1
0
Fork 0
forked from fte/fteqw
fteqw/engine/common/gl_q2bsp.c

7933 lines
203 KiB
C
Raw Normal View History

#include "quakedef.h"
#ifndef SERVERONLY
#include "glquake.h"
#endif
#include "com_mesh.h"
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
#include "com_bih.h"
#define MAX_Q3MAP_INDICES 0x8000000 //just a sanity limit
#define MAX_Q3MAP_VERTEXES 0x800000 //just a sanity limit
//#define MAX_CM_PATCH_VERTS (4096)
//#define MAX_CM_FACES (MAX_Q2MAP_FACES)
#ifdef FTE_TARGET_WEB
#define MAX_CM_PATCHES (0x1000) //fixme
#else
#define MAX_CM_PATCHES (0x10000) //fixme
#endif
//#define MAX_CM_LEAFFACES (MAX_Q2MAP_LEAFFACES)
#define MAX_CM_AREAS MAX_Q2MAP_AREAS
//#define Q3SURF_NODAMAGE 0x00000001
//#define Q3SURF_SLICK 0x00000002
//#define Q3SURF_SKY 0x00000004
//#define Q3SURF_LADDER 0x00000008
//#define Q3SURF_NOIMPACT 0x00000010
//#define Q3SURF_NOMARKS 0x00000020
//#define Q3SURF_FLESH 0x00000040
#define Q3SURF_NODRAW 0x00000080 // don't generate a drawsurface at all
//#define Q3SURF_HINT 0x00000100
#define Q3SURF_SKIP 0x00000200 // completely ignore, allowing non-closed brushes
//#define Q3SURF_NOLIGHTMAP 0x00000400
//#define Q3SURF_POINTLIGHT 0x00000800
//#define Q3SURF_METALSTEPS 0x00001000
//#define Q3SURF_NOSTEPS 0x00002000
#define Q3SURF_NONSOLID 0x00004000 // don't collide against curves with this set
//#define Q3SURF_LIGHTFILTER 0x00008000
//#define Q3SURF_ALPHASHADOW 0x00010000
//#define Q3SURF_NODLIGHT 0x00020000
//#define Q3SURF_DUST 0x00040000
cvar_t q3bsp_surf_meshcollision_flag = CVARD("q3bsp_surf_meshcollision_flag", "0x80000000", "The surfaceparm flag(s) that enables q3bsp trisoup collision");
cvar_t q3bsp_surf_meshcollision_force = CVARD("q3bsp_surf_meshcollision_force", "0", "Force mesh-based collisions on all q3bsp trisoup surfaces.");
cvar_t q3bsp_mergeq3lightmaps = CVARD("q3bsp_mergelightmaps", "1", "Specifies whether to merge lightmaps into atlases in order to boost performance. Unfortunately this breaks tcgen on lightmap passes - if you care, set this to 0.");
cvar_t q3bsp_ignorestyles = CVARD("q3bsp_ignorestyles", "0", "Ignores multiple lightstyles in Raven's q3bsp variant(and derivatives) for better batch/rendering performance.");
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
cvar_t q3bsp_bihtraces = CVARFD("_q3bsp_bihtraces", /*FIXME: generate BIH leafs more carefully*/"0", CVAR_RENDERERLATCH, "Uses runtime-generated bih collision culling for faster traces.");
#if Q3SURF_NODRAW != TI_NODRAW
#error "nodraw isn't constant"
#endif
extern cvar_t r_shadow_bumpscale_basetexture;
//these are in model.c (or gl_model.c)
qboolean Mod_LoadVertexes (model_t *loadmodel, qbyte *mod_base, lump_t *l);
qboolean Mod_LoadVertexNormals (model_t *loadmodel, bspx_header_t *bspx, qbyte *mod_base, lump_t *l);
qboolean Mod_LoadEdges (model_t *loadmodel, qbyte *mod_base, lump_t *l, qboolean lm);
qboolean Mod_LoadMarksurfaces (model_t *loadmodel, qbyte *mod_base, lump_t *l, qboolean lm);
qboolean Mod_LoadSurfedges (model_t *loadmodel, qbyte *mod_base, lump_t *l);
void Mod_LoadEntities (model_t *loadmodel, qbyte *mod_base, lump_t *l);
extern void BuildLightMapGammaTable (float g, float c);
#if defined(Q2BSPS) || defined(Q3BSPS)
static qboolean CM_NativeTrace(model_t *model, int forcehullnum, const framestate_t *framestate, const vec3_t axis[3], const vec3_t start, const vec3_t end, const vec3_t mins, const vec3_t maxs, qboolean capsule, unsigned int contents, trace_t *trace);
static unsigned int CM_NativeContents(struct model_s *model, int hulloverride, const framestate_t *framestate, const vec3_t axis[3], const vec3_t p, const vec3_t mins, const vec3_t maxs);
static unsigned int Q2BSP_PointContents(model_t *mod, const vec3_t axis[3], const vec3_t p);
static int CM_PointCluster (model_t *mod, const vec3_t p, int *area);
static void CM_InfoForPoint (struct model_s *mod, vec3_t pos, int *area, int *cluster, unsigned int *contentbits);
struct cminfo_s;
void CM_Init(void);
static qboolean VARGS CM_AreasConnected (struct model_s *mod, unsigned int area1, unsigned int area2);
static size_t CM_WriteAreaBits (struct model_s *mod, qbyte *buffer, size_t buffersize, int area, qboolean merge);
static qbyte *CM_ClusterPVS (struct model_s *mod, int cluster, pvsbuffer_t *buffer, pvsmerge_t merge);
static qbyte *CM_ClusterPHS (struct model_s *mod, int cluster, pvsbuffer_t *buffer);
//for gamecode to control portals/areas
static void CM_SetAreaPortalState (model_t *mod, unsigned int portalnum, unsigned int area1, unsigned int area2, qboolean open);
//for saved games to write the raw state.
static size_t CM_SaveAreaPortalBlob (model_t *mod, void **data);
static size_t CM_LoadAreaPortalBlob (model_t *mod, void *ptr, size_t ptrsize);
2023-03-27 17:22:00 +00:00
#ifdef HAVE_SERVER
static unsigned int Q23BSP_FatPVS(model_t *mod, const vec3_t org, pvsbuffer_t *buffer, qboolean merge);
static qboolean Q23BSP_EdictInFatPVS(model_t *mod, const struct pvscache_s *ent, const qbyte *pvs, const int *areas);
static void Q23BSP_FindTouchedLeafs(model_t *mod, struct pvscache_s *ent, const float *mins, const float *maxs);
2023-03-27 17:22:00 +00:00
static qboolean CM_HeadnodeVisible (struct model_s *mod, int nodenum, const qbyte *visbits);
#endif
#ifdef HAVE_CLIENT
static void CM_PrepareFrame(model_t *mod, refdef_t *refdef, int area, int viewclusters[2], pvsbuffer_t *vis, qbyte **entvis_out, qbyte **surfvis_out);
extern void Q2BSP_GenerateShadowMesh(model_t *mod, dlight_t *dl, const qbyte *lightvis, qbyte *litvis, void(*callback)(msurface_t*));
extern void Q3BSP_GenerateShadowMesh(model_t *mod, dlight_t *dl, const qbyte *lightvis, qbyte *litvis, void(*callback)(msurface_t*));
#endif
#endif
float RadiusFromBounds (const vec3_t mins, const vec3_t maxs)
{
int i;
vec3_t corner;
for (i=0 ; i<3 ; i++)
{
corner[i] = fabs(mins[i]) > fabs(maxs[i]) ? fabs(mins[i]) : fabs(maxs[i]);
}
return Length (corner);
}
void CalcSurfaceExtents (model_t *mod, msurface_t *s)
{
float mins[2], maxs[2], val;
int i,j, e;
mvertex_t *v;
mtexinfo_t *tex;
int bmins[2], bmaxs[2];
added qccgui icon (courtesy of shpuld) isolated transfers a little. added preliminary support for r1q2 and q2pro protocols (no zlib). fixed some missing q2 hud configstrings. fixed models attached to particles not appearing the first time around. fixed halos appearing on {foo fullbright textures. fix particle crash bug from missing particlecubes. implement r_converteffectinfo (instead of r_importeffectinfo with r_exportalleffects) for convenience. also now more complete and more accurate. loadfont no longer auto-allocates slot 0. fix issues with r_dynamic -1 don't restart audio on vid_reload (only reload it). this fixes audio issues with capturedemo with gamedir changes. fix possible crash from loading sounds during an audio reload/restart. now listens on both ports 27500 and 26000 when running quake, by default. sv_port or -port args will override. work around surfedge corruption in some instance of the slide4 map that dimman found. attempt to nudge players to travel at at least 200qu when travelling vertically through portals. this should help alieviate gravity-induced stuttering issues. fix nested function types used in conjunction with op_state. now assumes static within the parent. optimise array indexes slightly. fix issue with ptr[float].foo fix some qcc bugs to do with implicit type conversions fix a portal / angle protocol extension issue, so legacy clients can use portals, they just can't see the other side (nor predict them). fix q2 victory.pcx issue at the end of the game. use abort/sigabort instead of sigsegv on linux sys_errors. this should help get people to report stuff better. reformat the nq server advertising to be more helpful. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4996 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-12-28 17:41:39 +00:00
int idx;
mins[0] = mins[1] = 999999;
maxs[0] = maxs[1] = -999999;
tex = s->texinfo;
for (i=0 ; i<s->numedges ; i++)
{
e = mod->surfedges[s->firstedge+i];
added qccgui icon (courtesy of shpuld) isolated transfers a little. added preliminary support for r1q2 and q2pro protocols (no zlib). fixed some missing q2 hud configstrings. fixed models attached to particles not appearing the first time around. fixed halos appearing on {foo fullbright textures. fix particle crash bug from missing particlecubes. implement r_converteffectinfo (instead of r_importeffectinfo with r_exportalleffects) for convenience. also now more complete and more accurate. loadfont no longer auto-allocates slot 0. fix issues with r_dynamic -1 don't restart audio on vid_reload (only reload it). this fixes audio issues with capturedemo with gamedir changes. fix possible crash from loading sounds during an audio reload/restart. now listens on both ports 27500 and 26000 when running quake, by default. sv_port or -port args will override. work around surfedge corruption in some instance of the slide4 map that dimman found. attempt to nudge players to travel at at least 200qu when travelling vertically through portals. this should help alieviate gravity-induced stuttering issues. fix nested function types used in conjunction with op_state. now assumes static within the parent. optimise array indexes slightly. fix issue with ptr[float].foo fix some qcc bugs to do with implicit type conversions fix a portal / angle protocol extension issue, so legacy clients can use portals, they just can't see the other side (nor predict them). fix q2 victory.pcx issue at the end of the game. use abort/sigabort instead of sigsegv on linux sys_errors. this should help get people to report stuff better. reformat the nq server advertising to be more helpful. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4996 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-12-28 17:41:39 +00:00
idx = e < 0;
if (idx)
e = -e;
if (e < 0 || e >= mod->numedges)
v = &mod->vertexes[0];
else
added qccgui icon (courtesy of shpuld) isolated transfers a little. added preliminary support for r1q2 and q2pro protocols (no zlib). fixed some missing q2 hud configstrings. fixed models attached to particles not appearing the first time around. fixed halos appearing on {foo fullbright textures. fix particle crash bug from missing particlecubes. implement r_converteffectinfo (instead of r_importeffectinfo with r_exportalleffects) for convenience. also now more complete and more accurate. loadfont no longer auto-allocates slot 0. fix issues with r_dynamic -1 don't restart audio on vid_reload (only reload it). this fixes audio issues with capturedemo with gamedir changes. fix possible crash from loading sounds during an audio reload/restart. now listens on both ports 27500 and 26000 when running quake, by default. sv_port or -port args will override. work around surfedge corruption in some instance of the slide4 map that dimman found. attempt to nudge players to travel at at least 200qu when travelling vertically through portals. this should help alieviate gravity-induced stuttering issues. fix nested function types used in conjunction with op_state. now assumes static within the parent. optimise array indexes slightly. fix issue with ptr[float].foo fix some qcc bugs to do with implicit type conversions fix a portal / angle protocol extension issue, so legacy clients can use portals, they just can't see the other side (nor predict them). fix q2 victory.pcx issue at the end of the game. use abort/sigabort instead of sigsegv on linux sys_errors. this should help get people to report stuff better. reformat the nq server advertising to be more helpful. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4996 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-12-28 17:41:39 +00:00
v = &mod->vertexes[mod->edges[e].v[idx]];
for (j=0 ; j<2 ; j++)
{
//doubles should replicate win32/x87 compiler 80-bit precision better. We have to hope that win64 compilers use the same precision.
val = DotProduct_Double(v->position, tex->vecs[j]) + tex->vecs[j][3];
if (val < mins[j])
mins[j] = val;
if (val > maxs[j])
maxs[j] = val;
}
}
for (i=0 ; i<2 ; i++)
{
bmins[i] = floor(mins[i]/(1<<s->lmshift));
bmaxs[i] = ceil(maxs[i]/(1<<s->lmshift));
s->texturemins[i] = bmins[i] << s->lmshift;
s->extents[i] = (bmaxs[i] - bmins[i]);
// if ( !(tex->flags & TEX_SPECIAL) && s->extents[i] > 17 ) //vanilla used 16(+1), glquake used 17(+1). FTE uses 255(+1), but we omit lightmapping instead of crashing if its larger than our limit, so we omit the check here. different engines use different limits here, many of them make no sense.
// Con_Printf ("Bad surface extents (texture %s, more than %i lightmap samples)\n", s->texinfo->texture->name, s->extents[i]);
s->extents[i] <<= s->lmshift;
}
}
void AddPointToBounds (const vec3_t v, vec3_t mins, vec3_t maxs)
{
int i;
vec_t val;
for (i=0 ; i<3 ; i++)
{
val = v[i];
if (val < mins[i])
mins[i] = val;
if (val > maxs[i])
maxs[i] = val;
}
}
void ClearBounds (vec3_t mins, vec3_t maxs)
{
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
mins[0] = mins[1] = mins[2] = FLT_MAX;
maxs[0] = maxs[1] = maxs[2] = -FLT_MAX;
}
void Mod_SortShaders(model_t *mod)
{
//surely this isn't still needed?
texture_t *textemp;
int i, j;
//sort loadmodel->textures
for (i = 0; i < mod->numtextures; i++)
{
for (j = i+1; j < mod->numtextures; j++)
{
if ((mod->textures[i]->shader && mod->textures[j]->shader) && (mod->textures[j]->shader->sort < mod->textures[i]->shader->sort))
{
textemp = mod->textures[j];
mod->textures[j] = mod->textures[i];
mod->textures[i] = textemp;
}
}
}
}
#if defined(Q2BSPS) || defined(Q3BSPS)
#ifdef IMAGEFMT_PCX
qbyte *ReadPCXPalette(qbyte *buf, int len, qbyte *out);
#endif
#ifdef SERVERONLY
#define Host_Error SV_Error
#endif
extern qbyte *mod_base;
#define capsuledist(dist,plane,mins,maxs) \
case shape_iscapsule: \
dist = DotProduct(trace_up, plane->normal); \
dist = dist*(trace_capsulesize[(dist<0)?1:2]) - trace_capsulesize[0]; \
dist = plane->dist - dist; \
break;
#ifdef Q2BSPS
#ifdef HAVE_CLIENT
static unsigned char q2_palette[256*3];
#endif
#endif
/*
typedef struct q2csurface_s
{
char name[16];
int flags;
int value;
} q2csurface_t;
*/
typedef struct {
char shader[64];
int brushNum;
int visibleSide; // the brush side that ray tests need to clip against (-1 == none)
} dfog_t;
#ifdef Q2BSPS
typedef struct
{
int numareaportals;
int firstareaportal;
} q2carea_t;
#endif
#ifdef Q3BSPS
typedef struct
{
int numareaportals[MAX_CM_AREAS];
} q3carea_t;
#endif
typedef struct
{
int floodnum; // if two areas have equal floodnums, they are connected
int floodvalid; // flags the area as having been visited (sequence numbers matching prv->floodvalid)
} careaflood_t;
typedef struct
{
int facetype;
int numverts;
int firstvert;
int shadernum;
union
{
struct
{
unsigned short cp[2];
unsigned short fixedres[2];
} patch;
struct
{
int firstindex;
int numindicies;
} soup;
};
} q3cface_t;
typedef struct cmodel_s
{
vec3_t mins, maxs;
vec3_t origin; // for sounds or lights
mnode_t *headnode;
mleaf_t *headleaf;
int numsurfaces;
int firstsurface;
int firstbrush; //q3 submodels are considered small enough that you will never need to walk any sort of tree.
int num_brushes;//the brushes are checked instead.
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
//these things are generated at load time.
int firstpatch;
int num_patches;
int firstcmesh;
int num_cmeshes;
} cmodel_t;
/*used to trace*/
static int checkcount;
typedef struct cminfo_s
{
int numbrushsides;
q2cbrushside_t *brushsides;
q2mapsurface_t *surfaces;
int numleafbrushes;
q2cbrush_t **leafbrushes;
int numcmodels;
cmodel_t *cmodels;
int numbrushes;
q2cbrush_t *brushes;
int numvisibility;
q2dvis_t *q2vis;
q3dvis_t *q3pvs;
q3dvis_t *q3phs;
qbyte *phscalced;
int numareas;
int floodvalid;
careaflood_t areaflood[MAX_CM_AREAS];
#ifdef Q3BSPS
//q3's areas are simple bidirectional area1/area2 pairs. refcounted (so two areas can have two doors/openings)
q3carea_t q3areas[MAX_CM_AREAS];
#endif
#ifdef Q2BSPS
//q2's areas have a list of portals that open into other areas.
q2carea_t *q2areas; //indexes into q2areaportals for flooding
size_t numq2areaportals;
q2dareaportal_t *q2areaportals;
//and this is the state that is actually changed. booleans.
qbyte q2portalopen[MAX_Q2MAP_AREAPORTALS]; //memset will work if it's a qbyte, really it should be a qboolean
#endif
//list of mesh surfaces within the leaf
q3cmesh_t cmeshes[MAX_CM_PATCHES];
int numcmeshes;
int *leafcmeshes;
int numleafcmeshes;
int maxleafcmeshes;
//FIXME: remove the below
//(deprecated) patch collisions
q3cpatch_t patches[MAX_CM_PATCHES];
int numpatches;
int *leafpatches;
int numleafpatches;
int maxleafpatches;
//FIXME: remove the above
qboolean mapisq3;
#ifdef Q3BSPS
//this is for loading stuff. it used to be globals, but we have threads now. and multiple q3bsps at the same time is a problem.
int numvertexes;
vecV_t *verts; //3points
vec2_t *vertstmexcoords;
vec2_t *vertlstmexcoords[MAXRLIGHTMAPS];
vec4_t *colors4f_array[MAXRLIGHTMAPS];
vec3_t *normals_array;
//vec3_t *map_svector_array;
//vec3_t *map_tvector_array;
index_t *surfindexes;
//int map_numsurfindexes;
q3cface_t *faces;
int numfaces;
#endif
#ifdef HAVE_CLIENT
int oldclusters[2];
qbyte *oldvis;
#endif
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
// struct bihnode_s *bihnodes;
} cminfo_t;
static q2mapsurface_t nullsurface;
cvar_t map_noareas = CVAR("map_noareas", "0"); //1 for lack of mod support.
cvar_t map_noCurves = CVARF("map_noCurves", "0", CVAR_CHEAT);
cvar_t map_autoopenportals = CVARD("map_autoopenportals", "0", "When set to 1, force-opens all area portals. Normally these start closed and are opened by doors when they move, but this requires the gamecode to signal this."); //1 for lack of mod support.
cvar_t r_subdivisions = CVAR("r_subdivisions", "2");
static int CM_NumInlineModels (model_t *model);
static cmodel_t *CM_InlineModel (model_t *model, char *name);
static void CM_InitBoxHull (void);
static void CM_FinalizeBrush(q2cbrush_t *brush);
static void FloodAreaConnections (cminfo_t *prv);
qboolean BoundsIntersect (const vec3_t mins1, const vec3_t maxs1, const vec3_t mins2, const vec3_t maxs2)
{
return (mins1[0] <= maxs2[0] && mins1[1] <= maxs2[1] && mins1[2] <= maxs2[2] &&
maxs1[0] >= mins2[0] && maxs1[1] >= mins2[1] && maxs1[2] >= mins2[2]);
}
#ifdef Q3BSPS
static int PlaneTypeForNormal ( vec3_t normal )
{
vec_t ax, ay, az;
// NOTE: should these have an epsilon around 1.0?
