This required the addition of a few exception handlers so to avoid #ifdef overuse I also added some #defines for non-Windows systems that allow using __try and __except directly in the code without #ifdef'ing them out.
UpdateSounds will not be called during screen wipes and the entire setup of this function suggests that this is not advisable at all.
The OpenAL stream updates were done deep inside this function implicitly.
This caused music to stop while a wipe was in progress. So in order to allow uninterrupted music playback during screen wipes the music updates need to be handled separately from sound updates and be called both in the main loop and the wipe loop.
I think that the OpenAL music updating should be offloaded to a separate thread but at least it's working now without causing interruptions during wipes.
- Viewheight change was delayed: Viewheight must be copied to player
structure as well as the PlayerPawn. Not sure if should actually use
deltaviewheight to spread the change out over a few tics instead of
being instant.
- Viewheight not preserved when travelling: player->viewheight must be
restored from pawn->ViewHeight, because the temporary player set it back
to the default viewheight.
- If palette index 255 happens to be white (e.g. as in Hexen), trying to
use white with DTA_FillColor would treat it as if you had never passed
it to DrawTexture().
- S_PrecacheLevel() must also mark currently playing sounds as
used. If we don't, the sound could be unloaded and the underlying
channel stopped without triggering a channel callback. That would leave
the code in s_sound.cpp thinking the sound is still playing even though
it isn't.
- Added an invalid channel check to FMODSoundRenderer::StopChannel() so
that orphan channels passed to it will be returned at least when
S_StopAllChannels() is called.
- OpenAL never actualy worked properly and was removed in later FMODs.
- Sound Manager was deprecated by Apple long ago and is not supported for
64-bit applications. It was also removed in later FMODs.
Tags are now handled by a tag manager class which stores sector/tag pairs. This way multiple entries can be added per sector.
Since UDMF does not have any arrays the additional tags are stored as a space separated string as 'MoreIDs'.
- When the OPL3 cores were added, DiskWriterIO was never updated to take
into account things like more than two OPL2 chips can be configured.
- DiskWriterIO no longer does any file writing directly. That function has
been split off into an OPLDump class, which has two specializations: one
for RDOS Play, and the other for DOSBox. If one chip is configured, it
dumps for a single OPL2, otherwise it dumps for an OPL3 (effectively
dual OPL2).
- TODO: Figure out why playback of raw OPL files doesn't sound nearly as
good as playing MIDI with the OPL emulation. It's probably something
simple I overlooked.
- Fixed: If you enlarged the game window (in windowed mode) so that the
window is bigger than the selected resolution, the menu would still take
its inputs from the portion in the upper left that matched the
resolution.
* make setting the line ID with P_TranslateLineDef explicit because there's one FraggleScript function that needs to work around the changes caused by this. There's also some functions setting only a temporary linedef. These would inevitably cause problems if the underlying data gets changed.
* remove FS function 'ChangeTag'. Fortunately this was just some long forgotten test stuff that can be removed without affecting any maps, but the feature would cause some serious problems in a more complex system.
With these changes it is guaranteed that after map setup the tag/ids won't change anymore.
This is done to encapsulate the gory details of tag search in one place so that the implementation of multiple tags per sector remains contained to a few isolated spots in the code.
This also moves the special 'tag == 0 -> activate backsector' handling into the iterator class.
- Old mess:
* Execute autoexec files right away.
* Execute -exec files right away.
* Execute command line commands right away.
- If, during any of the above, an unknown command or a set of an
unknown variable is encountered, store it for later.
- Pullin commands are directly executed and add to the list of files
to load.
* Do a little setup, including parsing CVARINFOs.
* Retry saved commands in case CVARINFO added a cvar they refer to.
- New, less messy, mess:
* Parse autoexec files into an array.
* Parse -exec files.
* Parse command line commands.
- During all of the above, exec commands are also parsed into the
array immediately rather than being saved for execution later.
- Pullin commands are parsed into a different array. The pullin
command doesn't actually do anything directly anymore.
* Add all the pullin files to the list of files to load.
