This is an incredibly costly way to do a debug check as it infests the entire VM design from top to bottom. These tags are basically useless for anything else but validating object pointers being passed to native functions (i.e. mismatches between definition and declaration) and that simply does not justify a feature that costs execution time in non-debug builds and added memory overhead everywhere.
Note that this commit does not remove the tags, it only discontinues their use.
Both files can now be included independently without causing problems.
This also required moving some inline functions into separate files and splitting off the GC definitions from dobject.h to ensure that r_defs does not need to pull in any part of the object hierarchy.
The entire setup was quite broken with each item using its own activation result and the ones of the subsequent items in the list as the return value.
This rendered the STANDSTILL check in the main function totally unpredictable because the value it depended on could come from any item in the list.
Changed it so that the main dispatcher function is part of sector_t and does the stepping through the list iteratively instead of letting each item recursively call its successor and let this function decide for each item alone whether it should be removed.
The broken setup also had the effect that any MusicChanger would trigger all following SecActEnter specials right on msp start.
There are a few which require explicit native construction or destruction that need to be exported to the VM, e.g. FCheckPosition.
The VM cannot handle this directly, it needs two special functions to be attached to handle such elements.
- disabled the Build map loader after finding out that it has been completely broken and nonfunctional for a long time. Since this has no real value it will probably removed entirely in an upcoming commit.
- made ModifyDamage calls iterative instead of recursive. With going through the VM they'd be too costly otherwise.
- small optimization: Detect empty VM functions right when entering the VM and shortcut them. This is to reduce the overhead of virtual placeholders, which in a few cases (e.g. CanCollideWith and ModifyDamage) can be called quite frequently.
If the calling code wants to recycle this it will have to pass a container variable to AActor::UnlinkFromWorld and AActor::LinkToWorld.
This was changed because keeping such data in a global variable is dangerous for a set of functions that can be called from a script.
Note that the scripted versions do not yet support saving of the touching_sectorlist.
For most attack functions this is wrong, it's only the Hexen fighter attack needing this particular value, so it has been split up into two return values now.
This will get called for both actors taking part in a collision, if one of the two calls returns false it will immediately abort PIT_CheckThing with no collision taking place at all.
- cleaned up the virtual function interface of APlayerPawn which still had many virtual declarations from old times when class properties were handled through virtual overrides. None of this makes sense these days anymore.
Interesting tidbit: The damage calculation in P_MinotaurSlam had been incorrect for the Heretic version since the friendly Hexen Dark Servant was added, but nobody ever noticed in 14 years...
Needless to say, this is simply too volatile and would require constant active maintenance, not to mention a huge amount of work up front to get going.
It also hid a nasty problem with the Destroy method. Due to the way the garbage collector works, Destroy cannot be exposed to scripts as-is. It may be called from scripts but it may not be overridden from scripts because the garbage collector can call this function after all data needed for calling a scripted override has already been destroyed because if that data is also being collected there is no guarantee that proper order of destruction is observed. So for now Destroy is just a normal native method to scripted classes
- implemented method calls from struct instances.
- optimized disassembly of VM call instructions to print the function's name at the end where it is more visible and does not need to be truncated. Also use the printable name for script functions here.
- added support for global variables to the code generator - not the compiler, though. For the handful of entries this is needed for it may just as well be done manually. So far FLevelLocals level is the only one being exported.
- fixed: The VM disassembler truncated 64 bit pointers to 15 digits because the output buffer was too small.
- resolve entire FxSequences instead of aborting on the first failed entry. This allows to output all errors at once.
- swapped parameters of two-parameter VelToAngle method, so that internal and script version are in line.
- fixed parameter asserts to handle NULL pointers properly.
- fixed: Script functions did not receive the function name when being created.
- relaxed the asserts for PARAM_STATE, because the VM knows nothing about ATAG_STATE. Any state variable's content (e.g. Actor.SeeState) will receive ATAG_GENERIC, rather than ATAG_STATE.
- added a 'NeedResult' flag so that certain operations can create shorter code if the result of the expression is not needed. So far only used for postdecrement/increment statements on local variables (which is the most frequent case where this matters.)
- fixed postincrement and decrement for local variables. Due to the result preservation semantics it created faulty code.
- removed 'self' as a dedicated token. Internally this gets handled as a normal but implicitly named variable so the token just gets in the way of proper processing.
- removed P_ prefix from SpawnMissile export.
- fixed a crash with misnamed function exports.
The way this was done was a major headache inducer, requiring reconstruction of the function each time the value was changed and in general made actor damage a major hassle.
There was a DECORATE wrapper to mimic the original behavior but this looked quite broken because it completely ignored the different semantics of both damage calculation types.
It also made it impossible to determine if damage was a function or a value.
This accessor has been reverted to what it should be, only returning the constant, which now is -1 for a damage function. I am sorry if this may break the odd mod out but a quick look over some DECORATE-heavy stuff showed that this was never combined in any of them so that accessing 'damage' in DECORATE code depended on an actual damage function.
To get proper damage, a future commit will add a DECORATE function which calls AActor::GetMissileDamage.
This is necessary to prevent moving sectors from altering the bridge's z-position. The bridge should remain at its current z, even if the sector change would cause floorz or ceilingz to be changed in a way that would make P_ZMovement adjust the bridge.
- don't let P_PushUp move any actors with the same z as the pushing actor. This should solve most monster pile-ups, but the entire logic here needs some serious rethinking. The only reason this doesn't cause more problems is some fudging in several other places.
- ALLOWTHRUFLAGS must be used on the puffs, added for the sake of compatibility with older mods. This applies to the following:
-- Bullets: THRUACTORS, THRUSPECIES
-- Rails: Same as bullets, but includes THRUGHOST.
- restored the original 3D floor code to retrieve the current floor in P_CheckPosition. The portal aware version was a bit too strict and could place the actor on the wrong side when moving at high speeds.
This got accidentally committed. Even if this gets extended to reach through portals it needs to be done differently. FMultiBlockLinesIterator can't guarantee to get every sector that's being touched.
This is for rendering the sprite properly in all areas the actor touches. The original thinglist is not sufficient for this and Boom's touching thinglist has other purposes and collects too much data.
This new list will only get filled in when the actor is actually crossing a portal plane, for the normal sector thinglist will still be used.
This piggybacks on the msecnode_t code which has been extended to be able to handle more than one list by passing the sector's membert pointers as parameters.
This required some changes to the Trace function because it turned out that the original was incapable of collecting the required information:
* actors are now also linked into blockmap blocks on both sides if they occupy the boundary of a sector portal.
* Trace will no longer set up parallel traces in all parts connected with sector portal, but only use one trace and relocate that on the actual boundary.
The only reason this even existed was that ZDoom's original VC projects used __fastcall. The CMake generated project do not, they stick to __cdecl.
Since no performance gain can be seen by using __fastcall the best course of action is to just remove all traces of it from the source and forget that it ever existed.
