reference, using C++ references, and not by using pointers
------------------------------------------------------------------------
r22547 | lordhoto | 2006-05-20 12:02:39 -0300 (Sat, 20 May 2006) | 2 lines

makes destructor of KyraEngine virtual

------------------------------------------------------------------------
r22546 | eriktorbjorn | 2006-05-20 10:59:31 -0300 (Sat, 20 May 2006) | 8 lines

Apparently, passing a file handle directly to a function causes that file
handle (a copy of the original?) to be destroyed at the end of the function,
and that closes the internal file handle, even though we want it to stay open.

Who am I to argue with how C++ handles its parameters? I've simply changed the
functions to take pointers to file handles instead. Seems to fix crashes in the
Macintosh resource extractor for me.

------------------------------------------------------------------------
r22544 | eriktorbjorn | 2006-05-20 08:24:45 -0300 (Sat, 20 May 2006) | 2 lines

Forgot to set g_initialized to true.

------------------------------------------------------------------------
r22543 | eriktorbjorn | 2006-05-20 07:59:25 -0300 (Sat, 20 May 2006) | 4 lines

After the GUI has finished, restore the old cursor palette (if any). For this
to work, cursor palette now has to be set using the new "palette manager". See
graphics/paletteman.cpp

------------------------------------------------------------------------
r22542 | eriktorbjorn | 2006-05-20 07:57:15 -0300 (Sat, 20 May 2006) | 2 lines

Removed stray printf() from earlier commit.

------------------------------------------------------------------------
r22541 | eriktorbjorn | 2006-05-20 06:27:56 -0300 (Sat, 20 May 2006) | 3 lines

Re-render the cursor when enabling/disabling cursor palette. Also, I don't
think _cursorHasOwnPalette is needed any more, so I've removed it.

------------------------------------------------------------------------
r22540 | eriktorbjorn | 2006-05-20 05:29:50 -0300 (Sat, 20 May 2006) | 3 lines

Set a cursor palette for the classic theme, as well. Otherwise we'll get the
wrong colours in many games.

------------------------------------------------------------------------
r22539 | eriktorbjorn | 2006-05-20 05:06:21 -0300 (Sat, 20 May 2006) | 9 lines

Our GUI uses the "cursor palette" feature, but the SDL backend did not use the
cursor palette in overlay mode. Now it does, which fixes bug #1476666. The
reason it worked at all before was that blitCursor() was called while still not
in overlay mode.

This, of course, means that the GUI can corrupt the in-game cursor palette. I
don't have any game that uses this feature, so I don't know if it's properly
restored again afterwards. But if it isn't, that should be fairly easy to fix.

------------------------------------------------------------------------
r22538 | eriktorbjorn | 2006-05-20 04:31:03 -0300 (Sat, 20 May 2006) | 2 lines

Fixed cursor drawing in the classic theme.

------------------------------------------------------------------------
r22537 | eriktorbjorn | 2006-05-19 19:49:23 -0300 (Fri, 19 May 2006) | 3 lines

Call the inherited handleScreenChanged() method, so that the widget can resize
itself on scaler changes.

------------------------------------------------------------------------
r22536 | eriktorbjorn | 2006-05-19 19:41:51 -0300 (Fri, 19 May 2006) | 4 lines

Call handleScreenChanged() *before* calling setEditString(). Otherwise
_leftPadding may be wrong, or even undefined, when getEditRect() is called,
which may cause ScummVM to trigger an "invalid rect" assertion.

------------------------------------------------------------------------
r22535 | sev | 2006-05-19 19:26:07 -0300 (Fri, 19 May 2006) | 2 lines

Slighty modified patch #1491859: "Possible fix for recent cursor regressions"

------------------------------------------------------------------------
r22534 | sev | 2006-05-19 19:25:14 -0300 (Fri, 19 May 2006) | 2 lines

Minor GUI tweaks

------------------------------------------------------------------------
r22533 | eriktorbjorn | 2006-05-19 04:34:24 -0300 (Fri, 19 May 2006) | 2 lines

Using assert(_something->init()) is a bad idea. Assertions can be disabled.