if ( normal[0] >= 1.0)
return PLANE_X;
if ( normal[1] >= 1.0 )
return PLANE_Y;
if ( normal[2] >= 1.0 )
return PLANE_Z;
ax = fabs( normal[0] );
ay = fabs( normal[1] );
az = fabs( normal[2] );
if ( ax >= ay && ax >= az )
return PLANE_ANYX;
if ( ay >= ax && ay >= az )
return PLANE_ANYY;
return PLANE_ANYZ;
}
void CategorizePlane ( mplane_t *plane )
{
int i;
plane->signbits = 0;
plane->type = PLANE_ANYZ;
for (i = 0; i < 3; i++)
{
if (plane->normal[i] < 0)
plane->signbits |= 1<<i;
if (plane->normal[i] == 1.0f)
plane->type = i;
}
plane->type = PlaneTypeForNormal(plane->normal);
}
static void PlaneFromPoints ( vec3_t verts[3], mplane_t *plane )
{
vec3_t v1, v2;
VectorSubtract( verts[1], verts[0], v1 );
VectorSubtract( verts[2], verts[0], v2 );
CrossProduct( v2, v1, plane->normal );
VectorNormalize( plane->normal );
plane->dist = DotProduct( verts[0], plane->normal );
}
/*
===============
Patch_FlatnessTest
===============
*/
static int Patch_FlatnessTest( float maxflat2, const float *point0, const float *point1, const float *point2 )
{
float d;
int ft0, ft1;
vec3_t t, n;
vec3_t v1, v2, v3;
VectorSubtract( point2, point0, n );
if( !VectorNormalize( n ) )
return 0;
VectorSubtract( point1, point0, t );
d = -DotProduct( t, n );
VectorMA( t, d, n, t );
if( DotProduct( t, t ) < maxflat2 )
return 0;
VectorAvg( point1, point0, v1 );
VectorAvg( point2, point1, v2 );
VectorAvg( v1, v2, v3 );
ft0 = Patch_FlatnessTest( maxflat2, point0, v1, v3 );
ft1 = Patch_FlatnessTest( maxflat2, v3, v2, point2 );
return 1 + (int)( floor( max( ft0, ft1 ) ) + 0.5f );
}
/*
===============
Patch_GetFlatness
===============
*/
static void Patch_GetFlatness( float maxflat, const float *points, int comp, const unsigned short *patch_cp, int *flat )
{
int i, p, u, v;
float maxflat2 = maxflat * maxflat;
flat[0] = flat[1] = 0;
for( v = 0; v < patch_cp[1] - 1; v += 2 )
{
for( u = 0; u < patch_cp[0] - 1; u += 2 )
{
p = v * patch_cp[0] + u;
i = Patch_FlatnessTest( maxflat2, &points[p*comp], &points[( p+1 )*comp], &points[( p+2 )*comp] );
flat[0] = max( flat[0], i );
i = Patch_FlatnessTest( maxflat2, &points[( p+patch_cp[0] )*comp], &points[( p+patch_cp[0]+1 )*comp], &points[( p+patch_cp[0]+2 )*comp] );
flat[0] = max( flat[0], i );
i = Patch_FlatnessTest( maxflat2, &points[( p+2*patch_cp[0] )*comp], &points[( p+2*patch_cp[0]+1 )*comp], &points[( p+2*patch_cp[0]+2 )*comp] );
flat[0] = max( flat[0], i );
i = Patch_FlatnessTest( maxflat2, &points[p*comp], &points[( p+patch_cp[0] )*comp], &points[( p+2*patch_cp[0] )*comp] );
flat[1] = max( flat[1], i );
i = Patch_FlatnessTest( maxflat2, &points[( p+1 )*comp], &points[( p+patch_cp[0]+1 )*comp], &points[( p+2*patch_cp[0]+1 )*comp] );
flat[1] = max( flat[1], i );
i = Patch_FlatnessTest( maxflat2, &points[( p+2 )*comp], &points[( p+patch_cp[0]+2 )*comp], &points[( p+2*patch_cp[0]+2 )*comp] );
flat[1] = max( flat[1], i );
}
}
}
/*
===============
Patch_Evaluate_QuadricBezier
===============
*/
static void Patch_Evaluate_QuadricBezier( float t, const vec_t *point0, const vec_t *point1, const vec_t *point2, vec_t *out, int comp )
{
int i;
vec_t qt = t * t;
vec_t dt = 2.0f * t, tt, tt2;
tt = 1.0f - dt + qt;
tt2 = dt - 2.0f * qt;
for( i = 0; i < comp; i++ )
out[i] = point0[i] * tt + point1[i] * tt2 + point2[i] * qt;
}
/*
===============
Patch_Evaluate
===============
*/
2023-04-17 02:39:30 +00:00
#define Patch_Evaluate(p,numcp,tess,dest, comp) Patch_EvaluateStride(p,comp,numcp,tess,dest,comp,comp)
static void Patch_EvaluateStride(const vec_t *p, int pstride, const unsigned short *numcp, const int *tess, vec_t *dest, int deststride, int comp)
{
int num_patches[2], num_tess[2];
int index[3], dstpitch, i, u, v, x, y;
float s, t, step[2];
vec_t *tvec, *tvec2;
const vec_t *pv[3][3];
vec4_t v1, v2, v3;
if (!tess[0] || !tess[1])
{ //not really a patch
2023-04-17 02:39:30 +00:00
for( u = 0; u < numcp[1]*numcp[0]; u++, dest += deststride, p += pstride)
for( i = 0; i < comp; i++ )
dest[i] = p[i];
return;
}
num_patches[0] = numcp[0] / 2;
num_patches[1] = numcp[1] / 2;
2023-04-17 02:39:30 +00:00
dstpitch = ( num_patches[0] * tess[0] + 1 ) * deststride;
step[0] = 1.0f / (float)tess[0];
step[1] = 1.0f / (float)tess[1];
for( v = 0; v < num_patches[1]; v++ )
{
// last patch has one more row
if( v < num_patches[1] - 1 )
num_tess[1] = tess[1];
else
num_tess[1] = tess[1] + 1;
for( u = 0; u < num_patches[0]; u++ )
{
// last patch has one more column
if( u < num_patches[0] - 1 )
num_tess[0] = tess[0];
else
num_tess[0] = tess[0] + 1;
index[0] = ( v * numcp[0] + u ) * 2;
index[1] = index[0] + numcp[0];
index[2] = index[1] + numcp[0];
// current 3x3 patch control points
for( i = 0; i < 3; i++ )
{
2023-04-17 02:39:30 +00:00
pv[i][0] = &p[( index[0]+i ) * pstride];
pv[i][1] = &p[( index[1]+i ) * pstride];
pv[i][2] = &p[( index[2]+i ) * pstride];
}
2023-04-17 02:39:30 +00:00
tvec = dest + v * tess[1] * dstpitch + u * tess[0] * deststride;
for( y = 0, t = 0.0f; y < num_tess[1]; y++, t += step[1], tvec += dstpitch )
{
Patch_Evaluate_QuadricBezier( t, pv[0][0], pv[0][1], pv[0][2], v1, comp );
Patch_Evaluate_QuadricBezier( t, pv[1][0], pv[1][1], pv[1][2], v2, comp );
Patch_Evaluate_QuadricBezier( t, pv[2][0], pv[2][1], pv[2][2], v3, comp );
2023-04-17 02:39:30 +00:00
for( x = 0, tvec2 = tvec, s = 0.0f; x < num_tess[0]; x++, s += step[0], tvec2 += deststride )
Patch_Evaluate_QuadricBezier( s, v1, v2, v3, tvec2, comp );
}
}
}
}
2023-04-17 02:39:30 +00:00
#ifdef TERRAIN
#include "gl_terrain.h"
patchtessvert_t *PatchInfo_Evaluate(const qcpatchvert_t *cp, const unsigned short patch_cp[2], const short subdiv[2], unsigned short *size)
{
int step[2], flat[2];
float subdivlevel;
unsigned int numverts;
patchtessvert_t *out;
int i;
if (subdiv[0]>=0 && subdiv[1]>=0)
{ //fixed
step[0] = subdiv[0];
step[1] = subdiv[1];
}
else
{
// find the degree of subdivision in the u and v directions
subdivlevel = bound(1, r_subdivisions.ival, 15);
Patch_GetFlatness ( subdivlevel, cp->v, sizeof(*cp)/sizeof(vec_t), patch_cp, flat );
step[0] = 1 << flat[0];
step[1] = 1 << flat[1];
}
if (!step[0] || !step[1])
{
size[0] = patch_cp[0];
size[1] = patch_cp[1];
}
else
{
size[0] = ( patch_cp[0] >> 1 ) * step[0] + 1;
size[1] = ( patch_cp[1] >> 1 ) * step[1] + 1;
}
if( size[0] <= 0 || size[1] <= 0 )
return NULL;
numverts = (unsigned int)size[0] * size[1];
// fill in
out = BZ_Malloc(sizeof(*out) * numverts);
for (i = 0; i < numverts*sizeof(*out)/sizeof(vec_t); i++)
((vec_t *)out)[i] = -1;
Patch_EvaluateStride ( cp->v, sizeof(*cp)/sizeof(vec_t), patch_cp, step, out->v, sizeof(*out)/sizeof(vec_t), countof(cp->v));
Patch_EvaluateStride ( cp->rgba, sizeof(*cp)/sizeof(vec_t), patch_cp, step, out->rgba, sizeof(*out)/sizeof(vec_t), countof(cp->rgba));
Patch_EvaluateStride ( cp->tc, sizeof(*cp)/sizeof(vec_t), patch_cp, step, out->tc, sizeof(*out)/sizeof(vec_t), countof(cp->tc));
return out;
}
unsigned int PatchInfo_EvaluateIndexes(const unsigned short *size, index_t *out_indexes)
{
int i, u, v, p;
// compute new indexes avoiding adding invalid triangles
unsigned int numindexes = 0;
index_t *indexes = out_indexes;
for (v = 0, i = 0; v < size[1]-1; v++)
{
for (u = 0; u < size[0]-1; u++, i += 6)
{
indexes[0] = p = v * size[0] + u;
indexes[1] = p + size[0];
indexes[2] = p + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
indexes[0] = p + 1;
indexes[1] = p + size[0];
indexes[2] = p + size[0] + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
}
}
return numindexes;
}
#endif
#define PLANE_NORMAL_EPSILON 0.00001
#define PLANE_DIST_EPSILON 0.01
static qboolean ComparePlanes( const vec3_t p1normal, vec_t p1dist, const vec3_t p2normal, vec_t p2dist )
{
if( fabs( p1normal[0] - p2normal[0] ) < PLANE_NORMAL_EPSILON
&& fabs( p1normal[1] - p2normal[1] ) < PLANE_NORMAL_EPSILON
&& fabs( p1normal[2] - p2normal[2] ) < PLANE_NORMAL_EPSILON
&& fabs( p1dist - p2dist ) < PLANE_DIST_EPSILON )
return true;
return false;
}
static void SnapVector( vec3_t normal )
{
int i;
for( i = 0; i < 3; i++ )
{
if( fabs( normal[i] - 1 ) < PLANE_NORMAL_EPSILON )
{
VectorClear( normal );
normal[i] = 1;
break;
}
if( fabs( normal[i] + 1 ) < PLANE_NORMAL_EPSILON )
{
VectorClear( normal );
normal[i] = -1;
break;
}
}
}
#define Q_rint( x ) ( ( x ) < 0 ? ( (int)( ( x )-0.5f ) ) : ( (int)( ( x )+0.5f ) ) )
static void SnapPlane( vec3_t normal, vec_t *dist )
{
SnapVector( normal );
if( fabs( *dist - Q_rint( *dist ) ) < PLANE_DIST_EPSILON )
{
*dist = Q_rint( *dist );
}
}
/*
===============================================================================
PATCH LOADING
===============================================================================
*/
#define MAX_FACET_PLANES 32
#define cm_subdivlevel 15
/*
* CM_CreateFacetFromPoints
*/
static int CM_CreateFacetFromPoints(q2cbrush_t *facet, vec3_t *verts, int numverts, q2mapsurface_t *shaderref, mplane_t *brushplanes )
{
int i, j;
int axis, dir;
vec3_t normal;
float d, dist;
mplane_t mainplane;
vec3_t vec, vec2;
int numbrushplanes;
// set default values for brush
facet->numsides = 0;
facet->brushside = NULL;
facet->contents = shaderref->c.value;
// calculate plane for this triangle
PlaneFromPoints( verts, &mainplane );
if( ComparePlanes( mainplane.normal, mainplane.dist, vec3_origin, 0 ) )
return 0;
// test a quad case
if( numverts > 3 )
{
d = DotProduct( verts[3], mainplane.normal ) - mainplane.dist;
if( d < -0.1 || d > 0.1 )
return 0;
if( 0 )
{
vec3_t v[3];
mplane_t plane;
// try different combinations of planes
for( i = 1; i < 4; i++ )
{
VectorCopy( verts[i], v[0] );
VectorCopy( verts[( i+1 )%4], v[1] );
VectorCopy( verts[( i+2 )%4], v[2] );
PlaneFromPoints( v, &plane );
if( fabs( DotProduct( mainplane.normal, plane.normal ) ) < 0.9 )
return 0;
}
}
}
numbrushplanes = 0;
// add front plane
SnapPlane( mainplane.normal, &mainplane.dist );
VectorCopy( mainplane.normal, brushplanes[numbrushplanes].normal );
brushplanes[numbrushplanes].dist = mainplane.dist; numbrushplanes++;
// calculate mins & maxs
ClearBounds( facet->absmins, facet->absmaxs );
for( i = 0; i < numverts; i++ )
AddPointToBounds( verts[i], facet->absmins, facet->absmaxs );
// add the axial planes
for( axis = 0; axis < 3; axis++ )
{
for( dir = -1; dir <= 1; dir += 2 )
{
for( i = 0; i < numbrushplanes; i++ )
{
if( brushplanes[i].normal[axis] == dir )
break;
}
if( i == numbrushplanes )
{
VectorClear( normal );
normal[axis] = dir;
if( dir == 1 )
dist = facet->absmaxs[axis];
else
dist = -facet->absmins[axis];
VectorCopy( normal, brushplanes[numbrushplanes].normal );
brushplanes[numbrushplanes].dist = dist; numbrushplanes++;
}
}
}
// add the edge bevels
for( i = 0; i < numverts; i++ )
{
j = ( i + 1 ) % numverts;
// k = ( i + 2 ) % numverts;
VectorSubtract( verts[i], verts[j], vec );
if( VectorNormalize( vec ) < 0.5 )
continue;
SnapVector( vec );
for( j = 0; j < 3; j++ )
{
if( vec[j] == 1 || vec[j] == -1 )
break; // axial
}
if( j != 3 )
continue; // only test non-axial edges
// try the six possible slanted axials from this edge
for( axis = 0; axis < 3; axis++ )
{
for( dir = -1; dir <= 1; dir += 2 )
{
// construct a plane
VectorClear( vec2 );
vec2[axis] = dir;
CrossProduct( vec, vec2, normal );
if( VectorNormalize( normal ) < 0.5 )
continue;
dist = DotProduct( verts[i], normal );
for( j = 0; j < numbrushplanes; j++ )
{
// if this plane has already been used, skip it
if( ComparePlanes( brushplanes[j].normal, brushplanes[j].dist, normal, dist ) )
break;
}
if( j != numbrushplanes )
continue;
// if all other points are behind this plane, it is a proper edge bevel
for( j = 0; j < numverts; j++ )
{
if( j != i )
{
d = DotProduct( verts[j], normal ) - dist;
if( d > 0.1 )
break; // point in front: this plane isn't part of the outer hull
}
}
if( j != numverts )
continue;
// add this plane
VectorCopy( normal, brushplanes[numbrushplanes].normal );
brushplanes[numbrushplanes].dist = dist; numbrushplanes++;
if( numbrushplanes == MAX_FACET_PLANES )
break;
}
}
}
return ( facet->numsides = numbrushplanes );
}
/*
* CM_CreatePatch
*/
static void CM_CreatePatch(model_t *loadmodel, q3cpatch_t *patch, q2mapsurface_t *shaderref, const vec_t *verts, const unsigned short *patch_cp, const unsigned short *patch_subdiv)
{
int step[2], size[2], flat[2];
int i, j, k ,u, v;
int numsides, totalsides;
q2cbrush_t *facets, *facet;
vecV_t *points;
vec3_t tverts[4];
qbyte *data;
mplane_t *brushplanes;
float subdivlevel;
patch->surface = shaderref;
if (patch_subdiv)
{ //fixed
step[0] = patch_subdiv[0];
step[1] = patch_subdiv[1];
}
else
{
// find the degree of subdivision in the u and v directions
subdivlevel = cm_subdivlevel;//r_subdivisions.value;
if ( subdivlevel < 1 )
subdivlevel = 1;
Patch_GetFlatness( subdivlevel, verts, sizeof(vecV_t)/sizeof(vec_t), patch_cp, flat );
step[0] = 1 << flat[0];
step[1] = 1 << flat[1];
}
if (!step[0] || !step[1])
{
size[0] = patch_cp[0];
size[1] = patch_cp[1];
}
else
{
size[0] = ( patch_cp[0] >> 1 ) * step[0] + 1;
size[1] = ( patch_cp[1] >> 1 ) * step[1] + 1;
}
if( size[0] <= 0 || size[1] <= 0 )
return;
data = BZ_Malloc( size[0] * size[1] * sizeof( vecV_t ) +
( size[0]-1 ) * ( size[1]-1 ) * 2 * ( sizeof( q2cbrush_t ) + 32 * sizeof( mplane_t ) ) );
points = ( vecV_t * )data; data += size[0] * size[1] * sizeof( vecV_t );
facets = ( q2cbrush_t * )data; data += ( size[0]-1 ) * ( size[1]-1 ) * 2 * sizeof( q2cbrush_t );
brushplanes = ( mplane_t * )data; data += ( size[0]-1 ) * ( size[1]-1 ) * 2 * MAX_FACET_PLANES * sizeof( mplane_t );
// fill in
Patch_Evaluate(verts, patch_cp, step, points[0], sizeof(vecV_t)/sizeof(vec_t));
totalsides = 0;
patch->numfacets = 0;
patch->facets = NULL;
ClearBounds( patch->absmins, patch->absmaxs );
// create a set of facets
for( v = 0; v < size[1]-1; v++ )
{
for( u = 0; u < size[0]-1; u++ )
{
i = v * size[0] + u;
VectorCopy( points[i], tverts[0] );
VectorCopy( points[i + size[0]], tverts[1] );
VectorCopy( points[i + size[0] + 1], tverts[2] );
VectorCopy( points[i + 1], tverts[3] );
for( i = 0; i < 4; i++ )
AddPointToBounds( tverts[i], patch->absmins, patch->absmaxs );
// try to create one facet from a quad
numsides = CM_CreateFacetFromPoints( &facets[patch->numfacets], tverts, 4, shaderref, brushplanes + totalsides );
if( !numsides )
{ // create two facets from triangles
VectorCopy( tverts[3], tverts[2] );
numsides = CM_CreateFacetFromPoints( &facets[patch->numfacets], tverts, 3, shaderref, brushplanes + totalsides );
if( numsides )
{
totalsides += numsides;
patch->numfacets++;
}
VectorCopy( tverts[2], tverts[0] );
VectorCopy( points[v *size[0] + u + size[0] + 1], tverts[2] );
numsides = CM_CreateFacetFromPoints( &facets[patch->numfacets], tverts, 3, shaderref, brushplanes + totalsides );
}
if( numsides )
{
totalsides += numsides;
patch->numfacets++;
}
}
}
if (patch->numfacets)
{
qbyte *data;
data = ZG_Malloc(&loadmodel->memgroup, patch->numfacets * sizeof( q2cbrush_t ) + totalsides * ( sizeof( q2cbrushside_t ) + sizeof( mplane_t ) ));
patch->facets = ( q2cbrush_t * )data; data += patch->numfacets * sizeof( q2cbrush_t );
memcpy( patch->facets, facets, patch->numfacets * sizeof( q2cbrush_t ) );
for( i = 0, k = 0, facet = patch->facets; i < patch->numfacets; i++, facet++ )
{
mplane_t *planes;
q2cbrushside_t *s;
facet->brushside = ( q2cbrushside_t * )data; data += facet->numsides * sizeof( q2cbrushside_t );
planes = ( mplane_t * )data; data += facet->numsides * sizeof( mplane_t );
for( j = 0, s = facet->brushside; j < facet->numsides; j++, s++ )
{
planes[j] = brushplanes[k++];
s->plane = &planes[j];
SnapPlane( s->plane->normal, &s->plane->dist );
CategorizePlane( s->plane );
s->surface = shaderref;
}
}
for( i = 0; i < 3; i++ )
{
// spread the mins / maxs by a pixel
patch->absmins[i] -= 1;
patch->absmaxs[i] += 1;
}
}
BZ_Free( points );
}
//======================================================
static qboolean CM_CreatePatchForFace (model_t *loadmodel, cminfo_t *prv, mleaf_t *leaf, int facenum, int *checkout)
{
size_t u;
q3cface_t *face;
q2mapsurface_t *surf;
q3cpatch_t *patch;
q3cmesh_t *cmesh;
face = &prv->faces[facenum];
if (face->numverts <= 0)
return true;
if (face->shadernum < 0 || face->shadernum >= loadmodel->numtextures)
return true;
surf = &prv->surfaces[face->shadernum];
if (!surf->c.value) //surface has no contents value, so can't ever block anything.
return true;
switch(face->facetype)
{
case MST_TRIANGLE_SOUP:
if (!face->soup.numindicies)
return true;
//only enable mesh collisions if its meant to be enabled.
//we haven't parsed any shaders, so we depend upon the stuff that the bsp compiler left lying around.
if (!(surf->c.flags & q3bsp_surf_meshcollision_flag.ival) && !q3bsp_surf_meshcollision_force.ival)
return true;
if (prv->numleafcmeshes >= prv->maxleafcmeshes)
{
prv->maxleafcmeshes *= 2;
prv->maxleafcmeshes += 16;
if (prv->numleafcmeshes > prv->maxleafcmeshes)
{ //detect overflow
Con_Printf (CON_ERROR "CM_CreateCMeshesForLeafs: map is insanely huge!\n");
return false;
}
prv->leafcmeshes = realloc(prv->leafcmeshes, sizeof(*prv->leafcmeshes) * prv->maxleafcmeshes);
}
// the patch was already built
if (checkout[facenum] != -1)
{
prv->leafcmeshes[prv->numleafcmeshes] = checkout[facenum];
cmesh = &prv->cmeshes[checkout[facenum]];
}
else
{
if (prv->numcmeshes >= MAX_CM_PATCHES)
{
Con_Printf (CON_ERROR "CM_CreatePatchesForLeafs: map has too many patches\n");
return false;
}
cmesh = &prv->cmeshes[prv->numcmeshes];
prv->leafcmeshes[prv->numleafcmeshes] = prv->numcmeshes;
checkout[facenum] = prv->numcmeshes++;
//gcc warns without this cast
cmesh->surface = surf;
cmesh->numverts = face->numverts;
cmesh->numincidies = face->soup.numindicies;
cmesh->xyz_array = ZG_Malloc(&loadmodel->memgroup, cmesh->numverts * sizeof(*cmesh->xyz_array) + cmesh->numincidies * sizeof(*cmesh->indicies));
cmesh->indicies = (index_t*)(cmesh->xyz_array + cmesh->numverts);
VectorCopy(prv->verts[face->firstvert+0], cmesh->xyz_array[0]);
VectorCopy(cmesh->xyz_array[0], cmesh->absmaxs);
VectorCopy(cmesh->xyz_array[0], cmesh->absmins);
for (u = 1; u < cmesh->numverts; u++)
{
VectorCopy(prv->verts[face->firstvert+u], cmesh->xyz_array[u]);
AddPointToBounds(cmesh->xyz_array[u], cmesh->absmins, cmesh->absmaxs);
}
for (u = 0; u < cmesh->numincidies; u++)
cmesh->indicies[u] = prv->surfindexes[face->soup.firstindex+u];
}
leaf->contents |= surf->c.value;
leaf->numleafcmeshes++;
prv->numleafcmeshes++;
break;
case MST_PATCH:
case MST_PATCH_FIXED:
if (face->patch.cp[0] <= 0 || face->patch.cp[1] <= 0)
return true;
if ( !surf->c.value || (surf->c.flags & Q3SURF_NONSOLID) )
return true;
if (prv->numleafpatches >= prv->maxleafpatches)
{
prv->maxleafpatches *= 2;
prv->maxleafpatches += 16;
if (prv->numleafpatches > prv->maxleafpatches)
{ //detect overflow
Con_Printf (CON_ERROR "CM_CreatePatchesForLeafs: map is insanely huge!\n");
return false;
}
prv->leafpatches = realloc(prv->leafpatches, sizeof(*prv->leafpatches) * prv->maxleafpatches);
}
// the patch was already built
if (checkout[facenum] != -1)
{
prv->leafpatches[prv->numleafpatches] = checkout[facenum];
patch = &prv->patches[checkout[facenum]];
}
else
{
if (prv->numpatches >= MAX_CM_PATCHES)
{
Con_Printf (CON_ERROR "CM_CreatePatchesForLeafs: map has too many patches\n");
return false;
}
patch = &prv->patches[prv->numpatches];
prv->leafpatches[prv->numleafpatches] = prv->numpatches;
checkout[facenum] = prv->numpatches++;
//gcc warns without this cast
CM_CreatePatch (loadmodel, patch, surf, (const vec_t *)(prv->verts + face->firstvert), face->patch.cp, (face->facetype==MST_PATCH_FIXED)?face->patch.fixedres:NULL );
}
leaf->contents |= patch->surface->c.value;
leaf->numleafpatches++;
prv->numleafpatches++;
break;
}
return true;
}
static qboolean CM_CreatePatchesForLeaf (model_t *loadmodel, cminfo_t *prv, mleaf_t *leaf, int *checkout)
{
int j, k;
leaf->numleafpatches = 0;
leaf->firstleafpatch = prv->numleafpatches;
leaf->numleafcmeshes = 0;
leaf->firstleafcmesh = prv->numleafcmeshes;
if (leaf->cluster == -1)
return true;
for (j=0 ; j<leaf->nummarksurfaces ; j++)
{
k = leaf->firstmarksurface[j] - loadmodel->surfaces;
if (k >= prv->numfaces)
{
Con_Printf (CON_ERROR "CM_CreatePatchesForLeafs: corrupt map\n");
break;
}
if (!CM_CreatePatchForFace (loadmodel, prv, leaf, k, checkout))
return false;
}
return true;
}
/*
=================
CM_CreatePatchesForLeafs
=================
*/
static qboolean CM_CreatePatchesForLeafs (model_t *loadmodel, cminfo_t *prv)
{
int i, k;
mleaf_t *leaf;
int *checkout = alloca(sizeof(int)*prv->numfaces);
if (map_noCurves.ival)
return true;
memset (checkout, -1, sizeof(int)*prv->numfaces);
for (i = prv->numcmodels; i-- > 0; )
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
prv->cmodels[i].firstpatch = prv->numpatches;
prv->cmodels[i].firstcmesh = prv->numcmeshes;
if (i == 0)
{ //worldmodel's leafs
for (k = 0, leaf = loadmodel->leafs; k < loadmodel->numleafs; k++, leaf++)
if (!CM_CreatePatchesForLeaf(loadmodel, prv, leaf, checkout))
return false;
}
else
{ //submodel uni-leaf thing.
