Original fix written by Lat`.

Write a new hack for getting sector->linecount from sector->lines in Lua, to put my mind at rest about it at last.

1) In sector_get, actually push the memory address of the lines array within sector_t, rather than push the value of "lines" itself (essentially, we we want a pointer to a double pointer, or rather a TRIPLE pointer haha)
2) In the sectorlines_* functions, use offsetof to shift the memory address so we can obtain the value of linecount within the sector_t struct, and dereference the result to obtain the value of linecount itself
3) ??? profit

Untested and uncompiled atm, but I have some confidence this might work

Fixed the thinkerlist issues by removing THINK_LIMBO. Delay-removed thinkers now stay in their list.

Also includes toaster's assorted fixes and improvements from target_painted_on_your_ass.

Signed-off-by: Nev3r <apophycens@gmail.com>

Set up a main thinker list and a polyobject mover list to test things up. Works so far, networking as well.

Signed-off-by: Nev3r <apophycens@gmail.com>

Hopefully I'm not missing anything.

Signed-off-by: Nev3r <apophycens@gmail.com>

* Fix a memory leak regarding implementation of SOC_ (improperly copypasted code from LUA_LoadLump!!)
* Optimise the repeated strlen usage into a single call, which is stored for later.

This now means:
* Lua strings longer than 1024 chars can now be read properly without awful crashes
* Lua strings with embedded zeros can be written/read without truncating anything (hopefully)

Fix SOC/Lua loading messages to display full names or even display at all (for PK3s at least), and otherwise some cleanup of existing code for the messages.

Also, I moved lua_lumploading on/offing to LUA_LoadFile.

* ArchiveTables: print an error if invalid key, to alert script author potentially
* UnArchiveTables: if the key is found to be nil after reading key and value, print an error and don't set them in the table

Apart from the fact that UnArchiveValue reads UINT16 for both anyway (which alone causes problems), but UINT8 isn't even enough to store the higher end of the object types list and definitely most of the states welp

To re-enable support for the above, uncomment the define HAVE_LUA_SEGS line in lua_script.h. Plain bbox userdata stuff is not disabled (though currently it's not used anyway)

my P_SearchBlockmap_* functions are now a single searchBlockmap function, you can choose between "objects" and "lines" with the first arg to decide what to iterate through. I also rearranged the argument order a bit for easy stack cleanup etc

I'll remove the old stuff later, don't worry, it's disabled for now

Stupid overlooked semicolons

Oh, and P_GetFFloorByID is now a thing, to assist with the above. Also added a suitable description for P_GetFFloorBySec

(yes, I changed my mind about doing it in public repo, probably too much hassle if we want 2.1.16 out soon)

(I'll fix sector.lines and ffloors and other things I missed in the public repo)

Still some issues with node.bbox, but the rest seems to work

segs[i] is the segs table, "for seg in segs.iterate" iterates through segs, you know the drill by now I hope

(no actual SLOC changes)

Fixes I_Assert failure crash due to hooks working differently now.

That's supposed to be run once a frame, not once per hook
per mobj per frame you moron. If you just run it seven
thousand times a frame, of course your framerate will drop.

Angles now go from 0 to 0xFFFF (360 degrees == FRACUNIT) instead
of using a full UINT32. Lua only has one number type, so signedness
gets in the way of using angle_t directly. This handling of angles
matches up with how ZDoom ACS scripting and the like does it.

I also changed all the integer casts and pushes of fixed_t to
their own macro in preperation for possible future seperation.