The Map loader may not access any global state at all - everything it can touch must be exchangable.
Furthermore, if we want to sandbox each level, there may be no direct access to any kind of global state whatsoever from the play code.
- moved parts of the render setup out of the separate render functions.
Things like particle and polyobject linking were duplicated several times for rendering different things in different renderers.
These things only need to be set up once before the renderer is started so it makes a lot more sense to consolidate them into one place outside the actual rendering code.
This had two different flags that were checked totally inconsistently, and one was not even saved.
Moved everything into a few subfunctions so that these checks do not have to be scattered all over the code.
There is no need to do this deep inside the renderer where it required code duplication and made it problematic to execute on multiple levels.
This is now being done before and after the top level call into the renderer in d_main.cpp.
This also serializes the interpolator itself to avoid problems with the Serialize functions adding the interpolations into the list which can only work with a single global instance.
Since currently there is only one level, this will obvciously only run once on that level for the time being.
This is mainly used for CCMDs and CVARs which either print some diagnostics or change some user-settable configuration.
This was done to ensure that this code only runs when the thinker itself is fully set up.
With a constructor there is no control about such things, if some common initialization needs to be done it has to be in the base constructor, but that makes the entire approach chosen here to ensure proper linking into the thinker chains impossible.
ZDoom originally did it that way, which resulted in a very inflexible system and required some awful hacks to let the serializer work with it - the corresponding bSerialOverride flag is now gone.
The only thinker class still having a constructor is DFraggleThinker, because it contains non-serializable data that needs to be initialized in a piece of code that always runs, regardless of whether the object is created explicitly or from a savegame.
This doesn't really write out any info for the pointer, if the level does not match it just errors out.
This is both for quick detection of badly used level data and for automatic restoring of the pointer from the serializer's working level.
This also removed the temporary workarounds in DAutomap and DLevelScript to restore these pointers when a savegame is loaded.
This was a relatively cheap change but removes a significant batch of references to the global variable, only making the entry points to the ACS interpreter relevant.
UI always runs on the primary level, so this does not need the ability to operate on multiple levels. Additionally, this can later be set to null when running play code so that scope violations result in an abort.
This also changes the action special interface to pass a Level parameter to the separate functions and makes a few other minor adjustments to the polyobject code.
This should later be done for everything else as well, but the map loader should really be free of global dependencies ASAP.
Also replace TThinkerIterator<AActor> with FThinkerIterator globally because this only adds pointless type checks - with all actor subclasses being scripted this class has become redundant.
This was yet another piece of code that lived or died with the assumption that there can only be one level, stored in global variables.
# Conflicts:
# src/p_saveg.cpp
There are several places where a temporary change of light mode is needed, all these made this change in the global level struct. Now the change is only local to the active draw info.
This involves passing the level explicitly to many functions. What was done here may seem a bit excessive but at least it covers everything.
Most importantly, the global ActiveThinker pointer has been moved into FLevelLocals and is now getting tracked properly by the level without using dangerous assumptions about how the game organizes its data.
This should be less of a drag on the playsim than having each light a separate actor. A quick check with ZDCMP2 showed that the light processing time was reduced to 1/3rd from 0.5 ms to 0.17 ms per tic.
It's also one native actor class less.
Since the SpawnedThings array is still available when polyobjects are spawned it makes no sense to create an expensive linked list in P_SpawnMapThing.
This can be done far better by scanning through the array again and collect all matching items in a second array.
This did no longer sort sprites in the same position reliably since the feature to render sprites which only partially are inside a sector was added.
With this, sprites in the same position are no longer guaranteed to be added to the render list in sequence.
Fixed by adding an 'order' field to AActor which gets incremented with each spawned actor and reset when a new level is started.
The software renderer will also need a variation of this fix but its data no longer has access to the defining actor when being sorted, so a bit more work is needed here.
* it was never saved in savegames, leaving the state of dead bodies undefined
* it shouldn't be subjected to pointer substitution because all it contains is old dead bodies, not live ones.
* only call hw_CheckViewArea if the result is not known yet.
* check the map up front if it even contains heightsecs. This allows to shortcut the above check entirely for maps without sector transfers and will allow further optimizations.
* the MAPINFO options now get handled in g_mapinfo.cpp and g_level.cpp, just like the rest of them as members of level_info_t and FLevelLocals.
* RecalcVertexHeights has been made a member of vertex_t and been moved to p_sectors.cpp.
* the dumpgeometry CCMD has been moved to p_setup.cpp
For some files that had the Doom Source license attached but saw heavy external contributions over the years I added a special note to license all original ZDoom code under BSD.
- replaced TStaticArray with regular TArrays.
They had incomplete implementations preventing proper cleanup of the level loading code. It makes more sense to add the missing methods to the regular TArray and use that.
This also makes some changes to how the game nodes are used to avoid creating a copy: If the head node's pointer is stored in a separate variable, no code needs to check which of the two arrays gets used.