leaf = prv->cmodels[i].headleaf;
if (leaf)
{
if (!CM_CreatePatchesForLeaf(loadmodel, prv, leaf, checkout))
return false;
for (k = 0; k < prv->cmodels[i].numsurfaces; k++)
CM_CreatePatchForFace(loadmodel, prv, leaf, prv->cmodels[i].firstsurface+k, checkout);
}
}
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
prv->cmodels[i].num_patches = prv->numpatches-prv->cmodels[i].firstpatch;
prv->cmodels[i].num_cmeshes = prv->numcmeshes-prv->cmodels[i].firstcmesh;
}
return true;
}
#endif
/*
===============================================================================
MAP LOADING
===============================================================================
*/
static void CMod_SetParent (mnode_t *node, mnode_t *parent)
{
node->parent = parent;
if (node->contents != -1)
return;
CMod_SetParent (node->children[0], node);
CMod_SetParent (node->children[1], node);
}
#ifdef Q2BSPS
/*
=================
CMod_LoadSubmodels
=================
*/
static qboolean CModQ2_LoadSubmodels (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)loadmodel->meshinfo;
q2dmodel_t *in;
cmodel_t *out;
int i, j, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no models\n");
return false;
}
if (count > SANITY_MAX_Q2MAP_MODELS)
{
Con_Printf (CON_ERROR "Map has too many models\n");
return false;
}
out = prv->cmodels = ZG_Malloc(&loadmodel->memgroup, count * sizeof(*prv->cmodels));
prv->numcmodels = count;
for (i=0 ; i<count ; i++, in++, out++)
{
for (j=0 ; j<3 ; j++)
{ // spread the mins / maxs by a pixel
out->mins[j] = LittleFloat (in->mins[j]) - 1;
out->maxs[j] = LittleFloat (in->maxs[j]) + 1;
out->origin[j] = LittleFloat (in->origin[j]);
}
out->headnode = loadmodel->nodes + LittleLong (in->headnode);
out->firstsurface = LittleLong (in->firstface);
out->numsurfaces = LittleLong (in->numfaces);
}
AddPointToBounds(prv->cmodels[0].mins, loadmodel->mins, loadmodel->maxs);
AddPointToBounds(prv->cmodels[0].maxs, loadmodel->mins, loadmodel->maxs);
return true;
}
/*
=================
CMod_LoadSurfaces
=================
*/
static qboolean CModQ2_LoadSurfaces (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q2texinfo_t *in;
q2mapsurface_t *out;
int i, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no surfaces\n");
return false;
}
// if (count > MAX_Q2MAP_TEXINFO)
// Host_Error ("Map has too many surfaces");
mod->numtexinfo = count;
out = prv->surfaces = ZG_Malloc(&mod->memgroup, count * sizeof(*prv->surfaces));
for ( i=0 ; i<count ; i++, in++, out++)
{
Q_strncpyz (out->c.name, in->texture, sizeof(out->c.name));
Q_strncpyz (out->rname, in->texture, sizeof(out->rname));
out->c.flags = LittleLong (in->flags);
out->c.value = LittleLong (in->value);
}
return true;
}
#ifdef HAVE_CLIENT
static texture_t *Mod_LoadWall(model_t *loadmodel, char *mapname, char *texname, char *shadername, unsigned int imageflags)
{
char name[MAX_QPATH];
q2miptex_t replacementwal;
texture_t *tex;
q2miptex_t *wal;
image_t *base;
Q_snprintfz (name, sizeof(name), "textures/%s.wal", texname);
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
wal = (void *)FS_LoadMallocFile (name, NULL);
if (!wal)
{
wal = &replacementwal;
memset(wal, 0, sizeof(*wal));
Q_strncpyz(wal->name, texname, sizeof(wal->name));
wal->width = 64;
wal->height = 64;
}
else
{
wal->width = LittleLong(wal->width);
wal->height = LittleLong(wal->height);
}
{
int i;
for (i = 0; i < MIPLEVELS; i++)
wal->offsets[i] = LittleLong(wal->offsets[i]);
}
wal->flags = LittleLong(wal->flags);
wal->contents = LittleLong(wal->contents);
wal->value = LittleLong(wal->value);
tex = ZG_Malloc(&loadmodel->memgroup, sizeof(texture_t));
tex->vwidth = tex->srcwidth = wal->width;
tex->vheight = tex->srcheight = wal->height;
if (!tex->vwidth || !tex->vheight || wal == &replacementwal)
{
imageflags |= IF_LOADNOW; //make sure the size is known BEFORE it returns.
if (wal->offsets[0])
base = R_LoadReplacementTexture(wal->name, "bmodels", imageflags, (qbyte *)wal+wal->offsets[0], wal->width, wal->height, TF_SOLID8);
else
base = R_LoadReplacementTexture(wal->name, "bmodels", imageflags, NULL, 0, 0, TF_INVALID);
}
else
base = NULL;
if (wal == &replacementwal)
{
if (base)
{
if (base->status == TEX_LOADED||base->status==TEX_LOADING)
{
tex->vwidth = base->width;
tex->vheight = base->height;
}
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
else
Con_Printf("Unable to load textures/%s.wal\n", wal->name);
}
}
else
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
{
qbyte *out;
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
unsigned int size =
(wal->width>>0)*(wal->height>>0) +
(wal->width>>1)*(wal->height>>1) +
(wal->width>>2)*(wal->height>>2) +
(wal->width>>3)*(wal->height>>3);
tex->srcdata = out = BZ_Malloc(size);
tex->srcfmt = TF_MIP4_8PAL24_T255;
tex->palette = q2_palette;
memcpy(out, (qbyte *)wal + wal->offsets[0], (wal->width>>0)*(wal->height>>0));
out += (wal->width>>0)*(wal->height>>0);
memcpy(out, (qbyte *)wal + wal->offsets[1], (wal->width>>1)*(wal->height>>1));
out += (wal->width>>1)*(wal->height>>1);
memcpy(out, (qbyte *)wal + wal->offsets[2], (wal->width>>2)*(wal->height>>2));
out += (wal->width>>2)*(wal->height>>2);
memcpy(out, (qbyte *)wal + wal->offsets[3], (wal->width>>3)*(wal->height>>3));
out += (wal->width>>3)*(wal->height>>3);
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
BZ_Free(wal);
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
}
return tex;
}
static qboolean CModQ2_LoadTexInfo (model_t *mod, qbyte *mod_base, lump_t *l, char *mapname) //yes I know these load from the same place
{
q2texinfo_t *in;
mtexinfo_t *out;
int i, j, count;
char *lwr;
char sname[MAX_QPATH];
int texcount;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf ("MOD_LoadBmodel: funny lump size in %s\n", mod->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
mod->textures = ZG_Malloc(&mod->memgroup, sizeof(texture_t *)*count);
texcount = 0;
mod->texinfo = out;
mod->numtexinfo = count;
if (in[0].nexttexinfo != -1)
{
for (i = 1; i < count && in[i].nexttexinfo == in[0].nexttexinfo; i++)
;
if (i == count)
{
Con_Printf("WARNING: invalid texture animations in \"%s\"\n", mod->name);
for (i = 0; i < count; i++)
in[i].nexttexinfo = -1;
}
}
for ( i=0 ; i<count ; i++, in++, out++)
{
out->flags = LittleLong (in->flags);
for (j=0 ; j<4 ; j++)
out->vecs[0][j] = LittleFloat (in->vecs[0][j]);
for (j=0 ; j<4 ; j++)
out->vecs[1][j] = LittleFloat (in->vecs[1][j]);
out->vecscale[0] = 1.0/Length (out->vecs[0]);
out->vecscale[1] = 1.0/Length (out->vecs[1]);
if (out->flags & TI_SKY)
Q_snprintfz(sname, sizeof(sname), "sky/%s", in->texture);
else
mvd: cl_autotrack_team cvar locks autotrack to a specific team. getting the team name to match can still be problematic when it contains non-ascii chars however. qw: fix recording mid-map. q2: add support for recording demos on q2 servers. q: fix setattachment not using the correct orientations. q2: now supports splitscreen, as well as increased model and sound limits. cl: fix crosshair not appearing in splitscreen. cl: splitscreen clients now get their own colour tints (like the bf command) snd: tweak audio to be a bit more usable in splitscreen by default. cl: added con_logcenterprint cvar, for shoving a copy of centerprints onto the console. by default only appears in single player. qc: add checkbuiltin builtin. for all those #0 builtins without their own extension name. gl: fix r_dynamic -1 bug that was painfully visible in AD. mdl: validate per-frame bounds of mdl files (stuff that would crash software renderers). sv: fix -port or +sv_port commandline args to override the port correctly. win: attempt to cope with windows symlinks enumerating with the wrong filesizes. gl: fix skyboxes not appearing properly. gl: fix sprite alpha/edge issue resulting in some invisible sprites in AD. gl: fix screenshot_mega, in combination with r_projection. yay for HUGE panoramic screenshots. q2: fix replacement textures issue. qw: fix download demonum/X issue, both in client and server. qw: fix multicast dimensions not always being honoured properly. nq: fix starting angles sometimes being wrong. menusys: I'm finally uploading my menusys library, plus example mod, which I'll now be providing like csaddon is. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4997 fc73d0e0-1445-4013-8a0c-d673dee63da5
2016-01-18 05:22:07 +00:00
Q_snprintfz(sname, sizeof(sname), "%s", in->texture);
if (out->flags & (TI_WARP))
mvd: cl_autotrack_team cvar locks autotrack to a specific team. getting the team name to match can still be problematic when it contains non-ascii chars however. qw: fix recording mid-map. q2: add support for recording demos on q2 servers. q: fix setattachment not using the correct orientations. q2: now supports splitscreen, as well as increased model and sound limits. cl: fix crosshair not appearing in splitscreen. cl: splitscreen clients now get their own colour tints (like the bf command) snd: tweak audio to be a bit more usable in splitscreen by default. cl: added con_logcenterprint cvar, for shoving a copy of centerprints onto the console. by default only appears in single player. qc: add checkbuiltin builtin. for all those #0 builtins without their own extension name. gl: fix r_dynamic -1 bug that was painfully visible in AD. mdl: validate per-frame bounds of mdl files (stuff that would crash software renderers). sv: fix -port or +sv_port commandline args to override the port correctly. win: attempt to cope with windows symlinks enumerating with the wrong filesizes. gl: fix skyboxes not appearing properly. gl: fix sprite alpha/edge issue resulting in some invisible sprites in AD. gl: fix screenshot_mega, in combination with r_projection. yay for HUGE panoramic screenshots. q2: fix replacement textures issue. qw: fix download demonum/X issue, both in client and server. qw: fix multicast dimensions not always being honoured properly. nq: fix starting angles sometimes being wrong. menusys: I'm finally uploading my menusys library, plus example mod, which I'll now be providing like csaddon is. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4997 fc73d0e0-1445-4013-8a0c-d673dee63da5
2016-01-18 05:22:07 +00:00
Q_strncatz(sname, "#WARP", sizeof(sname));
if (out->flags & TI_FLOWING)
Q_strncatz(sname, "#FLOW", sizeof(sname));
if (out->flags & TI_TRANS66)
Q_strncatz(sname, "#ALPHA=0.66", sizeof(sname));
else if (out->flags & TI_TRANS33)
Q_strncatz(sname, "#ALPHA=0.33", sizeof(sname));
else if (out->flags & TI_KINGPIN_ALPHATEST) //kingpin...
Q_strncatz(sname, "#MASK=0.666#MASKLT", sizeof(sname));
else if (out->flags & (TI_WARP))
Q_strncatz(sname, "#ALPHA=1", sizeof(sname));
if (in->nexttexinfo != -1) //used to ensure non-looping and looping don't conflict and get confused.
Q_strncatz(sname, "#ANIMLOOP", sizeof(sname));
//in q2, 'TEX_SPECIAL' is TI_LIGHT, and that conflicts.
out->flags &= ~TI_LIGHT;
if (out->flags & (TI_SKY|TI_TRANS33|TI_TRANS66|TI_WARP))
out->flags |= TEX_SPECIAL;
//compact the textures.
for (j=0; j < texcount; j++)
{
if (!strcmp(sname, mod->textures[j]->name))
{
out->texture = mod->textures[j];
break;
}
}
if (j == texcount) //load a new one
{
for (lwr = in->texture; *lwr; lwr++)
{
if (*lwr >= 'A' && *lwr <= 'Z')
*lwr = *lwr - 'A' + 'a';
}
out->texture = Mod_LoadWall (mod, mapname, in->texture, sname, (out->flags&TEX_SPECIAL)?0:IF_NOALPHA);
if (!out->texture || !out->texture->srcwidth || !out->texture->srcheight)
{
out->texture = ZG_Malloc(&mod->memgroup, sizeof(texture_t) + 16*16+8*8+4*4+2*2);
Con_Printf (CON_WARNING "Couldn't load \"%s.wal\"\n", in->texture);
memcpy(out->texture, r_notexture_mip, sizeof(texture_t) + 16*16+8*8+4*4+2*2);
}
Q_strncpyz(out->texture->name, sname, sizeof(out->texture->name));
mod->textures[texcount++] = out->texture;
}
// if (in->nexttexinfo != -1)
// {
// Con_DPrintf("FIXME: %s should animate to %s\n", in->texture, (in->nexttexinfo+(q2texinfo_t *)(mod_base + l->fileofs))->texture);
// }
}
in = (void *)(mod_base + l->fileofs);
out = mod->texinfo;
for (i=0 ; i<count ; i++)
{
if (in[i].nexttexinfo >= 0 && in[i].nexttexinfo < count)
out[i].texture->anim_next = out[in[i].nexttexinfo].texture;
}
for (i=0 ; i<count ; i++)
{
texture_t *tex;
if (!out[i].texture->anim_next)
continue;
out[i].texture->anim_total = 1;
for (tex = out[i].texture->anim_next ; tex && tex != out[i].texture && out[i].texture->anim_total < 100; tex=tex->anim_next)
out[i].texture->anim_total++;
}
mod->numtextures = texcount;
Mod_SortShaders(mod);
return true;
}
#endif
/*
static void CalcSurfaceExtents (msurface_t *s)
{
float mins[2], maxs[2], val;
int i,j, e;
mvertex_t *v;
mtexinfo_t *tex;
int bmins[2], bmaxs[2];
mins[0] = mins[1] = 999999;
maxs[0] = maxs[1] = -99999;
tex = s->texinfo;
for (i=0 ; i<s->numedges ; i++)
{
e = loadmodel->surfedges[s->firstedge+i];
if (e >= 0)
v = &loadmodel->vertexes[loadmodel->edges[e].v[0]];
else
v = &loadmodel->vertexes[loadmodel->edges[-e].v[1]];
for (j=0 ; j<2 ; j++)
{
val = v->position[0] * tex->vecs[j][0] +
v->position[1] * tex->vecs[j][1] +
v->position[2] * tex->vecs[j][2] +
tex->vecs[j][3];
if (val < mins[j])
mins[j] = val;
if (val > maxs[j])
maxs[j] = val;
}
}
for (i=0 ; i<2 ; i++)
{
bmins[i] = floor(mins[i]/16);
bmaxs[i] = ceil(maxs[i]/16);
s->texturemins[i] = bmins[i] * 16;
s->extents[i] = (bmaxs[i] - bmins[i]) * 16;
// if ( !(tex->flags & TEX_SPECIAL) && s->extents[i] > 512 )// 256 )
// Sys_Error ("Bad surface extents");
}
}*/
/*
=================
Mod_LoadFaces
=================
*/
#ifdef HAVE_CLIENT
static qboolean CModQ2_LoadFaces (model_t *mod, qbyte *mod_base, lump_t *l, lump_t *lightlump, qboolean lightofsisdouble, bspx_header_t *bspx)
{
dsface_t *in;
msurface_t *out;
int i, count, surfnum;
int planenum, side;
int ti;
int style;
struct decoupled_lm_info_s *decoupledlm;
unsigned int dcsize, lofs;
unsigned short lmshift, lmscale;
char buf[64];
lightmapoverrides_t overrides = {0};
overrides.defaultshift = LMSHIFT_DEFAULT;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf ("MOD_LoadBmodel: funny lump size in %s\n",mod->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&mod->memgroup, (count+6)*sizeof(*out)); //spare for skybox
mod->surfaces = out;
mod->numsurfaces = count;
Mod_LoadLighting(mod, bspx, mod_base, lightlump, lightofsisdouble, &overrides, sb_none);
if (overrides.offsets)
lmshift = overrides.defaultshift;
else
{
lmscale = atoi(Mod_ParseWorldspawnKey(mod, "lightmap_scale", buf, sizeof(buf)));
if (!lmscale)
lmshift = LMSHIFT_DEFAULT;
else
{
for(lmshift = 0; lmscale > 1; lmshift++)
lmscale >>= 1;
}
}
decoupledlm = BSPX_FindLump(bspx, mod_base, "DECOUPLED_LM", &dcsize); //RGB packed data
if (dcsize == count*sizeof(*decoupledlm))
mod->facelmvecs = ZG_Malloc(&mod->memgroup, count * sizeof(*mod->facelmvecs)); //seems good.
else
decoupledlm = NULL; //wrong size somehow... discard it.
for ( surfnum=0 ; surfnum<count ; surfnum++, in++, out++)
{
out->firstedge = LittleLong(in->firstedge);
out->numedges = (unsigned short)LittleShort(in->numedges);
out->flags = 0;
planenum = (unsigned short)LittleShort(in->planenum);
side = (unsigned short)LittleShort(in->side);
if (side)
out->flags |= SURF_PLANEBACK;
out->plane = mod->planes + planenum;
ti = (unsigned short)LittleShort (in->texinfo);
if (ti < 0 || ti >= mod->numtexinfo)
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: bad texinfo number\n");
return false;
}
out->texinfo = mod->texinfo + ti;
#ifndef SERVERONLY
if (out->texinfo->flags & TI_SKY)
{
out->flags |= SURF_DRAWSKY;
}
if (out->texinfo->flags & TI_WARP)
{
out->flags |= SURF_DRAWTURB|SURF_DRAWTILED;
}
#endif
if (overrides.shifts)
out->lmshift = overrides.shifts[surfnum];
else
out->lmshift = lmshift;
if (decoupledlm)
{
lofs = LittleLong(decoupledlm->lmoffset);
out->texturemins[0] = out->texturemins[1] = 0; // should be handled by the now-per-surface vecs[][3] value.
out->lmshift = 0; //redundant.
out->extents[0] = (unsigned short)LittleShort(decoupledlm->lmsize[0]) - 1;
out->extents[1] = (unsigned short)LittleShort(decoupledlm->lmsize[1]) - 1;
mod->facelmvecs[surfnum].lmvecs[0][0] = LittleFloat(decoupledlm->lmvecs[0][0]);
mod->facelmvecs[surfnum].lmvecs[0][1] = LittleFloat(decoupledlm->lmvecs[0][1]);
mod->facelmvecs[surfnum].lmvecs[0][2] = LittleFloat(decoupledlm->lmvecs[0][2]);
mod->facelmvecs[surfnum].lmvecs[0][3] = LittleFloat(decoupledlm->lmvecs[0][3]) + 0.5f; //sigh
mod->facelmvecs[surfnum].lmvecs[1][0] = LittleFloat(decoupledlm->lmvecs[1][0]);
mod->facelmvecs[surfnum].lmvecs[1][1] = LittleFloat(decoupledlm->lmvecs[1][1]);
mod->facelmvecs[surfnum].lmvecs[1][2] = LittleFloat(decoupledlm->lmvecs[1][2]);
mod->facelmvecs[surfnum].lmvecs[1][3] = LittleFloat(decoupledlm->lmvecs[1][3]) + 0.5f; //sigh
mod->facelmvecs[surfnum].lmvecscale[0] = 1.0f/Length(mod->facelmvecs[surfnum].lmvecs[0]); //luxels->qu
mod->facelmvecs[surfnum].lmvecscale[1] = 1.0f/Length(mod->facelmvecs[surfnum].lmvecs[1]);
decoupledlm++;
}
else
{
if (overrides.offsets)
lofs = overrides.offsets[surfnum];
else
lofs = LittleLong(in->lightofs);
CalcSurfaceExtents (mod, out);
if (overrides.extents)
{
out->extents[0] = overrides.extents[surfnum*2+0];
out->extents[1] = overrides.extents[surfnum*2+1];
}
}
// lighting info
if (overrides.styles16)
{
for (i=0 ; i<overrides.stylesperface ; i++)
{
style = overrides.styles16[surfnum*overrides.stylesperface+i];
if (style == 0xffff)
style = INVALID_LIGHTSTYLE;
else if (mod->lightmaps.maxstyle < style)
mod->lightmaps.maxstyle = style;
out->styles[i] = style;
}
}
else if (overrides.styles8)
{
for (i=0 ; i<overrides.stylesperface ; i++)
{
style = overrides.styles8[surfnum*overrides.stylesperface+i];
if (style == 0xff)
style = INVALID_LIGHTSTYLE;
else if (mod->lightmaps.maxstyle < style)
mod->lightmaps.maxstyle = style;
out->styles[i] = style;
}
}
else
{
for (i=0 ; i<4 ; i++)
{
style = in->styles[i];
if (style == 0xff)
style = INVALID_LIGHTSTYLE;
else if (mod->lightmaps.maxstyle < style)
mod->lightmaps.maxstyle = style;
out->styles[i] = style;
}
}
for ( ; i<MAXCPULIGHTMAPS ; i++)
out->styles[i] = INVALID_LIGHTSTYLE;
if (lofs == ~0u)
out->samples = NULL;
else if (lightofsisdouble)
out->samples = mod->lightdata + (lofs/2);
else
out->samples = mod->lightdata + lofs;
// set the drawing flags
if (out->texinfo->flags & TI_WARP)
{
out->flags |= SURF_DRAWTURB;
for (i=0 ; i<2 ; i++)
{
out->extents[i] = 16384;
out->texturemins[i] = -8192;
}
}
}
return true;
}
#endif
/*
=================
CMod_LoadNodes
=================
*/
static qboolean CModQ2_LoadNodes (model_t *mod, qbyte *mod_base, lump_t *l)
{
q2dnode_t *in;
int child;
mnode_t *out;
int i, j, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map has no nodes\n");
return false;
}
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Map has too many nodes\n");
return false;
}
out = ZG_Malloc(&mod->memgroup, sizeof(mnode_t)*count);
mod->nodes = out;
mod->numnodes = count;
for (i=0 ; i<count ; i++, out++, in++)
{
memset(out, 0, sizeof(*out));
for (j=0 ; j<3 ; j++)
{
out->minmaxs[j] = LittleShort (in->mins[j]);
out->minmaxs[3+j] = LittleShort (in->maxs[j]);
}
out->plane = mod->planes + LittleLong(in->planenum);
out->firstsurface = (unsigned short)LittleShort (in->firstface);
out->numsurfaces = (unsigned short)LittleShort (in->numfaces);
out->contents = -1; // differentiate from leafs
for (j=0 ; j<2 ; j++)
{
child = LittleLong (in->children[j]);
out->childnum[j] = child;
if (child < 0)
out->children[j] = (mnode_t *)(mod->leafs + -1-child);
else
out->children[j] = mod->nodes + child;
}
}
CMod_SetParent (mod->nodes, NULL); // sets nodes and leafs
return true;
}
/*
=================
CMod_LoadBrushes
=================
*/
static qboolean CModQ2_LoadBrushes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q2dbrush_t *in;
q2cbrush_t *out;
int i, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > SANITY_MAX_MAP_BRUSHES)
{
Con_Printf (CON_ERROR "Map has too many brushes");
return false;
}
prv->brushes = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+1));
out = prv->brushes;
prv->numbrushes = count;
for (i=0 ; i<count ; i++, out++, in++)
{
//FIXME: missing bounds checks
out->brushside = &prv->brushsides[LittleLong(in->firstside)];
out->numsides = LittleLong(in->numsides);
out->contents = LittleLong(in->contents);
CM_FinalizeBrush(out);
}
return true;
}
/*
=================
CMod_LoadLeafs
=================
*/
static qboolean CModQ2_LoadLeafs (model_t *mod, qbyte *mod_base, lump_t *l)
{
int i, j;
mleaf_t *out;
q2dleaf_t *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no leafs\n");
return false;
}
// need to save space for box planes
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Map has too many leafs\n");
return false;
}
out = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+1));
mod->numclusters = 0;
mod->leafs = out;
mod->numleafs = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
memset(out, 0, sizeof(*out));
for (j=0 ; j<3 ; j++)
{
out->minmaxs[j] = LittleShort (in->mins[j]);
out->minmaxs[3+j] = LittleShort (in->maxs[j]);
}
out->contents = LittleLong (in->contents);
out->cluster = (unsigned short)LittleShort (in->cluster);
if (out->cluster == 0xffff)
out->cluster = -1;
out->area = (unsigned short)LittleShort (in->area);
out->firstleafbrush = (unsigned short)LittleShort (in->firstleafbrush);
out->numleafbrushes = (unsigned short)LittleShort (in->numleafbrushes);
out->firstmarksurface = mod->marksurfaces +
(unsigned short)LittleShort(in->firstleafface);
out->nummarksurfaces = (unsigned short)LittleShort(in->numleaffaces);
if (out->cluster >= mod->numclusters)
mod->numclusters = out->cluster + 1;
}
out = mod->leafs;
mod->pvsbytes = ((mod->numclusters + 31)>>3)&~3;
if (out[0].contents != Q2CONTENTS_SOLID)
{
Con_Printf (CON_ERROR "Map leaf 0 is not CONTENTS_SOLID\n");
return false;
}
return true;
}
/*
=================
CMod_LoadPlanes
=================
*/
static qboolean CModQ2_LoadPlanes (model_t *mod, qbyte *mod_base, lump_t *l)
{
int i, j;
mplane_t *out;
dplane_t *in;
int count;
int bits;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no planes\n");
return false;
}
// need to save space for box planes
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Map has too many planes (%i)\n", count);
return false;
}
mod->planes = out = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);
mod->numplanes = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
bits = 0;
for (j=0 ; j<3 ; j++)
{
out->normal[j] = LittleFloat (in->normal[j]);
if (out->normal[j] < 0)
bits |= 1<<j;
}
out->dist = LittleFloat (in->dist);
out->type = LittleLong (in->type);
out->signbits = bits;
}
return true;
}
/*
=================
CMod_LoadLeafBrushes
=================
*/
static qboolean CModQ2_LoadLeafBrushes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
q2cbrush_t **out;
unsigned short *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no planes\n");
return false;
}
// need to save space for box planes
if (count > SANITY_MAX_MAP_LEAFBRUSHES)
{
Con_Printf (CON_ERROR "Map has too many leafbrushes\n");
return false;
}
//prv->numbrushes is because of submodels being weird.