------------------------------------------------------------------------
r22532 | eriktorbjorn | 2006-05-19 04:32:00 -0300 (Fri, 19 May 2006) | 2 lines

Added some BS2 and Simon notes.

------------------------------------------------------------------------
r22531 | fingolfin | 2006-05-19 04:20:27 -0300 (Fri, 19 May 2006) | 1 line

Added some FIXMEs to the kyra engine
------------------------------------------------------------------------
r22530 | fingolfin | 2006-05-19 04:12:38 -0300 (Fri, 19 May 2006) | 1 line

Fix warning
------------------------------------------------------------------------
r22529 | kirben | 2006-05-18 21:35:13 -0300 (Thu, 18 May 2006) | 1 line

Update HE version for another English version of socks
------------------------------------------------------------------------
r22528 | lordhoto | 2006-05-18 19:46:34 -0300 (Thu, 18 May 2006) | 5 lines

- load fonts for kyra3
- adds screen dim tables for kyra3
- adds palette / screen backup for the vqa player
- small changes in the vqa player to use the Kyra::Screen functions
- adds (partially implemented/unimplemented) main menu drawing/handling functions (nothing to see yet)
------------------------------------------------------------------------
r22527 | sev | 2006-05-18 18:48:47 -0300 (Thu, 18 May 2006) | 2 lines

Proper rendering of slider widget.

------------------------------------------------------------------------
r22526 | eriktorbjorn | 2006-05-18 18:46:07 -0300 (Thu, 18 May 2006) | 8 lines

Added player for the Kyra 3 VQA cutscenes, based on my earlier prototype. It
could use some cleanup, and there are a couple of TODOs sprinkled throughout
the code, but it seems to work reasonably well. Until the Kyra 3 main menu is
implemented, it won't actually be used though.

It uses the appendable audio stream class, which I have moved out of the SCUMM
engine.

------------------------------------------------------------------------
r22525 | sev | 2006-05-18 17:53:28 -0300 (Thu, 18 May 2006) | 8 lines

- Fix spelling in theme config
- Introduce embossing shadow
- Less round corners in popup widget
- Custom border in editable text widget
- Matched color in popup widget and editable text widget with mock-ups
- Less round corners in editable text widget
- Text padding in editable text widget

------------------------------------------------------------------------
r22524 | sev | 2006-05-18 16:38:54 -0300 (Thu, 18 May 2006) | 2 lines

Oops. Break indefinite loop.

------------------------------------------------------------------------
r22523 | sev | 2006-05-18 16:37:38 -0300 (Thu, 18 May 2006) | 2 lines

Plug memory leak

------------------------------------------------------------------------
r22522 | sev | 2006-05-18 13:44:44 -0300 (Thu, 18 May 2006) | 3 lines

Fix bug #1483272: "GUI: SCUMM pause dialog breaks upon scaler switch". Also
fix same problem in scumm help dialog.

------------------------------------------------------------------------
r22521 | sev | 2006-05-18 12:40:44 -0300 (Thu, 18 May 2006) | 2 lines

Properly set GFX mode selected in GUI. Fixes selection of 1x scaler.

------------------------------------------------------------------------
r22520 | eriktorbjorn | 2006-05-18 11:28:19 -0300 (Thu, 18 May 2006) | 3 lines

Initialise _resList to avoid crash (from freeing an invalid pointer) if the
data files are missing.

------------------------------------------------------------------------
r22519 | sev | 2006-05-18 10:59:20 -0300 (Thu, 18 May 2006) | 2 lines

--force-1x-overlay is no more. Remove leftovers.

------------------------------------------------------------------------
r22518 | eriktorbjorn | 2006-05-18 10:58:33 -0300 (Thu, 18 May 2006) | 2 lines

Renamed screenChanged() to handleScreenChanged().

------------------------------------------------------------------------
r22517 | sev | 2006-05-18 10:40:35 -0300 (Thu, 18 May 2006) | 4 lines

More GUI fixes:
- Center dialog title in scumm save/load dialog
- more space in add game dialog between dialog header and selected path

------------------------------------------------------------------------