* Do a little setup, including parsing CVARINFOs.
* Execute every command that was parsed in the preceding steps.
- Fixed: P_MoveThing had source and destination fog spawning backwards.
- Fixed a case where the NOTELEPORT flag would be ignored on A_Teleport.
- Added pointer selection to A_Teleport. Defaults to AAPTR_DEFAULT (calling actor). State jumps will only be done by the calling actor.
Both 'Adventures of Square' IWADs were missing an entry for base MAPINFO and as a result did not define the common editor numbers.
To prevent this, a new mindefaults MAPINFO was added to zdoom.pk3 which now gets loaded if IWADINFO does not specify a game-specific file.
This minimum setting sets all gamedefaults to a reasonable base value and defines all other things that are required to be defined.
This has an important implication:
Previously the config was loaded before IWADINFO so in order to allow the config to access the data this had to be switched around.
This means that zdoom.pk3 will not be looked for in the global IWAD search paths anymore, but since it shouldn't be there to begin with it should be an acceptable compromise.
As a result the old 'Group' property could be removed and all other means to get a section name were disabled.
As an example, if the code gets 'doom.doom2.commercial' it will use the following sections in this order:
global.autoload
doom.autoload
doom.doom2.autoload
doom.doom2.commercial.autoload.
- For when IWADs are too specific, filter by the base gametype too.
- Minor small edits to the sndinfo.txt files so that zipdir will notice
the changes, since it doesn't check path names when checking for
file differences.
- Previously, calling ReallocBuffer with a smaller buffer size than the
current one could overwrite unallocated memory. This required that the
string it was called on had more than one reference and therefore
required creating a new copy. The entire original string would be
copied, whether it fit in the new buffer or not.
- Multi-directory archives (e.g. zips) now support filtering lumps
depending on the loaded IWAD. The search rules are the same as for the
Autoload entries in the user's ini. For instance, if you are playing
Doom 2, the following filters will be applied:
* "filter/doom2/*"
* "filter/doom/*"
They will be renamed to strip out the "filter/doom2/" and "filter/doom/"
parts and will be ordered so they take precedence over any files not
inside a filter/ directory. Any files inside another filter/ directory
(e.g. "filter/hexen/*") will be ignored.
This helps the crash catcher when there are numerous wad files, for which it's possible that either the wad list, the map or the position is truncated. A more reliable alternative to this should be allocating the char buffer, but I never heard about a way to do this reliably during signal handling.
- changed monster unblocking logic to include players as well (i.e. a player being stuck inside another actor is allowed to move away from that other actor.)
* Changed the behavior of SetActorTeleFog.
- Don't force "null" to resolve to no actor since "none" is already defined as NULL (via FindClass). (This change also applies to the decorate properties.)
- Passing an empty actor name will keep the existing fog since there's otherwise no way set only one fog. Since "none" works to remove the fog, I see no reason not to have this option.
- Don't force "null" to resolve to no actor since "none" is already defined as NULL (via FindClass). (This change also applies to the decorate properties.)
- Passing an empty actor name will keep the existing fog since there's otherwise no way set only one fog. Since "none" works to remove the fog, I see no reason not to have this option.
* decided that the pain threshold should always be checked against the actual damage, even if it's down to 0, for consistency. This also restores the original behavior of using actual damage for checking the pain threshold which was altered by the introduction of the ALLOWPAIN and CAUSEPAIN flags.
* removed all newly added exceptions that excluded the player from checks for completely cancelled out damage.
* if anything during damage modification causes negative damage, no pain handling whatsoever will be initiated.
* made sure that TELEFRAG_DAMAGE will not be subjected to damage amount modification by protection items and any other kind of damage modification.
- Move MUSINFO change request out of FLevelLocals and into player_t. This
allows the MusicChanger actors to change music for each player
independantly. This is similar to PrBoom+, which switches depending on
the displayplayer. The difference being, we don't actually track the
music other players are listening to. (Which might not be a bad idea to
implement at some point.)
- Moved a few fields in player_t for better packing.