out = prv->leafbrushes = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+prv->numbrushes));
prv->numleafbrushes = count;
for ( i=0 ; i<count ; i++, in++, out++)
*out = prv->brushes + (unsigned short)(short)LittleShort (*in);
return true;
}
/*
=================
CMod_LoadBrushSides
=================
*/
static qboolean CModQ2_LoadBrushSides (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
unsigned int i, j;
q2cbrushside_t *out;
q2dbrushside_t *in;
int count;
int num;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
// need to save space for box planes
if (count > SANITY_MAX_MAP_BRUSHSIDES)
{
Con_Printf (CON_ERROR "Map has too many brushsides (%i)\n", count);
return false;
}
out = prv->brushsides = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);
prv->numbrushsides = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
num = (unsigned short)LittleShort (in->planenum);
out->plane = &mod->planes[num];
j = (unsigned short)LittleShort (in->texinfo);
if (j >= mod->numtexinfo)
out->surface = &nullsurface;
else
out->surface = &prv->surfaces[j];
}
return true;
}
/*
=================
CMod_LoadAreas
=================
*/
static qboolean CModQ2_LoadAreas (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
q2carea_t *out;
q2darea_t *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > MAX_Q2MAP_AREAS)
{
Con_Printf (CON_ERROR "Map has too many areas\n");
return false;
}
out = prv->q2areas = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);;
prv->numareas = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
out->numareaportals = LittleLong (in->numareaportals);
out->firstareaportal = LittleLong (in->firstareaportal);
}
return true;
}
/*
=================
CMod_LoadAreaPortals
=================
*/
static qboolean CModQ2_LoadAreaPortals (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
q2dareaportal_t *out;
q2dareaportal_t *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > MAX_Q2MAP_AREAS)
{
Con_Printf (CON_ERROR "Map has too many areas\n");
return false;
}
out = prv->q2areaportals = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);
prv->numq2areaportals = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
out->portalnum = LittleLong (in->portalnum);
out->otherarea = LittleLong (in->otherarea);
}
return true;
}
/*
=================
CMod_LoadVisibility
=================
*/
static qboolean CModQ2_LoadVisibility (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
prv->numvisibility = l->filelen;
// if (l->filelen > MAX_Q2MAP_VISIBILITY)
// {
// Con_Printf (CON_ERROR "Map has too large visibility lump\n");
// return false;
// }
prv->q2vis = ZG_Malloc(&mod->memgroup, l->filelen);
memcpy (prv->q2vis, mod_base + l->fileofs, l->filelen);
mod->vis = prv->q2vis;
prv->q2vis->numclusters = LittleLong (prv->q2vis->numclusters);
for (i=0 ; i<prv->q2vis->numclusters ; i++)
{
prv->q2vis->bitofs[i][0] = LittleLong (prv->q2vis->bitofs[i][0]);
prv->q2vis->bitofs[i][1] = LittleLong (prv->q2vis->bitofs[i][1]);
}
mod->numclusters = prv->q2vis->numclusters;
mod->pvsbytes = ((mod->numclusters + 31)>>3)&~3;
return true;
}
#endif //q2bsps
#ifdef Q3BSPS
static qboolean CModQ3_LoadMarksurfaces (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
int i, j, count;
int *in;
msurface_t **out;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "CModQ3_LoadMarksurfaces: funny lump size in %s\n",loadmodel->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&loadmodel->memgroup, count*sizeof(*out));
loadmodel->marksurfaces = out;
loadmodel->nummarksurfaces = count;
for ( i=0 ; i<count ; i++)
{
j = LittleLong(in[i]);
if (j < 0 || j >= loadmodel->numsurfaces)
{
Con_Printf (CON_ERROR "Mod_ParseMarksurfaces: bad surface number\n");
return false;
}
out[i] = loadmodel->surfaces + j;
}
return true;
}
static qboolean CModQ3_LoadSubmodels (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q3dmodel_t *in;
cmodel_t *out;
int i, j, count;
q2cbrush_t **leafbrush;
mleaf_t *bleaf;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no models\n");
return false;
}
if (count > SANITY_MAX_Q2MAP_MODELS)
{
Con_Printf (CON_ERROR "Map has too many models\n");
return false;
}
out = prv->cmodels = ZG_Malloc(&mod->memgroup, count * sizeof(*prv->cmodels));
prv->numcmodels = count;
if (count > 1)
bleaf = ZG_Malloc(&mod->memgroup, (count-1) * sizeof(*bleaf));
else
bleaf = NULL;
prv->mapisq3 = true;
for (i=0 ; i<count ; i++, in++, out++)
{
for (j=0 ; j<3 ; j++)
{ // spread the mins / maxs by a pixel
out->mins[j] = LittleFloat (in->mins[j]) - 1;
out->maxs[j] = LittleFloat (in->maxs[j]) + 1;
out->origin[j] = (out->maxs[j] + out->mins[j])/2;
}
out->firstsurface = LittleLong (in->firstsurface);
out->numsurfaces = LittleLong (in->num_surfaces);
makefile: attempt to fix freetype when not using makelibs (should make it slightly easier for people to compile with msys2 without needing to resort to cmake). emenu: clean up hexen2's maplist options slightly. emenu: modelviewer should now be slightly more friendly (click+wasd to move around). particles: fix up randomised s coords. csqc: try to fix issue with applycustomskin not refcounting properly. client: [s_]precache and (new) mod_precache cvars can be set to 2 to precache the resources after load, for faster loading at the expense of some early stutter, without risking later mid-game stuttering. gltf: add support for morphweights in a cpu-fallback path. don't expect good performance on surfaces with morphtargets for now. gtlf: add some support for gltf1 files. far from perfect. shaders: gltf1 semantics handling shaders: const correctness iqmtool: fix up mdl skin export. iqmtool: integrate the engine's gltf2 loader. works with animated models, but unanimated ones suffer from basepose-different-from-bindpose issues. q3bsp: hopefully fixed bih traces. still disabled for now. qc: change default value of pr_gc_threaded to 1. qcext: add the '__deprecated' keyword to various symbols in fteextensions.qc, now that fteqcc supports it. ssqc: spit out a more readable error for WriteByte(MSG_CSQC,...) outside of SendEntity. ssqc: add registercommand builtin, for consistency with menuqc and csqc (though only one can register any single command). sv: report userinfo/serverinfo sizes (some clients still have arbitrary limits, plus its nice to see how abusive things are) sv: try to optimise sv_cullentities_trace a little. movechain: relink moved ents. csqc: add spriteframe builtin, for freecs to use instead of more ugly less reliable hacks. menuqc: fopen("tls://host:port", FILE_STREAM) should now open a tls stream. tcp:// should also work. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5703 fc73d0e0-1445-4013-8a0c-d673dee63da5
2020-06-12 23:29:58 +00:00
out->firstbrush = LittleLong(in->firstbrush);
out->num_brushes = LittleLong(in->num_brushes);
if (!i)
{
out->headnode = mod->nodes;
out->headleaf = NULL;
}
else
{
//create a new leaf to hold the brushes and be directly clipped
out->headleaf = bleaf;
out->headnode = NULL;
bleaf->numleafbrushes = LittleLong ( in->num_brushes );
bleaf->firstleafbrush = prv->numleafbrushes;
bleaf->contents = 0;
leafbrush = &prv->leafbrushes[prv->numleafbrushes];
for ( j = 0; j < bleaf->numleafbrushes; j++, leafbrush++ )
{
*leafbrush = prv->brushes + LittleLong ( in->firstbrush ) + j;
bleaf->contents |= (*leafbrush)->contents;
}
prv->numleafbrushes += bleaf->numleafbrushes;
bleaf++;
}
//submodels
}
AddPointToBounds(prv->cmodels[0].mins, mod->mins, mod->maxs);
AddPointToBounds(prv->cmodels[0].maxs, mod->mins, mod->maxs);
return true;
}
static qboolean CModQ3_LoadShaders (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
dq3shader_t *in;
q2mapsurface_t *out;
int i, count;
texture_t *tex;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no shaders\n");
return false;
}
// else if (count > MAX_Q2MAP_TEXINFO)
// Host_Error ("Map has too many shaders");
mod->numtexinfo = count;
out = prv->surfaces = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
mod->textures = ZG_Malloc(&mod->memgroup, (sizeof(texture_t*)+sizeof(mtexinfo_t)+sizeof(texture_t))*(count*2+1)); //+1 is 'noshader' for flares.
mod->texinfo = (mtexinfo_t*)(mod->textures+(count*2+1));
tex = (texture_t*)(mod->texinfo+(count*2+1));
mod->numtextures = count*2+1;
for ( i=0 ; i<count ; i++, in++, out++ )
{
out->c.flags = LittleLong ( in->surfflags );
out->c.value = LittleLong ( in->contents );
mod->texinfo[i].texture = tex+i;
mod->texinfo[i].flags = prv->surfaces[i].c.flags;
Q_strncpyz(mod->texinfo[i].texture->name, in->shadername, sizeof(mod->texinfo[i].texture->name));
mod->textures[i] = mod->texinfo[i].texture;
}
for ( i=0, in-=count ; i<count ; i++, in++ )
{
mod->texinfo[i+count].texture = tex+i+count;
mod->texinfo[i+count].flags = prv->surfaces[i].c.flags;
Q_strncpyz(mod->texinfo[i+count].texture->name, in->shadername, sizeof(mod->texinfo[i+count].texture->name));
mod->textures[i+count] = mod->texinfo[i+count].texture;
}
//and for flares, which are not supported at this time.
mod->texinfo[count*2].texture = tex+count*2;
mod->texinfo[i+count].flags = 0;
Q_strncpyz(mod->texinfo[count*2].texture->name, "noshader", sizeof(mod->texinfo[count*2].texture->name));
mod->textures[count*2] = mod->texinfo[count*2].texture;
return true;
}
static qboolean CModQ3_LoadVertexes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q3dvertex_t *in;
vecV_t *out;
vec3_t *nout;
//, *sout, *tout;
int i, count, j;
vec2_t *lmout, *stout;
vec4_t *cout;
extern cvar_t gl_overbright;
extern qbyte lmgamma[256];
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "CMOD_LoadVertexes: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > MAX_Q3MAP_VERTEXES)
{
Con_Printf (CON_ERROR "Map has too many vertexes\n");
return false;
}
BuildLightMapGammaTable(1, 1<<(2-gl_overbright.ival));
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
stout = ZG_Malloc(&mod->memgroup, count*sizeof(*stout));
lmout = ZG_Malloc(&mod->memgroup, count*sizeof(*lmout));
cout = ZG_Malloc(&mod->memgroup, count*sizeof(*cout));
nout = ZG_Malloc(&mod->memgroup, count*sizeof(*nout));
// sout = ZG_Malloc(&mod->memgroup, count*sizeof(*nout));
// tout = ZG_Malloc(&mod->memgroup, count*sizeof(*nout));
prv->verts = out;
prv->vertstmexcoords = stout;
for (i = 0; i < MAXRLIGHTMAPS; i++)
{
prv->vertlstmexcoords[i] = lmout;
prv->colors4f_array[i] = cout;
}
prv->normals_array = nout;
// prv->svector_array = sout;
// prv->tvector_array = tout;
prv->numvertexes = count;
for ( i=0 ; i<count ; i++, in++)
{
for ( j=0 ; j < 3 ; j++)
{
out[i][j] = LittleFloat ( in->point[j] );
nout[i][j] = LittleFloat (in->normal[j]);
}
for ( j=0 ; j < 2 ; j++)
{
stout[i][j] = LittleFloat ( ((float *)in->texcoords)[j] );
lmout[i][j] = LittleFloat ( ((float *)in->texcoords)[j+2] );
}
cout[i][0] = (lmgamma[in->color[0]]<<gl_overbright.ival)/255.0f;
cout[i][1] = (lmgamma[in->color[1]]<<gl_overbright.ival)/255.0f;
cout[i][2] = (lmgamma[in->color[2]]<<gl_overbright.ival)/255.0f;
cout[i][3] = in->color[3]/255.0f;
}
// if (r_lightmap_saturation.value != 1.0f)
// SaturateR8G8B8(cout, count*4, r_lightmap_saturation.value);
return true;
}
#ifdef RFBSPS
static qboolean CModRBSP_LoadVertexes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
rbspvertex_t *in;
vecV_t *out;
vec3_t *nout;
//, *sout, *tout;
int i, count, j;
vec2_t *lmout, *stout;
vec4_t *cout;
int sty;
extern qbyte lmgamma[256];
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "CMOD_LoadVertexes: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > MAX_Q3MAP_VERTEXES)
{
Con_Printf (CON_ERROR "Map has too many vertexes\n");
return false;
}
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
stout = ZG_Malloc(&mod->memgroup, count*sizeof(*stout));
lmout = ZG_Malloc(&mod->memgroup, MAXRLIGHTMAPS*count*sizeof(*lmout));
cout = ZG_Malloc(&mod->memgroup, MAXRLIGHTMAPS*count*sizeof(*cout));
nout = ZG_Malloc(&mod->memgroup, count*sizeof(*nout));
// sout = ZG_Malloc(&mod->memgroup, count*sizeof(*sout));
// tout = ZG_Malloc(&mod->memgroup, count*sizeof(*tout));
prv->verts = out;
prv->vertstmexcoords = stout;
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
prv->vertlstmexcoords[sty] = lmout + sty*count;
prv->colors4f_array[sty] = cout + sty*count;
}
prv->normals_array = nout;
// prv->svector_array = sout;
// prv->tvector_array = tout;
prv->numvertexes = count;
for ( i=0 ; i<count ; i++, in++)
{
for ( j=0 ; j < 3 ; j++)
{
out[i][j] = LittleFloat ( in->point[j] );
nout[i][j] = LittleFloat (in->normal[j]);
}
for ( j=0 ; j < 2 ; j++)
{
stout[i][j] = LittleFloat (in->stcoords[j]);
for (sty = 0; sty < min(MAXRLIGHTMAPS, RBSP_STYLESPERSURF); sty++)
prv->vertlstmexcoords[sty][i][j] = LittleFloat(in->lmtexcoords[sty][j]);
}
for (sty = 0; sty < min(MAXRLIGHTMAPS, RBSP_STYLESPERSURF); sty++)
{
prv->colors4f_array[sty][i][0] = lmgamma[in->color[sty][0]]/255.0f;
prv->colors4f_array[sty][i][1] = lmgamma[in->color[sty][1]]/255.0f;
prv->colors4f_array[sty][i][2] = lmgamma[in->color[sty][2]]/255.0f;
prv->colors4f_array[sty][i][3] = in->color[sty][3]/255.0f;
}
}
return true;
}
#endif
#ifndef SERVERONLY
static qboolean CModQ3_LoadIndexes (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)loadmodel->meshinfo;
int i, count;
int *in;
index_t *out;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n", loadmodel->name);
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1 || count >= MAX_Q3MAP_INDICES)
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: too many indicies in %s: %i\n",
loadmodel->name, count);
return false;
}
out = ZG_Malloc(&loadmodel->memgroup, count*sizeof(*out));
prv->surfindexes = out;
// prv->numsurfindexes = count;
for ( i=0 ; i<count ; i++)
out[i] = LittleLong (in[i]);
return true;
}
#endif
/*
=================
CMod_LoadFaces
=================
*/
static qboolean CModQ3_LoadFaces (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q3dface_t *in;
q3cface_t *out;
int i, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Map has too many faces\n");
return false;
}
out = BZ_Malloc ( count*sizeof(*out) );
prv->faces = out;
prv->numfaces = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
out->facetype = LittleLong ( in->facetype );
out->shadernum = LittleLong ( in->shadernum );
out->numverts = LittleLong ( in->num_vertices );
out->firstvert = LittleLong ( in->firstvertex );
if (out->facetype == MST_PATCH || out->facetype == MST_PATCH_FIXED)
{
unsigned int pw = LittleLong ( in->patchwidth );
unsigned int ph = LittleLong ( in->patchheight );
out->patch.cp[0] = pw&0xffff;
out->patch.cp[1] = ph&0xffff;
out->patch.fixedres[0] = pw>>16;
out->patch.fixedres[1] = ph>>16;
}
else
{
out->soup.firstindex = LittleLong(in->firstindex);
out->soup.numindicies = LittleLong(in->num_indexes);
}
}
mod->numsurfaces = i;
return true;
}
#ifdef RFBSPS
static qboolean CModRBSP_LoadFaces (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
rbspface_t *in;
q3cface_t *out;
int i, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Map has too many faces\n");
return false;
}
out = BZ_Malloc ( count*sizeof(*out) );
prv->faces = out;
prv->numfaces = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
out->facetype = LittleLong ( in->facetype );
out->shadernum = LittleLong ( in->shadernum );
out->numverts = LittleLong ( in->num_vertices );
out->firstvert = LittleLong ( in->firstvertex );
if (out->facetype == MST_PATCH || out->facetype == MST_PATCH_FIXED)
{
unsigned int pw = LittleLong ( in->patchwidth );
unsigned int ph = LittleLong ( in->patchheight );
out->patch.cp[0] = pw&0xffff;
out->patch.cp[1] = ph&0xffff;
out->patch.fixedres[0] = pw>>16;
out->patch.fixedres[1] = ph>>16;
}
else
{
out->soup.firstindex = LittleLong(in->firstindex);
out->soup.numindicies = LittleLong(in->num_indexes);
}
}
mod->numsurfaces = i;
return true;
}
#endif
#ifndef SERVERONLY
/*
=================
Mod_LoadFogs
=================
*/
static qboolean CModQ3_LoadFogs (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
dfog_t *in;
mfog_t *out;
q2cbrush_t *brush;
q2cbrushside_t *visibleside, *brushsides;
int i, j, count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n", mod->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
mod->fogs = out;
mod->numfogs = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
if ( LittleLong ( in->visibleSide ) == -1 )
{
continue;
}
brush = prv->brushes + LittleLong ( in->brushNum );
brushsides = brush->brushside;
visibleside = brushsides + LittleLong ( in->visibleSide );
out->visibleplane = visibleside->plane;
Q_strncpyz(out->shadername, in->shader, sizeof(out->shadername));
out->numplanes = brush->numsides;
out->planes = ZG_Malloc(&mod->memgroup, out->numplanes*sizeof(cplane_t *));
for ( j = 0; j < out->numplanes; j++ )
{
out->planes[j] = brushsides[j].plane;
}
}
return true;
}
mfog_t *Mod_FogForOrigin(model_t *wmodel, vec3_t org)
{
int i, j;
mfog_t *ret;
float dot;
if (!wmodel || wmodel->loadstate != MLS_LOADED)
return NULL;
for ( i=0 , ret=wmodel->fogs ; i<wmodel->numfogs ; i++, ret++)
{
if (!ret->shader)
continue;
for (j = 0; j < ret->numplanes; j++)
{
dot = DotProduct(ret->planes[j]->normal, org);
if (dot - ret->planes[j]->dist > 0)
break;
}
if (j == ret->numplanes)
{
return ret;
}
}
return NULL;
}
//Convert a patch in to a list of glpolys
#define MAX_ARRAY_VERTS 65535
static index_t tempIndexesArray[MAX_ARRAY_VERTS*6];
static void GL_SizePatchFixed(mesh_t *mesh, int patchwidth, int patchheight, int numverts, int firstvert, cminfo_t *prv)
{
unsigned short patch_cp[2];
int step[2], size[2];
patch_cp[0] = patchwidth&0xffff;
patch_cp[1] = patchheight&0xffff;
if (patch_cp[0] <= 0 || patch_cp[1] <= 0 )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
// allocate space for mesh
step[0] = patchwidth>>16;
step[1] = patchheight>>16;
if (!step[0] || !step[1])
{
size[0] = patch_cp[0];
size[1] = patch_cp[1];
}
else
{
size[0] = (patch_cp[0] / 2) * step[0] + 1;
size[1] = (patch_cp[1] / 2) * step[1] + 1;
}
mesh->numvertexes = size[0] * size[1];
mesh->numindexes = (size[0]-1) * (size[1]-1) * 6;
}
static void GL_SizePatch(mesh_t *mesh, int patchwidth, int patchheight, int numverts, int firstvert, cminfo_t *prv)
{
unsigned short patch_cp[2];
int step[2], size[2], flat[2];
float subdivlevel;
patch_cp[0] = patchwidth;
patch_cp[1] = patchheight;
if (patch_cp[0] <= 0 || patch_cp[1] <= 0 )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
subdivlevel = r_subdivisions.value;
if ( subdivlevel < 1 )
subdivlevel = 1;
// find the degree of subdivision in the u and v directions
Patch_GetFlatness ( subdivlevel, prv->verts[firstvert], sizeof(vecV_t)/sizeof(vec_t), patch_cp, flat );
// allocate space for mesh
step[0] = (1 << flat[0]);
step[1] = (1 << flat[1]);
size[0] = (patch_cp[0] / 2) * step[0] + 1;
size[1] = (patch_cp[1] / 2) * step[1] + 1;
mesh->numvertexes = size[0] * size[1];
mesh->numindexes = (size[0]-1) * (size[1]-1) * 6;
}
//mesh_t *GL_CreateMeshForPatch ( model_t *mod, q3dface_t *surf )
static void GL_CreateMeshForPatch (model_t *mod, mesh_t *mesh, int patchwidth, int patchheight, int numverts, int firstvert)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int numindexes, step[2], size[2], flat[2], i, u, v, p;
unsigned short patch_cp[2];
index_t *indexes;
float subdivlevel;
int sty;
patch_cp[0] = patchwidth;
patch_cp[1] = patchheight;
if (patch_cp[0] <= 0 || patch_cp[1] <= 0 )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
subdivlevel = r_subdivisions.value;
if ( subdivlevel < 1 )
subdivlevel = 1;
// find the degree of subdivision in the u and v directions
Patch_GetFlatness ( subdivlevel, prv->verts[firstvert], sizeof(vecV_t)/sizeof(vec_t), patch_cp, flat );
// allocate space for mesh
step[0] = (1 << flat[0]);
step[1] = (1 << flat[1]);
size[0] = (patch_cp[0] / 2) * step[0] + 1;
size[1] = (patch_cp[1] / 2) * step[1] + 1;
numverts = size[0] * size[1];
if ( numverts < 0 || numverts > MAX_ARRAY_VERTS )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
if (mesh->numvertexes != numverts)
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
// fill in
Patch_Evaluate ( prv->verts[firstvert], patch_cp, step, mesh->xyz_array[0], sizeof(vecV_t)/sizeof(vec_t));
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
if (mesh->colors4f_array[sty])
Patch_Evaluate ( prv->colors4f_array[sty][firstvert], patch_cp, step, mesh->colors4f_array[sty][0], 4 );
}
Patch_Evaluate ( prv->normals_array[firstvert], patch_cp, step, mesh->normals_array[0], 3 );
Patch_Evaluate ( prv->vertstmexcoords[firstvert], patch_cp, step, mesh->st_array[0], 2 );
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
if (mesh->lmst_array[sty])
Patch_Evaluate ( prv->vertlstmexcoords[sty][firstvert], patch_cp, step, mesh->lmst_array[sty][0], 2 );
}
// compute new indexes avoiding adding invalid triangles
numindexes = 0;
indexes = tempIndexesArray;
for (v = 0, i = 0; v < size[1]-1; v++)
{
for (u = 0; u < size[0]-1; u++, i += 6)
{
indexes[0] = p = v * size[0] + u;
indexes[1] = p + size[0];
indexes[2] = p + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
indexes[0] = p + 1;
indexes[1] = p + size[0];
indexes[2] = p + size[0] + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
}
}
// allocate and fill index table
mesh->numindexes = numindexes;
memcpy (mesh->indexes, tempIndexesArray, numindexes * sizeof(index_t) );
}
static void GL_CreateMeshForPatchFixed (model_t *mod, mesh_t *mesh, int patchwidth, int patchheight, int numverts, int firstvert)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int numindexes, step[2], size[2], i, u, v, p;
unsigned short patch_cp[2];
index_t *indexes;
float subdivlevel;
int sty;
patch_cp[0] = patchwidth&0xffff;
patch_cp[1] = patchheight&0xffff;
if (patch_cp[0] <= 0 || patch_cp[1] <= 0 )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
subdivlevel = r_subdivisions.value;
if ( subdivlevel < 1 )
subdivlevel = 1;
// allocate space for mesh
step[0] = patchwidth>>16;
step[1] = patchheight>>16;
if (!step[0] || !step[1])
{ //explicit CPs only.
size[0] = patch_cp[0];
size[1] = patch_cp[1];
}
else
{
size[0] = (patch_cp[0] / 2) * step[0] + 1;
size[1] = (patch_cp[1] / 2) * step[1] + 1;
}
numverts = size[0] * size[1];
if ( numverts < 0 || numverts > MAX_ARRAY_VERTS )
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
if (mesh->numvertexes != numverts)
{
mesh->numindexes = 0;
mesh->numvertexes = 0;
return;
}
// fill in
Patch_Evaluate ( prv->verts[firstvert], patch_cp, step, mesh->xyz_array[0], sizeof(vecV_t)/sizeof(vec_t));
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
if (mesh->colors4f_array[sty])
Patch_Evaluate ( prv->colors4f_array[sty][firstvert], patch_cp, step, mesh->colors4f_array[sty][0], 4 );
}
Patch_Evaluate ( prv->normals_array[firstvert], patch_cp, step, mesh->normals_array[0], 3 );
Patch_Evaluate ( prv->vertstmexcoords[firstvert], patch_cp, step, mesh->st_array[0], 2 );
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
if (mesh->lmst_array[sty])
Patch_Evaluate ( prv->vertlstmexcoords[sty][firstvert], patch_cp, step, mesh->lmst_array[sty][0], 2 );
}
// compute new indexes avoiding adding invalid triangles
numindexes = 0;
indexes = tempIndexesArray;
for (v = 0, i = 0; v < size[1]-1; v++)
{
for (u = 0; u < size[0]-1; u++, i += 6)
{
indexes[0] = p = v * size[0] + u;
indexes[1] = p + size[0];
indexes[2] = p + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
indexes[0] = p + 1;
indexes[1] = p + size[0];
indexes[2] = p + size[0] + 1;
// if ( !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[1]]) &&
// !VectorEquals(mesh->xyz_array[indexes[0]], mesh->xyz_array[indexes[2]]) &&
// !VectorEquals(mesh->xyz_array[indexes[1]], mesh->xyz_array[indexes[2]]) )
{
indexes += 3;
numindexes += 3;
}
}
}
// allocate and fill index table
mesh->numindexes = numindexes;
memcpy (mesh->indexes, tempIndexesArray, numindexes * sizeof(index_t) );
}
#ifdef RFBSPS
static void CModRBSP_BuildSurfMesh(model_t *mod, msurface_t *out, builddata_t *bd)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
rbspface_t *in = (rbspface_t*)(bd+1);
int idx = (out - mod->surfaces) - mod->firstmodelsurface;
int sty;
in += idx;
if (LittleLong(in->facetype) == MST_PATCH)
{
GL_CreateMeshForPatch(mod, out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex));
}
else if (LittleLong(in->facetype) == MST_PATCH_FIXED)
{
GL_CreateMeshForPatchFixed(mod, out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex));
}
else if (LittleLong(in->facetype) == MST_PLANAR || LittleLong(in->facetype) == MST_TRIANGLE_SOUP)
{
unsigned int fv = LittleLong(in->firstvertex), i;
for (i = 0; i < out->mesh->numvertexes; i++)
{
VectorCopy(prv->verts[fv + i], out->mesh->xyz_array[i]);
Vector2Copy(prv->vertstmexcoords[fv + i], out->mesh->st_array[i]);
for (sty = 0; sty < MAXRLIGHTMAPS; sty++)
{
Vector2Copy(prv->vertlstmexcoords[sty][fv + i], out->mesh->lmst_array[sty][i]);
Vector4Copy(prv->colors4f_array[sty][fv + i], out->mesh->colors4f_array[sty][i]);
}
VectorCopy(prv->normals_array[fv + i], out->mesh->normals_array[i]);
}
fv = LittleLong(in->firstindex);
for (i = 0; i < out->mesh->numindexes; i++)
{
out->mesh->indexes[i] = prv->surfindexes[fv + i];
}
}
else
{
/* //flare
int r, g, b;
extern index_t r_quad_indexes[6];
static vec2_t st[4] = {{0,0},{0,1},{1,1},{1,0}};
mesh = out->mesh = (mesh_t *)Hunk_Alloc(sizeof(mesh_t));
mesh->xyz_array = (vecV_t *)Hunk_Alloc(sizeof(vecV_t)*4);
mesh->colors4b_array = (byte_vec4_t *)Hunk_Alloc(sizeof(byte_vec4_t)*4);
mesh->numvertexes = 4;
mesh->indexes = r_quad_indexes;
mesh->st_array = st;
mesh->numindexes = 6;
VectorCopy (in->lightmap_origin, mesh->xyz_array[0]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[1]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[2]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[3]);
r = LittleFloat(in->lightmap_vecs[0][0]) * 255.0f;
r = bound (0, r, 255);
g = LittleFloat(in->lightmap_vecs[0][1]) * 255.0f;
g = bound (0, g, 255);
b = LittleFloat(in->lightmap_vecs[0][2]) * 255.0f;
b = bound (0, b, 255);
mesh->colors4b_array[0][0] = r;
mesh->colors4b_array[0][1] = g;
mesh->colors4b_array[0][2] = b;
mesh->colors4b_array[0][3] = 255;
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[1]);
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[2]);
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[3]);
*/
}
Mod_AccumulateMeshTextureVectors(out->mesh);
Mod_NormaliseTextureVectors(out->mesh->normals_array, out->mesh->snormals_array, out->mesh->tnormals_array, out->mesh->numvertexes, false);
}
#endif
static void CModQ3_BuildSurfMesh(model_t *mod, msurface_t *out, builddata_t *bd)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int idx = (out - mod->surfaces) - mod->firstmodelsurface;
q3dface_t *in = (q3dface_t*)(bd+1) + idx;
int facetype = LittleLong(in->facetype);
if (facetype == MST_PATCH)
{
GL_CreateMeshForPatch(mod, out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex));
}
else if (facetype == MST_PATCH_FIXED)
{
GL_CreateMeshForPatchFixed(mod, out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex));
}
else if (facetype == MST_PLANAR || facetype == MST_TRIANGLE_SOUP)
{
unsigned int fv = LittleLong(in->firstvertex), fi = LittleLong(in->firstindex), i;
for (i = 0; i < out->mesh->numvertexes; i++)
{
VectorCopy(prv->verts[fv + i], out->mesh->xyz_array[i]);
Vector2Copy(prv->vertstmexcoords[fv + i], out->mesh->st_array[i]);
Vector2Copy(prv->vertlstmexcoords[0][fv + i], out->mesh->lmst_array[0][i]);
Vector4Copy(prv->colors4f_array[0][fv + i], out->mesh->colors4f_array[0][i]);
VectorCopy(prv->normals_array[fv + i], out->mesh->normals_array[i]);
}
for (i = 0; i < out->mesh->numindexes; i++)
{
out->mesh->indexes[i] = prv->surfindexes[fi + i];
}
}
else
{
/* //flare
int r, g, b;
extern index_t r_quad_indexes[6];
static vec2_t st[4] = {{0,0},{0,1},{1,1},{1,0}};
mesh = out->mesh = (mesh_t *)Hunk_Alloc(sizeof(mesh_t));
mesh->xyz_array = (vecV_t *)Hunk_Alloc(sizeof(vecV_t)*4);
mesh->colors4b_array = (byte_vec4_t *)Hunk_Alloc(sizeof(byte_vec4_t)*4);
mesh->numvertexes = 4;
mesh->indexes = r_quad_indexes;
mesh->st_array = st;
mesh->numindexes = 6;
VectorCopy (in->lightmap_origin, mesh->xyz_array[0]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[1]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[2]);
VectorCopy (in->lightmap_origin, mesh->xyz_array[3]);
r = LittleFloat(in->lightmap_vecs[0][0]) * 255.0f;
r = bound (0, r, 255);
g = LittleFloat(in->lightmap_vecs[0][1]) * 255.0f;
g = bound (0, g, 255);
b = LittleFloat(in->lightmap_vecs[0][2]) * 255.0f;
b = bound (0, b, 255);
mesh->colors4b_array[0][0] = r;
mesh->colors4b_array[0][1] = g;
mesh->colors4b_array[0][2] = b;
mesh->colors4b_array[0][3] = 255;
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[1]);
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[2]);
Vector4Copy(mesh->colors4b_array[0], mesh->colors4b_array[3]);
*/
}
Mod_AccumulateMeshTextureVectors(out->mesh);
Mod_NormaliseTextureVectors(out->mesh->normals_array, out->mesh->snormals_array, out->mesh->tnormals_array, out->mesh->numvertexes, false);
}
static qboolean CModQ3_LoadRFaces (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
extern cvar_t r_vertexlight;
q3dface_t *in;
msurface_t *out;
mplane_t *pl;
int facetype;
int count;
int surfnum;
int fv;
int sty;
mesh_t *mesh;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",mod->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
pl = ZG_Malloc(&mod->memgroup, count*sizeof(*pl));//create a new array of planes for speed.
mesh = ZG_Malloc(&mod->memgroup, count*sizeof(*mesh));
mod->surfaces = out;
mod->numsurfaces = count;
mod->lightmaps.first = 0;
for (surfnum = 0; surfnum < count; surfnum++, out++, in++, pl++)
{
out->plane = pl;
facetype = LittleLong(in->facetype);
out->texinfo = mod->texinfo + LittleLong(in->shadernum);
out->lightmaptexturenums[0] = LittleLong(in->lightmapnum);
if (facetype == MST_FLARE)
out->texinfo = mod->texinfo + mod->numtexinfo*2;
else if (out->lightmaptexturenums[0] < 0 /*|| facetype == MST_TRIANGLE_SOUP*/ || r_vertexlight.value)
out->texinfo += mod->numtexinfo; //various surfaces use a different version of the same shader (with all the lightmaps collapsed)
out->light_s[0] = LittleLong(in->lightmap_offs[0]);
out->light_t[0] = LittleLong(in->lightmap_offs[1]);
out->styles[0] = INVALID_LIGHTSTYLE;
out->vlstyles[0] = INVALID_VLIGHTSTYLE;
for (sty = 1; sty < MAXRLIGHTMAPS; sty++)
{
out->styles[sty] = INVALID_LIGHTSTYLE;
out->vlstyles[sty] = INVALID_VLIGHTSTYLE;
out->lightmaptexturenums[sty] = -1;
}
for (; sty < MAXCPULIGHTMAPS; sty++)
out->styles[sty] = INVALID_LIGHTSTYLE;
out->lmshift = LMSHIFT_DEFAULT;
//fixme: determine texturemins from lightmap_origin
out->extents[0] = (LittleLong(in->lightmap_width)-1)<<out->lmshift;
out->extents[1] = (LittleLong(in->lightmap_height)-1)<<out->lmshift;
out->samples=NULL;
if (mod->lightmaps.count < out->lightmaptexturenums[0]+1)
mod->lightmaps.count = out->lightmaptexturenums[0]+1;
fv = LittleLong(in->firstvertex);
{
vec3_t v[3];
VectorCopy(prv->verts[fv+0], v[0]);
VectorCopy(prv->verts[fv+1], v[1]);
VectorCopy(prv->verts[fv+2], v[2]);
PlaneFromPoints(v, pl);
CategorizePlane(pl);
}
if (prv->surfaces[LittleLong(in->shadernum)].c.value == 0 || prv->surfaces[LittleLong(in->shadernum)].c.value & Q3CONTENTS_TRANSLUCENT)
//q3dm10's thingie is 0
out->flags |= SURF_DRAWALPHA;
if (mod->texinfo[LittleLong(in->shadernum)].flags & TI_SKY)
out->flags |= SURF_DRAWSKY;
if (LittleLong(in->fognum) == -1 || !mod->numfogs)
out->fog = NULL;
else
out->fog = mod->fogs + LittleLong(in->fognum);
if (prv->surfaces[LittleLong(in->shadernum)].c.flags & (Q3SURF_NODRAW | Q3SURF_SKIP))
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = 0;
out->mesh->numvertexes = 0;
}
else if (facetype == MST_PATCH)
{
out->mesh = &mesh[surfnum];
GL_SizePatch(out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex), prv);
}
else if (facetype == MST_PATCH_FIXED)
{
out->mesh = &mesh[surfnum];
GL_SizePatchFixed(out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex), prv);
}
else if (facetype == MST_PLANAR || facetype == MST_TRIANGLE_SOUP)
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = LittleLong(in->num_indexes);
out->mesh->numvertexes = LittleLong(in->num_vertices);
/*
Mod_AccumulateMeshTextureVectors(out->mesh);
*/
}
else
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = 6;
out->mesh->numvertexes = 4;
}
}
Mod_SortShaders(mod);
return true;
}
#ifdef RFBSPS
static qboolean CModRBSP_LoadRFaces (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
extern cvar_t r_vertexlight;
rbspface_t *in;
msurface_t *out;
mplane_t *pl;
int facetype;
int count;
int surfnum;
int fv;
int j;
mesh_t *mesh;
int maxstyle = q3bsp_ignorestyles.ival?1:min(MAXRLIGHTMAPS, RBSP_STYLESPERSURF);
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",mod->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&mod->memgroup, count*sizeof(*out));
pl = ZG_Malloc(&mod->memgroup, count*sizeof(*pl));//create a new array of planes for speed.
mesh = ZG_Malloc(&mod->memgroup, count*sizeof(*mesh));
mod->surfaces = out;
mod->numsurfaces = count;
for (surfnum = 0; surfnum < count; surfnum++, out++, in++, pl++)
{
out->plane = pl;
facetype = LittleLong(in->facetype);
out->texinfo = mod->texinfo + LittleLong(in->shadernum);
for (j = 0; j < maxstyle; j++)
{
out->lightmaptexturenums[j] = LittleLong(in->lightmapnum[j]);
out->light_s[j] = LittleLong(in->lightmap_offs[0][j]);
out->light_t[j] = LittleLong(in->lightmap_offs[1][j]);
out->styles[j] = (in->lm_styles[j]!=255)?in->lm_styles[j]:INVALID_LIGHTSTYLE;
out->vlstyles[j] = (in->vt_styles[j]!=255)?in->vt_styles[j]:INVALID_VLIGHTSTYLE;
if (mod->lightmaps.count < out->lightmaptexturenums[j]+1)
mod->lightmaps.count = out->lightmaptexturenums[j]+1;
}
for (; j < MAXRLIGHTMAPS; j++)
{
out->lightmaptexturenums[j] = -1;
out->light_s[j] = 0;
out->light_t[j] = 0;
out->styles[j] = INVALID_LIGHTSTYLE;
out->vlstyles[j] = INVALID_VLIGHTSTYLE;
}
for (; j < MAXCPULIGHTMAPS; j++)
out->styles[j] = INVALID_LIGHTSTYLE;
if (facetype == MST_FLARE)
out->texinfo = mod->texinfo + mod->numtexinfo*2;
else if (out->lightmaptexturenums[0]<0 || r_vertexlight.value)
out->texinfo += mod->numtexinfo; //soup/vertex light uses a different version of the same shader (with all the lightmaps collapsed)
out->lmshift = LMSHIFT_DEFAULT;
out->extents[0] = (LittleLong(in->lightmap_width)-1)<<out->lmshift;
out->extents[1] = (LittleLong(in->lightmap_height)-1)<<out->lmshift;
out->samples=NULL;
fv = LittleLong(in->firstvertex);
{
vec3_t v[3];
VectorCopy(prv->verts[fv+0], v[0]);
VectorCopy(prv->verts[fv+1], v[1]);
VectorCopy(prv->verts[fv+2], v[2]);
PlaneFromPoints(v, pl);
CategorizePlane(pl);
}
if (prv->surfaces[in->shadernum].c.value == 0 || prv->surfaces[in->shadernum].c.value & Q3CONTENTS_TRANSLUCENT)
//q3dm10's thingie is 0
out->flags |= SURF_DRAWALPHA;
if (mod->texinfo[in->shadernum].flags & TI_SKY)
out->flags |= SURF_DRAWSKY;
if (in->fognum < 0 || in->fognum >= mod->numfogs)
out->fog = NULL;
else
out->fog = mod->fogs + in->fognum;
if (prv->surfaces[LittleLong(in->shadernum)].c.flags & (Q3SURF_NODRAW | Q3SURF_SKIP))
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = 0;
out->mesh->numvertexes = 0;
}
else if (facetype == MST_PATCH)
{
out->mesh = &mesh[surfnum];
GL_SizePatch(out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex), prv);
}
else if (facetype == MST_PATCH_FIXED)
{
out->mesh = &mesh[surfnum];
GL_SizePatchFixed(out->mesh, LittleLong(in->patchwidth), LittleLong(in->patchheight), LittleLong(in->num_vertices), LittleLong(in->firstvertex), prv);
}
else if (facetype == MST_PLANAR || facetype == MST_TRIANGLE_SOUP)
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = LittleLong(in->num_indexes);
out->mesh->numvertexes = LittleLong(in->num_vertices);
/*
Mod_AccumulateMeshTextureVectors(out->mesh);
*/
}
else
{
out->mesh = &mesh[surfnum];
out->mesh->numindexes = 6;
out->mesh->numvertexes = 4;
}
}
Mod_SortShaders(mod);
return true;
}
#endif
#endif
static qboolean CModQ3_LoadNodes (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
int i, j, count, p;
q3dnode_t *in;
mnode_t *out;
//dnode_t
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",loadmodel->name);
return false;
}
count = l->filelen / sizeof(*in);
out = ZG_Malloc(&loadmodel->memgroup, count*sizeof(*out));
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Too many nodes on map\n");
return false;
}
loadmodel->nodes = out;
loadmodel->numnodes = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
for (j=0 ; j<3 ; j++)
{
out->minmaxs[j] = LittleLong (in->mins[j]);
out->minmaxs[3+j] = LittleLong (in->maxs[j]);
}
AddPointToBounds(out->minmaxs, loadmodel->mins, loadmodel->maxs);
AddPointToBounds(out->minmaxs+3, loadmodel->mins, loadmodel->maxs);
p = LittleLong(in->plane);
out->plane = loadmodel->planes + p;
out->firstsurface = 0;//LittleShort (in->firstface);
out->numsurfaces = 0;//LittleShort (in->numfaces);
out->contents = -1;
for (j=0 ; j<2 ; j++)
{
p = LittleLong (in->children[j]);
out->childnum[j] = p;
if (p >= 0)
{
out->children[j] = loadmodel->nodes + p;
}
else
out->children[j] = (mnode_t *)(loadmodel->leafs + (-1 - p));
}
}
CMod_SetParent (loadmodel->nodes, NULL); // sets nodes and leafs
return true;
}
static qboolean CModQ3_LoadBrushes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
q3dbrush_t *in;
q2cbrush_t *out;
int i, count;
int shaderref;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count > SANITY_MAX_MAP_BRUSHES)
{
Con_Printf (CON_ERROR "Map has too many brushes");
return false;
}
prv->brushes = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+1));
out = prv->brushes;
prv->numbrushes = count;
for (i=0 ; i<count ; i++, out++, in++)
{
shaderref = LittleLong ( in->shadernum );
out->contents = prv->surfaces[shaderref].c.value;
out->brushside = &prv->brushsides[LittleLong ( in->firstside )];
out->numsides = LittleLong ( in->num_sides );
CM_FinalizeBrush(out);
}
return true;
}
static qboolean CModQ3_LoadLeafs (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i, j;
mleaf_t *out;
q3dleaf_t *in;
int count;
q2cbrush_t **brush;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no leafs\n");
return false;
}
// need to save space for box planes
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Too many leaves on map");
return false;
}
out = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+1));
mod->leafs = out;
mod->numleafs = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
for (j = 0; j < 3; j++)
{
out->minmaxs[0+j] = LittleLong(in->mins[j]);
out->minmaxs[3+j] = LittleLong(in->maxs[j]);
}
out->cluster = LittleLong(in->cluster);
out->area = LittleLong(in->area);
// out->firstleafface = LittleLong(in->firstleafsurface);
// out->numleaffaces = LittleLong(in->num_leafsurfaces);
out->contents = 0;
out->firstleafbrush = LittleLong(in->firstleafbrush);
out->numleafbrushes = LittleLong(in->num_leafbrushes);
out->firstmarksurface = mod->marksurfaces + LittleLong(in->firstleafsurface);
out->nummarksurfaces = LittleLong(in->num_leafsurfaces);
if (out->minmaxs[0] > out->minmaxs[3+0] || out->minmaxs[1] > out->minmaxs[3+1] ||
out->minmaxs[2] > out->minmaxs[3+2])// || VectorEquals (out->minmaxs, out->minmaxs+3))
{
out->nummarksurfaces = 0;
}
brush = &prv->leafbrushes[out->firstleafbrush];
for (j=0 ; j<out->numleafbrushes ; j++)
{
out->contents |= brush[j]->contents;
}
if (out->area >= prv->numareas)
{
prv->numareas = out->area + 1;
}
}
return true;
}
static qboolean CModQ3_LoadPlanes (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
int i, j;
mplane_t *out;
Q3PLANE_t *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",loadmodel->name);
return false;
}
count = l->filelen / sizeof(*in);
if (count > SANITY_LIMIT(*out))
{
Con_Printf (CON_ERROR "Too many planes on map (%i)\n", count);
return false;
}
loadmodel->planes = out = ZG_Malloc(&loadmodel->memgroup, sizeof(*out) * count);
loadmodel->numplanes = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
for (j=0 ; j<3 ; j++)
{
out->normal[j] = LittleFloat (in->n[j]);
}
out->dist = LittleFloat (in->d);
CategorizePlane(out);
}
return true;
}
static qboolean CModQ3_LoadLeafBrushes (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
q2cbrush_t **out;
int *in;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
if (count < 1)
{
Con_Printf (CON_ERROR "Map with no leafbrushes\n");
return false;
}
// need to save space for box planes
if (count > SANITY_MAX_MAP_LEAFBRUSHES)
{
Con_Printf (CON_ERROR "Map has too many leafbrushes\n");
return false;
}
//prv->numbrushes is because of submodels being weird.
out = prv->leafbrushes = ZG_Malloc(&mod->memgroup, sizeof(*out) * (count+prv->numbrushes));
prv->numleafbrushes = count;
for ( i=0 ; i<count ; i++, in++, out++)
*out = prv->brushes + (unsigned int)LittleLong (*in);
return true;
}
static qboolean CModQ3_LoadBrushSides (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i, j;
q2cbrushside_t *out;
q3dbrushside_t *in;
int count;
int num;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
// need to save space for box planes
if (count > SANITY_MAX_MAP_BRUSHSIDES)
{
Con_Printf (CON_ERROR "Map has too many brushsides (%i)\n", count);
return false;
}
out = prv->brushsides = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);
prv->numbrushsides = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
num = LittleLong (in->planenum);
out->plane = &mod->planes[num];
j = LittleLong (in->texinfo);
if (j >= mod->numtexinfo)
{
Con_Printf (CON_ERROR "Bad brushside texinfo\n");
return false;
}
out->surface = &prv->surfaces[j];
}
return true;
}
#ifdef RFBSPS
static qboolean CModRBSP_LoadBrushSides (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i, j;
q2cbrushside_t *out;
rbspbrushside_t *in;
int count;
int num;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size\n");
return false;
}
count = l->filelen / sizeof(*in);
// need to save space for box planes
if (count > SANITY_MAX_MAP_BRUSHSIDES)
{
Con_Printf (CON_ERROR "Map has too many brushsides (%i)\n", count);
return false;
}
out = prv->brushsides = ZG_Malloc(&mod->memgroup, sizeof(*out) * count);
prv->numbrushsides = count;
for ( i=0 ; i<count ; i++, in++, out++)
{
num = LittleLong (in->planenum);
out->plane = &mod->planes[num];
j = LittleLong (in->texinfo);
if (j >= mod->numtexinfo)
{
Con_Printf (CON_ERROR "Bad brushside texinfo\n");
return false;
}
out->surface = &prv->surfaces[j];
}
return true;
}
#endif
static qboolean CModQ3_LoadVisibility (model_t *mod, qbyte *mod_base, lump_t *l)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
unsigned int numclusters;
if (l->filelen == 0)
{
int i;
#if 0
//the 'correct' code
numclusters = 0;
for (i = 0; i < mod->numleafs; i++)
if (numclusters < mod->leafs[i].cluster+1)
numclusters = mod->leafs[i].cluster+1;
numclusters++;
#else
//but its much faster to merge all leafs into a single pvs cluster. no vis is no vis.
numclusters = 8*sizeof(int);
for (i = 0; i < mod->numleafs; i++)
mod->leafs[i].cluster = !!mod->leafs[i].cluster;
#endif
prv->q3pvs = ZG_Malloc(&mod->memgroup, sizeof(*prv->q3pvs) + (numclusters+7)/8 * numclusters);
memset (prv->q3pvs, 0xff, sizeof(*prv->q3pvs) + (numclusters+7)/8 * numclusters);
prv->q3pvs->numclusters = numclusters;
prv->numvisibility = 0;
prv->q3pvs->rowsize = (prv->q3pvs->numclusters+7)/8;
}
else
{
prv->numvisibility = l->filelen;
prv->q3pvs = ZG_Malloc(&mod->memgroup, l->filelen);
mod->vis = (q2dvis_t *)prv->q3pvs;
memcpy (prv->q3pvs, mod_base + l->fileofs, l->filelen);
numclusters = prv->q3pvs->numclusters = LittleLong (prv->q3pvs->numclusters);
prv->q3pvs->rowsize = LittleLong (prv->q3pvs->rowsize);
}
mod->numclusters = numclusters;
mod->pvsbytes = ((mod->numclusters + 31)>>3)&~3;
return true;
}
#ifndef SERVERONLY
static void CModQ3_LoadLighting (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
qbyte *in = mod_base + l->fileofs;
qbyte *out;
unsigned int samples = l->filelen;
int m, s, t;
int mapstride = loadmodel->lightmaps.width*3;
int mapsize = mapstride*loadmodel->lightmaps.height;
int maps;
int merge;
int mergestride;
extern cvar_t gl_overbright;
float scale = (1<<(2-gl_overbright.ival));
loadmodel->lightmaps.fmt = LM_L8;
//round up the samples, in case the last one is partial.
maps = ((samples+mapsize-1)&~(mapsize-1)) / mapsize;
//q3 maps have built in 4-fold overbright.
//if we're not rendering with that, we need to brighten the lightmaps in order to keep the darker parts the same brightness. we loose the 2 upper bits. those bright areas become uniform and indistinct.
gl_overbright.flags |= CVAR_RENDERERLATCH;
BuildLightMapGammaTable(1, (1<<(2-gl_overbright.ival)));
loadmodel->lightmaps.mergew = 0;
loadmodel->lightmaps.mergeh = 0;
loadmodel->engineflags |= MDLF_NEEDOVERBRIGHT;
if (!samples)
return;
loadmodel->lightmaps.fmt = LM_RGB8;
if (loadmodel->lightmaps.deluxemapping)
maps /= 2;
{
int limitw = sh_config.texture2d_maxsize / loadmodel->lightmaps.width;
int limith = sh_config.texture2d_maxsize / loadmodel->lightmaps.height;
if (!q3bsp_mergeq3lightmaps.ival)
{
limitw = 1;
limith = 1;
}
loadmodel->lightmaps.mergeh = loadmodel->lightmaps.mergew = 1;
while (loadmodel->lightmaps.mergew*loadmodel->lightmaps.mergeh < maps)
{ //this could probably be smarter.
if (loadmodel->lightmaps.mergew*2 <= limitw && loadmodel->lightmaps.mergew < loadmodel->lightmaps.mergeh)
loadmodel->lightmaps.mergew *= 2;
else if (loadmodel->lightmaps.mergeh*2 <= limith)
loadmodel->lightmaps.mergeh *= 2;
else if (loadmodel->lightmaps.mergew*2 <= limitw)
loadmodel->lightmaps.mergew *= 2;
else
break; //can't expand in either direction.
}
}
merge = loadmodel->lightmaps.mergew*loadmodel->lightmaps.mergeh;
mergestride = loadmodel->lightmaps.mergew*mapstride;
//q3bsp itself does not support deluxemapping.
//the way it works is by interleaving the data in lightmap+deluxemap pairs.
//the surface data makes no references to the deluxemap maps, they're implied by lightmap+1
//if no surface references an odd lightmap index then we know we have deluxemaps... assuming there are at least two lightmaps.
//q3map2 likes generating null lightmaps, so beware false positives.
//note that external lighting makes this even more fun.
//if we have deluxemapping data then we split it here. beware externals.
if (loadmodel->lightmaps.deluxemapping)
{
m = merge;
while (m < maps)
m += merge;
loadmodel->lightdata = ZG_Malloc(&loadmodel->memgroup, mapsize*m*2);
loadmodel->lightdatasize = mapsize*m*2;
}
else
{
m = merge;
while (m < maps)
m += merge;
loadmodel->lightdata = ZG_Malloc(&loadmodel->memgroup, mapsize*m);
loadmodel->lightdatasize = mapsize*m;
}
if (!loadmodel->lightdata)
return;
//be careful here, q3bsp deluxemapping is done using interleaving. we want to unoverbright ONLY lightmaps and not deluxemaps.
for (m = 0; m < maps; m++)
{
out = loadmodel->lightdata;
//figure out which merged lightmap we're putting it into
out += (m/merge)*merge*mapsize * (loadmodel->lightmaps.deluxemapping?2:1);
//and the submap
s = m%merge;
t = s/loadmodel->lightmaps.mergew;
s = s%loadmodel->lightmaps.mergew;
out += s*mapstride;
out += t*mergestride*loadmodel->lightmaps.height;
//q3bsp has 4-fold overbrights, so if we're not using overbrights then we basically need to scale the values up by 4
//this will require clamping, which can result in oversaturation of channels, meaning discolouration
for (t = 0; t < loadmodel->lightmaps.height; t++)
{
for (s = 0; s < loadmodel->lightmaps.width; s++)
{
float i;
vec3_t l;
l[0] = *in++;
l[1] = *in++;
l[2] = *in++;
VectorScale(l, scale, l); //it should be noted that this maths is wrong if you're trying to use srgb lightmaps.
i = max(l[0], max(l[1], l[2]));
if (i > 255)
VectorScale(l, 255/i, l); //clamp the brightest channel, scaling the others down to retain chromiance.
*out++ = l[0];
*out++ = l[1];
*out++ = l[2];
}
out += mergestride-mapstride;
}
if (r_lightmap_saturation.value != 1.0f)
SaturateR8G8B8(out, mapsize, r_lightmap_saturation.value);
if (loadmodel->lightmaps.deluxemapping)
{
out -= mergestride*loadmodel->lightmaps.height;
out += merge*mapsize;
//no gamma for deluxemap
for (t = 0; t < loadmodel->lightmaps.height; t++)
{
for (s = 0; s < loadmodel->lightmaps.width; s++)
{
*out++ = in[0];
*out++ = in[1];
*out++ = in[2];
in += 3;
}
out += mergestride-mapstride;
}
}
}
/*for (; m%merge; m++)
{
out = loadmodel->lightdata;
//figure out which merged lightmap we're putting it into
out += (m/merge)*merge*mapsize * (loadmodel->lightmaps.deluxemapping?2:1);
//and the submap
out += (m%merge)*mapsize;
for(s = 0; s < mapsize; s+=3)
{
out[s+0] = 0;
out[s+1] = 255;
out[s+2] = 0;
}
}*/
}
static qboolean CModQ3_LoadLightgrid (model_t *loadmodel, qbyte *mod_base, lump_t *l)
{
dq3gridlight_t *in;
dq3gridlight_t *out;
q3lightgridinfo_t *grid;
int count;
in = (void *)(mod_base + l->fileofs);
if (l->filelen % sizeof(*in))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",loadmodel->name);
return false;
}
count = l->filelen / sizeof(*in);
grid = ZG_Malloc(&loadmodel->memgroup, sizeof(q3lightgridinfo_t) + count*sizeof(*out));
grid->lightgrid = (dq3gridlight_t*)(grid+1);
out = grid->lightgrid;
loadmodel->lightgrid = grid;
grid->numlightgridelems = count;
// lightgrid is all 8 bit
memcpy ( out, in, count*sizeof(*out) );
return true;
}
#ifdef RFBSPS
static qboolean CModRBSP_LoadLightgrid (model_t *loadmodel, qbyte *mod_base, lump_t *elements, lump_t *indexes)
{
unsigned short *iin;
rbspgridlight_t *ein;
unsigned short *iout;
rbspgridlight_t *eout;
q3lightgridinfo_t *grid;
int ecount;
int icount;
int i;
ein = (void *)(mod_base + elements->fileofs);
iin = (void *)(mod_base + indexes->fileofs);
if (indexes->filelen % sizeof(*iin) || elements->filelen % sizeof(*ein))
{
Con_Printf (CON_ERROR "MOD_LoadBmodel: funny lump size in %s\n",loadmodel->name);
return false;
}
icount = indexes->filelen / sizeof(*iin);
ecount = elements->filelen / sizeof(*ein);
grid = ZG_Malloc(&loadmodel->memgroup, sizeof(q3lightgridinfo_t) + ecount*sizeof(*eout) + icount*sizeof(*iout));
grid->rbspelements = (rbspgridlight_t*)((char *)grid + sizeof(q3lightgridinfo_t));
grid->rbspindexes = (unsigned short*)((char *)grid + sizeof(q3lightgridinfo_t) + ecount*sizeof(*eout));
eout = grid->rbspelements;
iout = grid->rbspindexes;
loadmodel->lightgrid = grid;
grid->numlightgridelems = icount;
// elements are all 8 bit
memcpy ( eout, ein, ecount*sizeof(*eout) );
for (i = 0; i < icount; i++)
iout[i] = LittleShort(iin[i]);
return true;
}
#endif
#endif
#endif
#if !defined(SERVERONLY) && (defined(Q2BSPS) || defined(Q3BSPS))
#ifdef IMAGEFMT_PCX
qbyte *ReadPCXPalette(qbyte *buf, int len, qbyte *out);
static int CM_GetQ2Palette (void)
{
char *f;
size_t sz;
sz = FS_LoadFile("pics/colormap.pcx", (void**)&f);
if (!f)
{
Con_Printf (CON_WARNING "Couldn't find pics/colormap.pcx\n");
return -1;
}
if (!ReadPCXPalette(f, sz, q2_palette))
{
Con_Printf (CON_WARNING "Couldn't read pics/colormap.pcx\n");
FS_FreeFile(f);
return -1;
}
FS_FreeFile(f);
#if 0
{
float inf;
qbyte palette[768];
qbyte *pal;
int i;
pal = q2_palette;
for (i=0 ; i<768 ; i++)
{
inf = ((pal[i]+1)/256.0)*255 + 0.5;
if (inf < 0)
inf = 0;
if (inf > 255)
inf = 255;
palette[i] = inf;
}
memcpy (q2_palette, palette, sizeof(palette));
}
#endif
return 0;
}
#endif
#endif
#if 0
static void CM_OpenAllPortals(model_t *mod, char *ents) //this is a compleate hack. About as compleate as possible.
{ //q2 levels contain a thingie called area portals. Basically, doors can seperate two areas and
//the engine knows when this portal is open, and weather to send ents from both sides of the door
//or not. It's not just ents, but also walls. We want to just open them by default and hope the
//progs knows how to close them.
char style[8];
char name[64];
if (!map_autoopenportals.value)
return;
while(*ents)
{
if (*ents == '{') //an entity
{
ents++;
*style = '\0';
*name = '\0';
while (*ents)
{
ents = COM_Parse(ents);
if (!strcmp(com_token, "classname"))
{
ents = COM_ParseOut(ents, name, sizeof(name));
}
else if (!strcmp(com_token, "style"))
{
ents = COM_ParseOut(ents, style, sizeof(style));
}
else if (*com_token == '}')
break;
else
ents = COM_Parse(ents); //other field
ents++;
}
if (!strcmp(name, "func_areaportal"))
{
CMQ2_SetAreaPortalState(mod, atoi(style), true);
}
}
ents++;
}
}
#endif
2023-03-27 17:22:00 +00:00
#if defined(Q3BSPS)
static void CalcClusterPHS(cminfo_t *prv, int cluster)
{
int j, k, l, index;
int bitbyte;
unsigned int *dest, *src;
qbyte *scan;
int numclusters = prv->q3pvs->numclusters;
int rowbytes = prv->q3pvs->rowsize;
int rowwords = rowbytes / sizeof(int);
scan = (qbyte *)prv->q3pvs->data;
dest = (unsigned int *)(prv->q3phs->data);
dest += rowwords*cluster;
scan += rowbytes*cluster;
for (j=0 ; j<rowbytes ; j++)
{
bitbyte = scan[j];
if (!bitbyte)
continue;
for (k=0 ; k<8 ; k++)
{
if (! (bitbyte & (1<<k)) )
continue;
// OR this pvs row into the phs
index = (j<<3) + k;
if (index >= numclusters)
{
// if (!buggytools)
// Con_Printf ("CM_CalcPHS: Bad bit(s) in PVS (%i >= %i)\n", index, numclusters); // pad bits should be 0
// buggytools = true;
}
else
{
src = (unsigned int *)(prv->q3pvs->data) + index*rowwords;
for (l=0 ; l<rowwords ; l++)
dest[l] |= src[l];
}
}
}
prv->phscalced[cluster>>3] |= 1<<(cluster&7);
}
2023-03-27 17:22:00 +00:00
#endif
#if defined(HAVE_SERVER) && defined(Q3BSPS)
static void CMQ3_CalcPHS (model_t *mod)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int rowbytes, rowwords;
int i, j, k, l, index;
int bitbyte;
unsigned int *dest, *src;
qbyte *scan;
int count, vcount;
int numclusters;
qboolean buggytools = false;
extern cvar_t sv_calcphs;
Con_DPrintf ("Building PHS...\n");
prv->q3phs = ZG_Malloc(&mod->memgroup, sizeof(*prv->q3phs) + prv->q3pvs->rowsize * prv->q3pvs->numclusters);
rowwords = prv->q3pvs->rowsize / sizeof(int);
rowbytes = prv->q3pvs->rowsize;
memset ( prv->q3phs, 0, sizeof(*prv->q3phs) + prv->q3pvs->rowsize * prv->q3pvs->numclusters );
prv->q3phs->rowsize = prv->q3pvs->rowsize;
prv->q3phs->numclusters = numclusters = prv->q3pvs->numclusters;
if (!numclusters)
return;
vcount = 0;
for (i=0 ; i<numclusters ; i++)
{
scan = CM_ClusterPVS (mod, i, NULL, PVM_FAST);
for (j=0 ; j<numclusters ; j++)
{
if ( scan[j>>3] & (1<<(j&7)) )
{
vcount++;
}
}
}
count = 0;
scan = (qbyte *)prv->q3pvs->data;
dest = (unsigned int *)(prv->q3phs->data);
if (sv_calcphs.ival >= 2)
{ //delay-calculate it.
prv->phscalced = ZG_Malloc(&mod->memgroup, (prv->q3pvs->numclusters+7)/8);
memcpy(dest, scan, rowbytes*numclusters);
Con_DPrintf ("Average clusters visible / total: %i / %i\n"
, vcount/numclusters, numclusters);
}
else if (!sv_calcphs.ival)
{ //disable calcs (behaves like broadcast so just fill with 1s)
memset(dest, 0xff, rowbytes*numclusters);
}
else
{ //the original slow logic like q3.
for (i=0 ; i<numclusters ; i++, dest += rowwords, scan += rowbytes)
{
memcpy (dest, scan, rowbytes);
for (j=0 ; j<rowbytes ; j++)
{
bitbyte = scan[j];
if (!bitbyte)
continue;
for (k=0 ; k<8 ; k++)
{
if (! (bitbyte & (1<<k)) )
continue;
// OR this pvs row into the phs
index = (j<<3) + k;
if (index >= numclusters)
{
if (!buggytools)
Con_Printf ("CM_CalcPHS: Bad bit(s) in PVS (%i >= %i)\n", index, numclusters); // pad bits should be 0
buggytools = true;
}
else
{
src = (unsigned int *)(prv->q3pvs->data) + index*rowwords;
for (l=0 ; l<rowwords ; l++)
dest[l] |= src[l];
}
}
}
for (j=0 ; j<numclusters ; j++)
if ( ((qbyte *)dest)[j>>3] & (1<<(j&7)) )
count++;
}
Con_DPrintf ("Average clusters visible / hearable / total: %i / %i / %i\n"
, vcount/numclusters, count/numclusters, numclusters);
}
}
#endif
/*
static qbyte *CM_LeafnumPVS (model_t *model, int leafnum, qbyte *buffer, unsigned int buffersize)
{
return CM_ClusterPVS(model, CM_LeafCluster(model, leafnum), buffer, buffersize);
}
*/
#ifndef SERVERONLY
#define GLQ2BSP_LightPointValues GLQ1BSP_LightPointValues
extern int r_dlightframecount;
static void Q2BSP_MarkLights (dlight_t *light, dlightbitmask_t bit, mnode_t *node)
{
mplane_t *splitplane;
float dist;
msurface_t *surf;
int i;
if (node->contents != -1)
{
mleaf_t *leaf = (mleaf_t *)node;
msurface_t **mark;
i = leaf->nummarksurfaces;
mark = leaf->firstmarksurface;
while(i--!=0)
{
surf = *mark++;
if (surf->dlightframe != r_dlightframecount)
{
surf->dlightbits = 0;
surf->dlightframe = r_dlightframecount;
}
surf->dlightbits |= bit;
}
return;
}
splitplane = node->plane;
dist = DotProduct (light->origin, splitplane->normal) - splitplane->dist;
if (dist > light->radius)
{
Q2BSP_MarkLights (light, bit, node->children[0]);
return;
}
if (dist < -light->radius)
{
Q2BSP_MarkLights (light, bit, node->children[1]);
return;
}
// mark the polygons
surf = cl.worldmodel->surfaces + node->firstsurface;
for (i=0 ; i<node->numsurfaces ; i++, surf++)
{
if (surf->dlightframe != r_dlightframecount)
{
surf->dlightbits = 0u;
surf->dlightframe = r_dlightframecount;
}
surf->dlightbits |= bit;
}
Q2BSP_MarkLights (light, bit, node->children[0]);
Q2BSP_MarkLights (light, bit, node->children[1]);
}
#ifndef SERVERONLY
static void GLR_Q2BSP_StainNode_r (model_t *model, mnode_t *node, float *parms)
{
mplane_t *splitplane;
float dist;
msurface_t *surf;
int i;
if (node->contents != -1)
return;
splitplane = node->plane;
dist = DotProduct ((parms+1), splitplane->normal) - splitplane->dist;
if (dist > (*parms))
{
GLR_Q2BSP_StainNode_r (model, node->children[0], parms);
return;
}
if (dist < (-*parms))
{
GLR_Q2BSP_StainNode_r (model, node->children[1], parms);
return;
}
// mark the polygons
surf = model->surfaces + node->firstsurface;
for (i=0 ; i<node->numsurfaces ; i++, surf++)
{
if (surf->flags&~(SURF_DONTWARP|SURF_PLANEBACK))
continue;
Surf_StainSurf(model, surf, parms);
}
GLR_Q2BSP_StainNode_r (model, node->children[0], parms);
GLR_Q2BSP_StainNode_r (model, node->children[1], parms);
}
static void GLR_Q2BSP_StainNode (model_t *model, float *parms)
{
GLR_Q2BSP_StainNode_r(model, model->rootnode, parms);
}
#endif
#endif
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
static void CM_BuildBIH(model_t *mod, int submodel)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
cmodel_t *sub = &prv->cmodels[submodel];
struct bihleaf_s *bihleaf, *l;
size_t bihleafs, i;
//undo any bih damage on the copy
mod->cnodes = NULL;
mod->funcs.NativeTrace = CM_NativeTrace;
mod->funcs.NativeContents = CM_NativeContents;
mod->funcs.PointContents = Q2BSP_PointContents;
if (!q3bsp_bihtraces.ival)
return; //skip this. fall back on other stuff.
if (mod->fromgame != fg_quake3)
return;
bihleafs = sub->num_brushes;
for (i = 0; i < sub->num_patches; i++)
bihleafs += prv->patches[sub->firstpatch + i].numfacets;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
for (i = 0; i < sub->num_cmeshes; i++)
bihleafs += prv->cmeshes[sub->firstcmesh + i].numincidies/3;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
bihleaf = l = BZ_Malloc(sizeof(*bihleaf)*bihleafs);
//now we have enough storage, spit them out providing bounds info.
for (i = 0; i < sub->num_brushes; i++)
{
q2cbrush_t *b = &prv->brushes[sub->firstbrush+i];
l->type = BIH_BRUSH;
l->data.brush = b;
l->data.contents = b->contents;
VectorCopy(b->absmins, l->mins);
VectorCopy(b->absmaxs, l->maxs);
l++;
}
#ifdef Q3BSPS
for (i = 0; i < sub->num_patches; i++)
{
q3cpatch_t *p = &prv->patches[sub->firstpatch+i];
size_t j;
for (j = 0; j < p->numfacets; j++)
{
q2cbrush_t *b = &p->facets[j];
l->type = BIH_PATCHBRUSH;
l->data.patchbrush = b;
l->data.contents = b->contents;
VectorCopy(b->absmins, l->mins);
VectorCopy(b->absmaxs, l->maxs);
l++;
}
}
#endif
#ifdef Q3BSPS
for (i = 0; i < sub->num_cmeshes; i++)
{
q3cmesh_t *m = &prv->cmeshes[sub->firstcmesh+i];
size_t j;
for (j = 0; j+2 < m->numincidies; j+=3)
{
index_t *v = m->indicies+j;
vec_t *v1 = m->xyz_array[v[0]], *v2 = m->xyz_array[v[1]], *v3 = m->xyz_array[v[2]];
l->type = BIH_TRIANGLE;
l->data.tri.xyz = m->xyz_array;
l->data.tri.indexes = v;
l->data.contents = m->surface->c.value;
VectorCopy(v1, l->mins);
VectorCopy(v1, l->maxs);
AddPointToBounds(v2, l->mins, l->maxs);
AddPointToBounds(v3, l->mins, l->maxs);
l++;
}
}
#endif
BIH_Build(mod, bihleaf, l-bihleaf);
BZ_Free(bihleaf);
}
/*
==================
CM_LoadMap
Loads in the map and all submodels
==================
*/
static cmodel_t *CM_LoadMap (model_t *mod, qbyte *filein, size_t filelen, qboolean clientload)
{
unsigned *buf;
int i;
q2dheader_t header;
int length;
qboolean noerrors = true;
model_t *wmod = mod;
char loadname[32];
qbyte *mod_base = (qbyte *)filein;
bspx_header_t *bspx = NULL;
unsigned int checksum;
#ifdef Q3BSPS
extern cvar_t gl_overbright;
#endif
#ifndef SERVERONLY
void (*buildmeshes)(model_t *mod, msurface_t *surf, builddata_t *cookie) = NULL;
qbyte *facedata = NULL;
unsigned int facesize = 0;
#endif
cminfo_t *prv;
COM_FileBase (mod->name, loadname, sizeof(loadname));
// free old stuff
mod->meshinfo = prv = ZG_Malloc(&mod->memgroup, sizeof(*prv));
prv->numcmodels = 0;
prv->numvisibility = 0;
mod->type = mod_brush;
if (!mod->name[0])
{
prv->cmodels = ZG_Malloc(&mod->memgroup, 1 * sizeof(*prv->cmodels));
mod->leafs = ZG_Malloc(&mod->memgroup, 1 * sizeof(*mod->leafs));
mod->funcs.AreasConnected = CM_AreasConnected;
prv->numcmodels = 1;
prv->numareas = 1;
mod->checksum = mod->checksum2 = 0;
prv->cmodels[0].headnode = (mnode_t*)mod->leafs; //directly start with the empty leaf
return &prv->cmodels[0]; // cinematic servers won't have anything at all
}
//
// load the file
//
buf = (unsigned *)filein;
length = filelen;
if (!buf)
{
Con_Printf (CON_ERROR "Couldn't load %s\n", mod->name);
return NULL;
}
2023-02-20 06:13:40 +00:00
checksum = LittleLong (CalcHashInt(&hash_md4, buf, length));
header = *(q2dheader_t *)(buf);
header.ident = LittleLong(header.ident);
header.version = LittleLong(header.version);
ClearBounds(mod->mins, mod->maxs);
switch(header.version)
{
default:
Con_Printf (CON_ERROR "Quake 2 or Quake 3 based BSP with unknown header (%s: %i should be %i or %i)\n"
, mod->name, header.version, BSPVERSION_Q2, BSPVERSION_Q3);
return NULL;
break;
#ifdef Q3BSPS
#ifdef RFBSPS
case BSPVERSION_RBSP: //rbsp/fbsp
#endif
case BSPVERSION_RTCW: //rtcw
case BSPVERSION_Q3:
#ifdef RFBSPS
if (header.ident == (('F'<<0)+('B'<<8)+('S'<<16)+('P'<<24)))
{
mod->lightmaps.width = 512;
mod->lightmaps.height = 512;
}
else
#endif
{
mod->lightmaps.width = 128;
mod->lightmaps.height = 128;
}
prv->mapisq3 = true;
mod->fromgame = fg_quake3;
for (i=0 ; i<Q3LUMPS_TOTAL ; i++)
{
#ifdef RFBSPS
if (i == RBSPLUMP_LIGHTINDEXES && header.version != BSPVERSION_RBSP)
{
header.lumps[i].filelen = 0;
header.lumps[i].fileofs = 0;
}
else
#endif
{
header.lumps[i].filelen = LittleLong (header.lumps[i].filelen);
header.lumps[i].fileofs = LittleLong (header.lumps[i].fileofs);
if (header.lumps[i].filelen && header.lumps[i].fileofs + header.lumps[i].filelen > filelen)
{
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
Con_Printf (CON_ERROR "WARNING: q3bsp %s truncated (lump %i, %i+%i > %u)\n", mod->name, i, header.lumps[i].fileofs, header.lumps[i].filelen, (unsigned int)filelen);
header.lumps[i].filelen = filelen - header.lumps[i].fileofs;
if (header.lumps[i].filelen < 0)
header.lumps[i].filelen = 0;
}
}
}
/*
#ifndef SERVERONLY
GLMod_LoadVertexes (mod, cmod_base, &header.lumps[Q3LUMP_DRAWVERTS]);
// GLMod_LoadEdges (mod, cmod_base, &header.lumps[Q3LUMP_EDGES]);
// GLMod_LoadSurfedges (mod, cmod_base, &header.lumps[Q3LUMP_SURFEDGES]);
GLMod_LoadLighting (mod, cmod_base, &header.lumps[Q3LUMP_LIGHTMAPS]);
#endif
CModQ3_LoadShaders (mod, cmod_base, &header.lumps[Q3LUMP_SHADERS]);
CModQ3_LoadPlanes (mod, cmod_base, &header.lumps[Q3LUMP_PLANES]);
CModQ3_LoadLeafBrushes (mod, cmod_base, &header.lumps[Q3LUMP_LEAFBRUSHES]);
CModQ3_LoadBrushes (mod, cmod_base, &header.lumps[Q3LUMP_BRUSHES]);
CModQ3_LoadBrushSides (mod, cmod_base, &header.lumps[Q3LUMP_BRUSHSIDES]);
#ifndef SERVERONLY
CMod_LoadTexInfo (mod, cmod_base, &header.lumps[Q3LUMP_SHADERS]);
CMod_LoadFaces (mod, cmod_base, &header.lumps[Q3LUMP_SURFACES]);
// GLMod_LoadMarksurfaces (mod, cmod_base, &header.lumps[Q3LUMP_LEAFFACES]);
#endif
CMod_LoadVisibility (mod, cmod_base, &header.lumps[Q3LUMP_VISIBILITY]);
CModQ3_LoadSubmodels (mod, cmod_base, &header.lumps[Q3LUMP_MODELS]);
CModQ3_LoadLeafs (mod, cmod_base, &header.lumps[Q3LUMP_LEAFS]);
CModQ3_LoadNodes (mod, cmod_base, &header.lumps[Q3LUMP_NODES]);
// CMod_LoadAreas (mod, cmod_base, &header.lumps[Q3LUMP_AREAS]);
// CMod_LoadAreaPortals (mod, cmod_base, &header.lumps[Q3LUMP_AREAPORTALS]);
CMod_LoadEntityString (mod, cmod_base, &header.lumps[Q3LUMP_ENTITIES]);
*/
prv->faces = NULL;
bspx = BSPX_Setup(mod, mod_base, filelen, header.lumps, Q3LUMPS_TOTAL);
//q3 maps have built in 4-fold overbright.
//if we're not rendering with that, we need to brighten the lightmaps in order to keep the darker parts the same brightness. we loose the 2 upper bits. those bright areas become uniform and indistinct.
//this is used for both the lightmap AND vertex lighting
//FIXME: when not using overbrights, we suffer a loss of precision.
gl_overbright.flags |= CVAR_RENDERERLATCH;
BuildLightMapGammaTable(1, (1<<(2-gl_overbright.ival)));
prv->mapisq3 = true;
noerrors = noerrors && CModQ3_LoadShaders (mod, mod_base, &header.lumps[Q3LUMP_SHADERS]);
noerrors = noerrors && CModQ3_LoadPlanes (mod, mod_base, &header.lumps[Q3LUMP_PLANES]);
#ifdef RFBSPS
if (header.version == BSPVERSION_RBSP)
{
noerrors = noerrors && CModRBSP_LoadBrushSides (mod, mod_base, &header.lumps[Q3LUMP_BRUSHSIDES]);
noerrors = noerrors && CModRBSP_LoadVertexes (mod, mod_base, &header.lumps[Q3LUMP_DRAWVERTS]);
}
else
#endif
{
noerrors = noerrors && CModQ3_LoadBrushSides (mod, mod_base, &header.lumps[Q3LUMP_BRUSHSIDES]);
noerrors = noerrors && CModQ3_LoadVertexes (mod, mod_base, &header.lumps[Q3LUMP_DRAWVERTS]);
}
noerrors = noerrors && CModQ3_LoadBrushes (mod, mod_base, &header.lumps[Q3LUMP_BRUSHES]);
noerrors = noerrors && CModQ3_LoadLeafBrushes (mod, mod_base, &header.lumps[Q3LUMP_LEAFBRUSHES]);
#ifdef RFBSPS
if (header.version == BSPVERSION_RBSP)
noerrors = noerrors && CModRBSP_LoadFaces (mod, mod_base, &header.lumps[Q3LUMP_SURFACES]);
else
#endif
noerrors = noerrors && CModQ3_LoadFaces (mod, mod_base, &header.lumps[Q3LUMP_SURFACES]);
if (noerrors)
Mod_LoadEntities (mod, mod_base, &header.lumps[Q3LUMP_ENTITIES]);
#ifndef SERVERONLY
if (qrenderer != QR_NONE)
{
#ifdef RFBSPS
if (header.version == BSPVERSION_RBSP)
noerrors = noerrors && CModRBSP_LoadLightgrid (mod, mod_base, &header.lumps[Q3LUMP_LIGHTGRID], &header.lumps[RBSPLUMP_LIGHTINDEXES]);
else
#endif
noerrors = noerrors && CModQ3_LoadLightgrid (mod, mod_base, &header.lumps[Q3LUMP_LIGHTGRID]);
noerrors = noerrors && CModQ3_LoadIndexes (mod, mod_base, &header.lumps[Q3LUMP_DRAWINDEXES]);
if (header.version != BSPVERSION_RTCW)
noerrors = noerrors && CModQ3_LoadFogs (mod, mod_base, &header.lumps[Q3LUMP_FOGS]);
else
mod->numfogs = 0;
facedata = (void *)(mod_base + header.lumps[Q3LUMP_SURFACES].fileofs);
#ifdef RFBSPS
if (header.version == BSPVERSION_RBSP)
{
noerrors = noerrors && CModRBSP_LoadRFaces (mod, mod_base, &header.lumps[Q3LUMP_SURFACES]);
buildmeshes = CModRBSP_BuildSurfMesh;
facesize = sizeof(rbspface_t);
mod->lightmaps.surfstyles = 4;
}
else
#endif
{
noerrors = noerrors && CModQ3_LoadRFaces (mod, mod_base, &header.lumps[Q3LUMP_SURFACES]);
buildmeshes = CModQ3_BuildSurfMesh;
facesize = sizeof(q3dface_t);
mod->lightmaps.surfstyles = 1;
}
if (noerrors)
{
i = header.lumps[Q3LUMP_LIGHTMAPS].filelen / (mod->lightmaps.width*mod->lightmaps.height*3);
mod->lightmaps.deluxemapping = !(i&1);
mod->lightmaps.count = max(mod->lightmaps.count, i);
mod->lightmaps.deluxemapping_modelspace = true; //we assume true for q3bsp.
for (i = 0; i < mod->numsurfaces && mod->lightmaps.deluxemapping; i++)
{
if (mod->surfaces[i].lightmaptexturenums[0] >= 0 && (mod->surfaces[i].lightmaptexturenums[0] & 1))
mod->lightmaps.deluxemapping = false;
}
{
char deluxeMaps[64], *key;
key = (char*)Mod_ParseWorldspawnKey(mod, "deluxeMaps", deluxeMaps, sizeof(deluxeMaps));
if (*key)
{
switch(atoi(key))
{
case 0:
mod->lightmaps.deluxemapping = false;
break;
case 1:
// mod->lightmaps.deluxemapping = true;
mod->lightmaps.deluxemapping_modelspace = true;
break;
case 2:
// mod->lightmaps.deluxemapping = true;
mod->lightmaps.deluxemapping_modelspace = false;
break;
}
}
}
}
if (noerrors)
CModQ3_LoadLighting (mod, mod_base, &header.lumps[Q3LUMP_LIGHTMAPS]); //fixme: duplicated loading.
}
#endif
noerrors = noerrors && CModQ3_LoadMarksurfaces (mod, mod_base, &header.lumps[Q3LUMP_LEAFSURFACES]);
noerrors = noerrors && CModQ3_LoadLeafs (mod, mod_base, &header.lumps[Q3LUMP_LEAFS]);
noerrors = noerrors && CModQ3_LoadNodes (mod, mod_base, &header.lumps[Q3LUMP_NODES]);
noerrors = noerrors && CModQ3_LoadSubmodels (mod, mod_base, &header.lumps[Q3LUMP_MODELS]);
noerrors = noerrors && CModQ3_LoadVisibility (mod, mod_base, &header.lumps[Q3LUMP_VISIBILITY]);
if (!noerrors)
{
if (prv->faces)
BZ_Free(prv->faces);
return NULL;
}
#ifdef HAVE_SERVER
mod->funcs.FatPVS = Q23BSP_FatPVS;
mod->funcs.EdictInFatPVS = Q23BSP_EdictInFatPVS;
mod->funcs.FindTouchedLeafs = Q23BSP_FindTouchedLeafs;
#endif
mod->funcs.ClusterPVS = CM_ClusterPVS;
mod->funcs.ClusterPHS = CM_ClusterPHS;
mod->funcs.ClusterForPoint = CM_PointCluster;
#ifdef HAVE_CLIENT
mod->funcs.LightPointValues = GLQ3_LightGrid;
mod->funcs.StainNode = GLR_Q2BSP_StainNode;
mod->funcs.MarkLights = Q2BSP_MarkLights;
mod->funcs.PrepareFrame = CM_PrepareFrame;
#ifdef RTLIGHTS
mod->funcs.GenerateShadowMesh = Q3BSP_GenerateShadowMesh;
#endif
#endif
mod->funcs.PointContents = Q2BSP_PointContents;
mod->funcs.NativeTrace = CM_NativeTrace;
mod->funcs.NativeContents = CM_NativeContents;
mod->funcs.InfoForPoint = CM_InfoForPoint;
mod->funcs.AreasConnected = CM_AreasConnected;
mod->funcs.SetAreaPortalState = CM_SetAreaPortalState;
mod->funcs.WriteAreaBits = CM_WriteAreaBits;
mod->funcs.LoadAreaPortalBlob = CM_LoadAreaPortalBlob;
mod->funcs.SaveAreaPortalBlob = CM_SaveAreaPortalBlob;
#ifdef HAVE_CLIENT
//light grid info
if (mod->lightgrid)
{
char gridsize[256], *key;
char val[64];
float maxs;
q3lightgridinfo_t *lg = mod->lightgrid;
key = (char*)Mod_ParseWorldspawnKey(mod, "gridsize", gridsize, sizeof(gridsize));
key = COM_ParseOut(key, val, sizeof(val));
lg->gridSize[0] = atof(val);
key = COM_ParseOut(key, val, sizeof(val));
lg->gridSize[1] = atof(val);
key = COM_ParseOut(key, val, sizeof(val));
lg->gridSize[2] = atof(val);
if ( lg->gridSize[0] < 1 || lg->gridSize[1] < 1 || lg->gridSize[2] < 1 )
{
lg->gridSize[0] = 64;
lg->gridSize[1] = 64;
lg->gridSize[2] = 128;
}
for ( i = 0; i < 3; i++ )
{
lg->gridMins[i] = lg->gridSize[i] * ceil( (prv->cmodels->mins[i] + 1) / lg->gridSize[i] );
maxs = lg->gridSize[i] * floor( (prv->cmodels->maxs[i] - 1) / lg->gridSize[i] );
lg->gridBounds[i] = (maxs - lg->gridMins[i])/lg->gridSize[i] + 1;
}
lg->gridBounds[3] = lg->gridBounds[1] * lg->gridBounds[0];
}
#endif
if (!CM_CreatePatchesForLeafs (mod, prv)) //for clipping
{
BZ_Free(prv->faces);
return NULL;
}
#ifdef HAVE_SERVER
CMQ3_CalcPHS(mod);
#endif
// BZ_Free(map_verts);
BZ_Free(prv->faces);
break;
#endif
#ifdef Q2BSPS
case BSPVERSION_Q2:
case BSPVERSION_Q2W:
mod->lightmaps.width = LMBLOCK_SIZE_MAX;
mod->lightmaps.height = LMBLOCK_SIZE_MAX;
prv->mapisq3 = false;
mod->engineflags |= MDLF_NEEDOVERBRIGHT;
for (i=0 ; i<Q2HEADER_LUMPS ; i++)
{
header.lumps[i].filelen = LittleLong (header.lumps[i].filelen);
header.lumps[i].fileofs = LittleLong (header.lumps[i].fileofs);
}
if (header.version == BSPVERSION_Q2W)
{
header.lumps[i].filelen = LittleLong (header.lumps[i].filelen);
header.lumps[i].fileofs = LittleLong (header.lumps[i].fileofs);
i++;
}
bspx = BSPX_Setup(mod, mod_base, filelen, header.lumps, i);
#if defined(HAVE_CLIENT) && defined(IMAGEFMT_PCX)
if (CM_GetQ2Palette())
memcpy(q2_palette, host_basepal, 768);
#endif
#ifdef HAVE_SERVER
mod->funcs.FatPVS = Q23BSP_FatPVS;
mod->funcs.EdictInFatPVS = Q23BSP_EdictInFatPVS;
mod->funcs.FindTouchedLeafs = Q23BSP_FindTouchedLeafs;
#endif
mod->funcs.LightPointValues = NULL;
mod->funcs.StainNode = NULL;
mod->funcs.MarkLights = NULL;
mod->funcs.ClusterPVS = CM_ClusterPVS;
mod->funcs.ClusterPHS = CM_ClusterPHS;
mod->funcs.ClusterForPoint = CM_PointCluster;
mod->funcs.PointContents = Q2BSP_PointContents;
mod->funcs.NativeTrace = CM_NativeTrace;
mod->funcs.NativeContents = CM_NativeContents;
mod->funcs.InfoForPoint = CM_InfoForPoint;
mod->funcs.AreasConnected = CM_AreasConnected;
mod->funcs.SetAreaPortalState = CM_SetAreaPortalState;
mod->funcs.WriteAreaBits = CM_WriteAreaBits;
mod->funcs.LoadAreaPortalBlob = CM_LoadAreaPortalBlob;
mod->funcs.SaveAreaPortalBlob = CM_SaveAreaPortalBlob;
mod->funcs.PrepareFrame = NULL;
switch(qrenderer)
{
case QR_NONE: //dedicated only
noerrors = noerrors && CModQ2_LoadSurfaces (mod, mod_base, &header.lumps[Q2LUMP_TEXINFO]);
noerrors = noerrors && CModQ2_LoadPlanes (mod, mod_base, &header.lumps[Q2LUMP_PLANES]);
noerrors = noerrors && CModQ2_LoadVisibility (mod, mod_base, &header.lumps[Q2LUMP_VISIBILITY]);
noerrors = noerrors && CModQ2_LoadBrushSides (mod, mod_base, &header.lumps[Q2LUMP_BRUSHSIDES]);
noerrors = noerrors && CModQ2_LoadBrushes (mod, mod_base, &header.lumps[Q2LUMP_BRUSHES]);
noerrors = noerrors && CModQ2_LoadLeafBrushes (mod, mod_base, &header.lumps[Q2LUMP_LEAFBRUSHES]);
noerrors = noerrors && CModQ2_LoadLeafs (mod, mod_base, &header.lumps[Q2LUMP_LEAFS]);
noerrors = noerrors && CModQ2_LoadNodes (mod, mod_base, &header.lumps[Q2LUMP_NODES]);
noerrors = noerrors && CModQ2_LoadSubmodels (mod, mod_base, &header.lumps[Q2LUMP_MODELS]);
noerrors = noerrors && CModQ2_LoadAreas (mod, mod_base, &header.lumps[Q2LUMP_AREAS]);
noerrors = noerrors && CModQ2_LoadAreaPortals (mod, mod_base, &header.lumps[Q2LUMP_AREAPORTALS]);
if (noerrors)
Mod_LoadEntities (mod, mod_base, &header.lumps[Q2LUMP_ENTITIES]);
break;
#ifdef HAVE_CLIENT
default:
{
// load into heap
noerrors = noerrors && Mod_LoadVertexes (mod, mod_base, &header.lumps[Q2LUMP_VERTEXES]);
noerrors = noerrors && Mod_LoadEdges (mod, mod_base, &header.lumps[Q2LUMP_EDGES], false);
noerrors = noerrors && Mod_LoadSurfedges (mod, mod_base, &header.lumps[Q2LUMP_SURFEDGES]);
noerrors = noerrors && CModQ2_LoadSurfaces (mod, mod_base, &header.lumps[Q2LUMP_TEXINFO]);
noerrors = noerrors && CModQ2_LoadPlanes (mod, mod_base, &header.lumps[Q2LUMP_PLANES]);
noerrors = noerrors && CModQ2_LoadTexInfo (mod, mod_base, &header.lumps[Q2LUMP_TEXINFO], loadname);
if (noerrors)
Mod_LoadEntities (mod, mod_base, &header.lumps[Q2LUMP_ENTITIES]);
noerrors = noerrors && CModQ2_LoadFaces (mod, mod_base, &header.lumps[Q2LUMP_FACES], &header.lumps[Q2LUMP_LIGHTING], header.version == BSPVERSION_Q2W, bspx);
Mod_LoadVertexNormals(mod, bspx, mod_base, (header.version == BSPVERSION_Q2W)?&header.lumps[19]:NULL);
noerrors = noerrors && Mod_LoadMarksurfaces (mod, mod_base, &header.lumps[Q2LUMP_LEAFFACES], false);
noerrors = noerrors && CModQ2_LoadVisibility (mod, mod_base, &header.lumps[Q2LUMP_VISIBILITY]);
noerrors = noerrors && CModQ2_LoadBrushSides (mod, mod_base, &header.lumps[Q2LUMP_BRUSHSIDES]);
noerrors = noerrors && CModQ2_LoadBrushes (mod, mod_base, &header.lumps[Q2LUMP_BRUSHES]);
noerrors = noerrors && CModQ2_LoadLeafBrushes (mod, mod_base, &header.lumps[Q2LUMP_LEAFBRUSHES]);
noerrors = noerrors && CModQ2_LoadLeafs (mod, mod_base, &header.lumps[Q2LUMP_LEAFS]);
noerrors = noerrors && CModQ2_LoadNodes (mod, mod_base, &header.lumps[Q2LUMP_NODES]);
noerrors = noerrors && CModQ2_LoadSubmodels (mod, mod_base, &header.lumps[Q2LUMP_MODELS]);
noerrors = noerrors && CModQ2_LoadAreas (mod, mod_base, &header.lumps[Q2LUMP_AREAS]);
noerrors = noerrors && CModQ2_LoadAreaPortals (mod, mod_base, &header.lumps[Q2LUMP_AREAPORTALS]);
if (!noerrors)
{
return NULL;
}
mod->funcs.LightPointValues = GLQ2BSP_LightPointValues;
mod->funcs.StainNode = GLR_Q2BSP_StainNode;
mod->funcs.MarkLights = Q2BSP_MarkLights;
mod->funcs.PrepareFrame = CM_PrepareFrame;
#ifdef RTLIGHTS
mod->funcs.GenerateShadowMesh = Q2BSP_GenerateShadowMesh;
#endif
}
break;
#endif
}
#endif
}
BSPX_LoadEnvmaps(mod, bspx, mod_base);
#ifdef Q3BSPS
{
int x, y;
for (x = 0; x < prv->numareas; x++)
for (y = 0; y < prv->numareas; y++)
prv->q3areas[x].numareaportals[y] = map_autoopenportals.ival;
}
#endif
#ifdef Q2BSPS
if (map_autoopenportals.value)
memset (prv->q2portalopen, 1, sizeof(prv->q2portalopen)); //open them all. Used for progs that havn't got a clue.
else
memset (prv->q2portalopen, 0, sizeof(prv->q2portalopen)); //make them start closed.
#endif
FloodAreaConnections (prv);
mod->checksum = mod->checksum2 = checksum;
mod->nummodelsurfaces = mod->numsurfaces;
memset(&mod->batches, 0, sizeof(mod->batches));
mod->vbos = NULL;
mod->numsubmodels = CM_NumInlineModels(mod);
mod->hulls[0].firstclipnode = prv->cmodels[0].headnode-mod->nodes;
mod->rootnode = prv->cmodels[0].headnode;
mod->nummodelsurfaces = prv->cmodels[0].numsurfaces;
#ifdef HAVE_CLIENT
prv->oldclusters[0] = prv->oldclusters[1] = -1;
if (qrenderer != QR_NONE)
{
builddata_t *bd = NULL;
if (buildmeshes)
{
bd = Z_Malloc(sizeof(*bd) + facesize*mod->nummodelsurfaces);
bd->buildfunc = buildmeshes;
memcpy(bd+1, facedata + mod->firstmodelsurface*facesize, facesize*mod->nummodelsurfaces);
}
COM_AddWork(WG_MAIN, ModBrush_LoadGLStuff, mod, bd, 0, 0);
}
#endif
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
//FIXME: q2bsp apparently doesn't report which brushes are part of which submodels.
//FIXME: ALL patches? not just worldmodel?
CM_BuildBIH(mod, 0);
for (i=1 ; i< mod->numsubmodels ; i++)
{
cmodel_t *bm;
char name[MAX_QPATH];
Q_snprintfz (name, sizeof(name), "*%i:%s", i, wmod->publicname);
mod = Mod_FindName (name);
*mod = *wmod;
mod->archive = NULL;
mod->entities_raw = NULL;
Reworked client support for DPP5+. less code now, its much more graceful. added waterfog command. waterfog overrides regular fog only when the view is in water. fixed 64bit printf format specifiers. should work better on winxp64. fixed some spec angle weirdness. fixed viewsize 99.99 weirdness with ezhud. fixed extra offset on the console (exhibited in 64bit builds, but not limited to). fixed .avi playback, can now actually display frames again. reimplemented line sparks. fixed r_editlights_save flipping the light's pitch. fixed issue with oggs failing to load. fixed condump to cope with unicode properly. made sv_bigcoords default except in quake. hexen2 kinda needs it for bsp angle precision. fixed nq server to not stall weirdly on map changes. fixed qwprogs svc_cdtrack not bugging out with nq clients on the server. fixed restart command to load the last map run by the server, instead of start.bsp (when idle) optimised d3d9 renderer a little. now uses less draw calls, especially with complex scenes. seems to get higher framerates than opengl now. fixed d3d9 renderer to not bug out quite so much when run fullscreen (shader subsystem is now correctly initialised). fixed a couple of bugs from font change. also now supports utf-8 in a few more places. r_editlights_reload no longer generates rtlights inside the void. this resolves a few glitches (but should also help framerates a little). fixed so corona-only lights won't generate shadowmaps and waste lots of time. removed lots of #defines from qclib. I should never have made them in the first place, but I was lazy. obviously there's more left that I cba to remove yet. fixed nested calls with variant-vectors. this fixes csaddon's light editor. fixed qcc hc calling conventions using redundant stores. disabled keywords can still be used by using __keyword instead. fixed ftegccgui grep feature. fixed motionless-dog qcc bug. tweaked qcc warnings a little. -Wall is now a viable setting. you should be able to fix all those warnings. fixed qw svc_intermission + dpp5+ clients bug. fixed annoying spam about disconnecting in hexen2. rewrote status command a little to cope with ipv6 addresses more gracefully fixed significant stall when hibernating/debugging a server with a player sitting on it. fixed truelightning. fixed rocketlight overriding pflags. fixed torches vanishing on vid_restart. fixed issue with decal scaling. fixed findentityfield builtin. fixed fteqcc issue with ptr+1 fixed use of arrays inside class functions. fixed/implemented fteqcc emulation of pointer opcodes. added __inout keyword to fteqcc, so that it doesn't feel so horrendous. fixed sizeof(*foo) fixed *struct = struct; fixed recursive structs. fixed fteqcc warning report. fixed sdl2 controller support, hopefully. attempted to implement xinput, including per-player audio playback. slightly fixed relaxed attitude to mouse focus when running fullscreen. fixed weird warnings/errors with 'ent.arrayhead' terms. now generates sane errors. implemented bindmaps (for csqc). fixed crashing bug with eprint builtin. implemented subset of music_playlist_* functionality. significant changes to music playback. fixed some more dpcsqc compat. fixed binds menu. now displays and accepts modifiers. fixed issues with huge lightmaps. fixed protocol determinism with dp clients connecting to fte servers. the initial getchallenge request now inhibits vanilla nq connection requests. implemented support for 'dupe' userinfo key, allowing clients to request client->server packet duplication. should probably queue them tbh. implemented sv_saveentfile command. fixed resume after breaking inside a stepped-over function. fixed erroneous footer after debugging. (I wonder just how many things I broke with these fixes) git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4946 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-07-26 10:56:18 +00:00
mod->submodelof = wmod;
Q_strncpyz(mod->publicname, name, sizeof(mod->publicname));
Q_snprintfz (mod->name, sizeof(mod->name), "*%i:%s", i, wmod->name);
memset(&mod->memgroup, 0, sizeof(mod->memgroup));
bm = CM_InlineModel (wmod, name);
mod->hulls[0].firstclipnode = -1; //no nodes,
if (bm->headleaf)
{
mod->leafs = bm->headleaf;
mod->nodes = NULL;
mod->hulls[0].firstclipnode = -1; //make it refer directly to the first leaf, for things that still use numbers.
mod->rootnode = (mnode_t*)bm->headleaf;
}
else
{
mod->leafs = wmod->leafs;
mod->nodes = wmod->nodes;
mod->hulls[0].firstclipnode = bm->headnode - mod->nodes; //determine the correct node index
mod->rootnode = bm->headnode;
}
mod->nummodelsurfaces = bm->numsurfaces;
mod->firstmodelsurface = bm->firstsurface;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
CM_BuildBIH(mod, i);
memset(&mod->batches, 0, sizeof(mod->batches));
mod->vbos = NULL;
VectorCopy (bm->maxs, mod->maxs);
VectorCopy (bm->mins, mod->mins);
#ifndef SERVERONLY
mod->radius = RadiusFromBounds (mod->mins, mod->maxs);
if (qrenderer != QR_NONE)
{
builddata_t *bd = NULL;
if (buildmeshes)
{
bd = Z_Malloc(sizeof(*bd) + facesize*mod->nummodelsurfaces);
bd->buildfunc = buildmeshes;
memcpy(bd+1, facedata + mod->firstmodelsurface*facesize, facesize*mod->nummodelsurfaces);
}
COM_AddWork(WG_MAIN, ModBrush_LoadGLStuff, mod, bd, i, 0);
}
#endif
COM_AddWork(WG_MAIN, Mod_ModelLoaded, mod, NULL, MLS_LOADED, 0);
}
#ifdef TERRAIN
wmod->terrain = Mod_LoadTerrainInfo(wmod, loadname, false);
#endif
return &prv->cmodels[0];
}
/*
==================
CM_InlineModel
==================
*/
static cmodel_t *CM_InlineModel (model_t *model, char *name)
{
cminfo_t *prv = (cminfo_t*)model->meshinfo;
int num;
if (!name)
Host_Error("Bad model\n");
else if (name[0] != '*')
Host_Error("Bad model\n");
num = atoi (name+1);
if (num < 1 || num >= prv->numcmodels)
Host_Error ("CM_InlineModel: bad number");
return &prv->cmodels[num];
}
static int CM_NumInlineModels (model_t *model)
{
cminfo_t *prv = (cminfo_t*)model->meshinfo;
return prv->numcmodels;
}
static int CM_LeafContents (model_t *model, int leafnum)
{
if (leafnum < 0 || leafnum >= model->numleafs)
Host_Error ("CM_LeafContents: bad number");
return model->leafs[leafnum].contents;
}
static int CM_LeafCluster (model_t *model, int leafnum)
{
if (leafnum < 0 || leafnum >= model->numleafs)
Host_Error ("CM_LeafCluster: bad number");
return model->leafs[leafnum].cluster;
}
static int CM_LeafArea (model_t *model, int leafnum)
{
if (leafnum < 0 || leafnum >= model->numleafs)
Host_Error ("CM_LeafArea: bad number");
return model->leafs[leafnum].area;
}
//=======================================================================
#define PlaneDiff(point,plane) (((plane)->type < 3 ? (point)[(plane)->type] : DotProduct((point), (plane)->normal)) - (plane)->dist)
static mplane_t box_planes[6];
static model_t box_model;
static q2cbrush_t box_brush;
static q2cbrushside_t box_sides[6];
static qboolean BM_NativeTrace(model_t *model, int forcehullnum, const framestate_t *framestate, const vec3_t axis[3], const vec3_t start, const vec3_t end, const vec3_t mins, const vec3_t maxs, qboolean capsule, unsigned int contents, trace_t *trace);
static unsigned int BM_NativeContents(struct model_s *model, int hulloverride, const framestate_t *framestate, const vec3_t axis[3], const vec3_t p, const vec3_t mins, const vec3_t maxs)
{
unsigned int j;
q2cbrushside_t *brushside = box_sides;
for ( j = 0; j < 6; j++, brushside++ )
{
if ( PlaneDiff (p, brushside->plane) > 0 )
return 0;
}
return FTECONTENTS_BODY;
}
/*
===================
CM_InitBoxHull
Set up the planes and nodes so that the six floats of a bounding box
can just be stored out and get a proper clipping hull structure.
===================
*/
static void CM_InitBoxHull (void)
{
int i;
mplane_t *p;
q2cbrushside_t *s;
box_model.funcs.NativeContents = BM_NativeContents;
box_model.funcs.NativeTrace = BM_NativeTrace;
box_model.loadstate = MLS_LOADED;
box_brush.contents = FTECONTENTS_BODY;
box_brush.numsides = 6;
box_brush.brushside = box_sides;
for (i=0 ; i<6 ; i++)
{
//the pointers
s = &box_sides[i];
p = &box_planes[i];
// brush sides
s->plane = p;
s->surface = &nullsurface;
// planes
p->type = ((i>=3)?i-3:i);
p->signbits = 0;
VectorClear (p->normal);
p->normal[p->type] = ((i>=3)?-1:1);
}
}
/*
===================
CM_HeadnodeForBox
To keep everything totally uniform, bounding boxes are turned into small
BSP trees instead of being compared directly.
===================
*/
static void CM_SetTempboxSize (const vec3_t mins, const vec3_t maxs)
{
box_planes[0].dist = maxs[0];
box_planes[1].dist = maxs[1];
box_planes[2].dist = maxs[2];
box_planes[3].dist = -mins[0];
box_planes[4].dist = -mins[1];
box_planes[5].dist = -mins[2];
}
model_t *CM_TempBoxModel(const vec3_t mins, const vec3_t maxs)
{
CM_SetTempboxSize(mins, maxs);
return &box_model;
}
/*
==================
CM_PointLeafnum_r
==================
*/
static int CM_PointLeafnum_r (model_t *mod, const vec3_t p, int num)
{
float d;
mnode_t *node;
mplane_t *plane;
while (num >= 0)
{
node = mod->nodes + num;
plane = node->plane;
if (plane->type < 3)
d = p[plane->type] - plane->dist;
else
d = DotProduct (plane->normal, p) - plane->dist;
if (d < 0)
num = node->childnum[1];
else
num = node->childnum[0];
}
return -1 - num;
}
2023-03-27 17:22:00 +00:00
#ifdef HAVE_SERVER
static int CM_PointLeafnum (model_t *mod, const vec3_t p)
{
if (!mod || mod->loadstate != MLS_LOADED)
return 0; // sound may call this without map loaded
return CM_PointLeafnum_r (mod, p, 0);
}
2023-03-27 17:22:00 +00:00
#endif
static int CM_PointCluster (model_t *mod, const vec3_t p, int *area)
{
int leaf;
if (!mod || mod->loadstate != MLS_LOADED)
return 0; // sound may call this without map loaded
leaf = CM_PointLeafnum_r (mod, p, 0);
if (area)
*area = CM_LeafArea(mod, leaf);
return CM_LeafCluster(mod, leaf);
}
static int CM_PointContents (model_t *mod, const vec3_t p);
static void CM_InfoForPoint (struct model_s *mod, vec3_t pos, int *area, int *cluster, unsigned int *contentbits)
{
int leaf = CM_PointLeafnum_r (mod, pos, 0);
*area = CM_LeafArea(mod, leaf);
*cluster = CM_LeafCluster(mod, leaf);
*contentbits = CM_LeafContents(mod, leaf);
//q3 needs to use brush contents (its leafs no longer need to strictly follow brushes)
if (mod->fromgame != fg_quake2)
*contentbits = CM_PointContents (mod, pos);
}
/*
=============
CM_BoxLeafnums
Fills in a list of all the leafs touched
=============
*/
static int leaf_count, leaf_maxcount;
static int *leaf_list;
static const float *leaf_mins, *leaf_maxs;
static int leaf_topnode;
static void CM_BoxLeafnums_r (model_t *mod, int nodenum)
{
mplane_t *plane;
mnode_t *node;
int s;
while (1)
{
if (nodenum < 0)
{
if (leaf_count >= leaf_maxcount)
{
// Com_Printf ("CM_BoxLeafnums_r: overflow\n");
return;
}
leaf_list[leaf_count++] = -1 - nodenum;
return;
}
node = &mod->nodes[nodenum];
plane = node->plane;
// s = BoxOnPlaneSide (leaf_mins, leaf_maxs, plane);
s = BOX_ON_PLANE_SIDE(leaf_mins, leaf_maxs, plane);
if (s == 1)
nodenum = node->childnum[0];
else if (s == 2)
nodenum = node->childnum[1];
else
{ // go down both
if (leaf_topnode == -1)
leaf_topnode = nodenum;
CM_BoxLeafnums_r (mod, node->childnum[0]);
nodenum = node->childnum[1];
}
}
}
static int CM_BoxLeafnums_headnode (model_t *mod, const vec3_t mins, const vec3_t maxs, int *list, int listsize, int headnode, int *topnode)
{
leaf_list = list;
leaf_count = 0;
leaf_maxcount = listsize;
leaf_mins = mins;
leaf_maxs = maxs;
leaf_topnode = -1;
CM_BoxLeafnums_r (mod, headnode);
if (topnode)
*topnode = leaf_topnode;
return leaf_count;
}
static int CM_BoxLeafnums (model_t *mod, const vec3_t mins, const vec3_t maxs, int *list, int listsize, int *topnode)
{
return CM_BoxLeafnums_headnode (mod, mins, maxs, list,
listsize, mod->hulls[0].firstclipnode, topnode);
}
/*
==================
CM_PointContents
==================
*/
static int CM_PointContents (model_t *mod, const vec3_t p)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i, j, contents;
mleaf_t *leaf;
q2cbrush_t *brush;
q2cbrushside_t *brushside;
if (!mod) // map not loaded
return 0;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
i = CM_PointLeafnum_r (mod, p, mod->hulls[0].firstclipnode);
if (mod->fromgame == fg_quake2)
contents = mod->leafs[i].contents; //q2 is simple.
else
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
leaf = &mod->leafs[i];
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
// if ( leaf->contents & CONTENTS_NODROP ) {
// contents = CONTENTS_NODROP;
// } else {
contents = 0;
// }
for (i = 0; i < leaf->numleafbrushes; i++)
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
brush = prv->leafbrushes[leaf->firstleafbrush + i];
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
// check if brush actually adds something to contents
if ( (contents & brush->contents) == brush->contents ) {
continue;
}
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
brushside = brush->brushside;
for ( j = 0; j < brush->numsides; j++, brushside++ )
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
if ( PlaneDiff (p, brushside->plane) > 0 )
break;
}
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
if (j == brush->numsides)
contents |= brush->contents;
}
}
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
#ifdef TERRAIN
if (mod->terrain)
contents |= Heightmap_PointContents(mod, NULL, p);
#endif
return contents;
}
static unsigned int CM_NativeContents(struct model_s *model, int hulloverride, const framestate_t *framestate, const vec3_t axis[3], const vec3_t p, const vec3_t mins, const vec3_t maxs)
{
cminfo_t *prv = (cminfo_t*)model->meshinfo;
int contents;
if (!DotProduct(mins, mins) && !DotProduct(maxs, maxs))
return CM_PointContents(model, p);
if (!model) // map not loaded
return 0;
{
int i, j, k;
mleaf_t *leaf;
q2cbrush_t *brush;
q2cbrushside_t *brushside;
vec3_t absmin, absmax;
int leaflist[64];
k = CM_BoxLeafnums (model, absmin, absmax, leaflist, 64, NULL);
contents = 0;
for (k--; k >= 0; k--)
{
leaf = &model->leafs[leaflist[k]];
if (model->fromgame != fg_quake2)
{ //q3 is more complex
for (i = 0; i < leaf->numleafbrushes; i++)
{
brush = prv->leafbrushes[leaf->firstleafbrush + i];
// check if brush actually adds something to contents
if ( (contents & brush->contents) == brush->contents ) {
continue;
}
brushside = brush->brushside;
for ( j = 0; j < brush->numsides; j++, brushside++ )
{
if ( PlaneDiff (p, brushside->plane) > 0 )
break;
}
if (j == brush->numsides)
contents |= brush->contents;
}
}
else //q2 is simple
contents |= leaf->contents;
}
}
return contents;
}
/*
===============================================================================
BOX TRACING
===============================================================================
*/
// 1/32 epsilon to keep floating point happy
#define DIST_EPSILON (0.03125)
static vec3_t trace_start, trace_end;
static vec3_t trace_mins, trace_maxs;
static vec3_t trace_extents;
static vec3_t trace_absmins, trace_absmaxs;
static vec3_t trace_up; //capsule points upwards in this direction
static vec3_t trace_capsulesize; //radius, up, down
static float trace_truefraction;
static float trace_nearfraction;
static trace_t trace_trace;
static int trace_contents;
static enum
{
shape_isbox,
shape_iscapsule,
shape_ispoint
} trace_shape; // optimized case
static void CM_FinalizeBrush(q2cbrush_t *brush)
{
vecV_t verts[256];
vec4_t planes[256];
int i, j;
ClearBounds(brush->absmins, brush->absmaxs);
for (i = 0; i < brush->numsides; i++)
{
VectorCopy(brush->brushside[i].plane->normal, planes[i]);
planes[i][3] = brush->brushside[i].plane->dist;
}
for (i = 0; i < brush->numsides; i++)
{
Too many changes, sorry. Change revision displays, use the SVN commit date instead of using __DATE__ (when there's no local changes). This should allow reproducible builds. Added s_al_disable cvar, to block openal and all the various problems people have had with it, without having to name an explicit fallback (which would vary by system). Add mastervolume cvar (for ss). Add r_shadows 2 (aka fake shadows - for ss). Add scr_loadingscreen_aspect -1 setting, to disable levelshots entirely, also disables the progress bar (for ss). Better support for some effectinfo hacks (for ss). Added dpcompat_nocsqcwarnings (because of lazy+buggy mods like ss). Rework the dpcsqc versions of project+unproject builtins for better compat (for ss). Added dpcompat_csqcinputeventtypes to block unexpected csqc input events (for ss). Better compat with DP's loadfont console command (for ss). Added dpcompat_smallerfonts cvar to replicate a DP bug (for ss). Detect dp's m_draw extension, to work around it (for ss). Cvar dpcompat_ignoremodificationtimes added. A value of 0 favour the most recently modified file, 1 will use DP-like alphabetically sorted preferences (for ss). loadfont builtin can now accept outline=1 in the sizes arg for slightly more readable fonts. Fix bbox calcs for rotated entities, fix needed for r_ignorenetpvs 0. Hackily parse emoji.json to provide :poop: etc suggestions. Skip prediction entirely when there's no local entity info. This fixes stair-smoothing in xonotic. screenshot_cubemap will now capture half-float images when saving to ktx or dds files. Fix support for xcf files larger than 4gb, mostly to avoid compiler warnings. Fixed size of gfx/loading.lmp when replacement textures are used. Added mipmap support for rg8 and l8a8 textures. r_hdr_framebuffer cvar updated to support format names instead of random negative numbers. Description updated to name some interesting ones. Perform autoupdate _checks_ ONLY with explicit user confirmation (actual updating already needed user confirmation, but this extra step should reduce the chances of us getting wrongly accused of exfiltrating user data if we're run in a sandbox - we ONLY ever included the updating engine's version in the checks, though there's nothing we can do to avoid sending the user's router's IP). Removed the 'summon satan all over your harddrive' quit message, in case paranoid security researchers are idiots and don't bother doing actual research. Removed the triptohell.info and fte.triptohell.info certificates, they really need to stop being self-signed. The updates domain is still self-signed for autoupdates. Video drivers are now able to report supported video resolutions, visible to menuqc. Currently only works with SDL2 builds. Added setmousepos builtin. Should work with glx+win32 build. VF_SKYROOM_CAMERA can now accept an extra two args, setviewprop(VF_SKYROOM_CAMERA, org, axis, degrees). Removed v_skyroom_origin+v_skyroom_orientation cvars in favour just v_skyroom, which should make it behave more like the 'fog' command (used when csqc isn't overriding). Added R_EndPolygonRibbon builtin to make it faster+easier to generate textured ribbon/cable/etc wide lines (for TW). sdl: Fix up sys_sdl.c's file enumeration to support wildcards in directories. edit command now displays end1.bin/end2.bin correctly, because we can. Finally add support for f_modified - though ruleset_allow_larger_models and ruleset_allow_overlong_sounds generally make it redundant. Fix threading race condition in sha1 lookups. Updated f_ruleset to include the same extra flags reported by ezquake. A mod's default.fmf file can now contain an eg 'mainconfig config.cfg' line (to explicitly set the main config saved with cfg_save_auto 1 etc). fmf: basegame steam:GameName/GameDir can be used to try to load a mod directory from an installed steam game. The resulting gamedir will be read-only. HOMEDIR CHANGE: use homedirs only if the basedir cannot be written or a homedir already exists, which should further reduce the probability of microsoft randomly uploading our data to their cloud (but mostly because its annoying to never know where your data is written). Fixed buf_cvarlist, should work in xonotic now, and without segfaults. Added an extra arg to URI_Get_Callback calls - the response size, also changed the tempstring to contain all bytes of the response, you need to be careful about nulls though. Try to work around nvidia's forced-panning bug on x11 when changing video modes. This might screw with other programs. sdl: support custom icons. sdl: support choosing a specific display. Added some documentation to menuqc builtins. menusys: use outlines for slightly more readable fonts. menusys: switch vid_width and vid_height combos into a single video mode combo to set both according to reported video modes. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5581 fc73d0e0-1445-4013-8a0c-d673dee63da5
2019-11-20 03:09:50 +00:00
//most brushes are axial, which can save some a little loadtime
if (planes[i][0] == 1)
brush->absmaxs[0] = planes[i][3];
else if (planes[i][1] == 1)
brush->absmaxs[1] = planes[i][3];
else if (planes[i][2] == 1)
brush->absmaxs[2] = planes[i][3];
else if (planes[i][0] == -1)
brush->absmins[0] = -planes[i][3];
else if (planes[i][1] == -1)
brush->absmins[1] = -planes[i][3];
else if (planes[i][2] == -1)
brush->absmins[2] = -planes[i][3];
else
{
j = Fragment_ClipPlaneToBrush(verts, countof(verts), planes, sizeof(planes[0]), brush->numsides, planes[i]);
while (j-- > 0)
AddPointToBounds(verts[j], brush->absmins, brush->absmaxs);
}
}
}
/*
================
CM_ClipBoxToBrush
================
*/
static void CM_ClipBoxToBrush (vec3_t mins, vec3_t maxs, vec3_t p1, vec3_t p2,
trace_t *trace, q2cbrush_t *brush)
{
int i, j;
mplane_t *plane, *clipplane;
float dist;
float enterfrac, leavefrac;
vec3_t ofs;
float d1, d2;
qboolean getout, startout;
float f;
q2cbrushside_t *side, *leadside;
float nearfrac=0;
enterfrac = -1;
leavefrac = 2;
clipplane = NULL;
if (!brush->numsides)
return;
getout = false;
startout = false;
leadside = NULL;
for (i=0 ; i<brush->numsides ; i++)
{
side = brush->brushside+i;
plane = side->plane;
switch(trace_shape)
{
default:
case shape_isbox: // general box case
// push the plane out apropriately for mins/maxs
// FIXME: use signbits into 8 way lookup for each mins/maxs
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = maxs[j];
else
ofs[j] = mins[j];
}
dist = DotProduct (ofs, plane->normal);
dist = plane->dist - dist;
break;
capsuledist(dist,plane,mins,maxs)
case shape_ispoint: // special point case
dist = plane->dist;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
d2 = DotProduct (p2, plane->normal) - dist;
if (d2 > 0)
getout = true; // endpoint is not in solid
if (d1 > 0)
startout = true;
// if completely in front of face, no intersection
if (d1 > 0 && d2 >= d1)
return;
if (d1 <= 0 && d2 <= 0)
continue;
// crosses face
if (d1 > d2)
{ // enter
f = (d1) / (d1-d2);
if (f > enterfrac)
{
enterfrac = f;
nearfrac = (d1-DIST_EPSILON) / (d1-d2);
clipplane = plane;
leadside = side;
}
}
else
{ // leave
f = (d1) / (d1-d2);
if (f < leavefrac)
leavefrac = f;
}
}
if (!startout)
{ // original point was inside brush
trace->startsolid = true;
if (!getout)
trace->allsolid = true;
return;
}
if (enterfrac <= leavefrac)
{
if (enterfrac > -1 && enterfrac <= trace_truefraction)
{
if (enterfrac < 0)
enterfrac = 0;
trace_nearfraction = nearfrac;
trace_truefraction = enterfrac;
trace->plane.dist = clipplane->dist;
VectorCopy(clipplane->normal, trace->plane.normal);
trace->surface = &(leadside->surface->c);
trace->contents = brush->contents;
}
}
}
#ifdef Q3BSPS
static void CM_ClipBoxToPlanes (vec3_t trmins, vec3_t trmaxs, vec3_t p1, vec3_t p2, trace_t *trace, vec3_t plmins, vec3_t plmaxs, mplane_t *plane, int numplanes, q2csurface_t *surf)
{
int i, j;
mplane_t *clipplane;
float dist;
float enterfrac, leavefrac;
vec3_t ofs;
float d1, d2;
qboolean getout, startout;
float f;
// q2cbrushside_t *side, *leadside;
static mplane_t bboxplanes[6] = //we change the dist, but nothing else
{
{{1, 0, 0}},
{{0, 1, 0}},
{{0, 0, 1}},
{{-1, 0, 0}},
{{0, -1, 0}},
{{0, 0, -1}},
};
float nearfrac=0;
enterfrac = -1;
leavefrac = 2;
clipplane = NULL;
getout = false;
startout = false;
// leadside = NULL;
for (i=0 ; i<numplanes ; i++, plane++)
{
switch(trace_shape)
{
default:
case shape_isbox: // general box case
// push the plane out apropriately for mins/maxs
// FIXME: special case for axial
// FIXME: use signbits into 8 way lookup for each mins/maxs
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = trmaxs[j];
else
ofs[j] = trmins[j];
}
dist = DotProduct (ofs, plane->normal);
dist = plane->dist - dist;
break;
capsuledist(dist,plane,trmins,trmaxs)
case shape_ispoint: // special point case
dist = plane->dist;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
d2 = DotProduct (p2, plane->normal) - dist;
if (d2 > 0)
getout = true; // endpoint is not in solid
if (d1 > 0)
startout = true;
// if completely in front of face, no intersection
if (d1 > 0 && d2 >= d1)
return;
if (d1 <= 0 && d2 <= 0)
continue;
// crosses face
if (d1 > d2)
{ // enter
f = (d1) / (d1-d2);
if (f > enterfrac)
{
enterfrac = f;
nearfrac = (d1-DIST_EPSILON) / (d1-d2);
clipplane = plane;
// leadside = side;
}
}
else
{ // leave
f = (d1) / (d1-d2);
if (f < leavefrac)
leavefrac = f;
}
}
//bevel the brush axially (to match the player's bbox), in case that wasn't already done
for (i=0, plane = bboxplanes; i<6 ; i++, plane++)
{
if (i < 3)
{ //positive normal
dist = trmins[i];
plane->dist = plmaxs[i];
dist = plane->dist - dist;
d1 = p1[i] - dist;
d2 = p2[i] - dist;
}
else
{ //negative normal
j = i-3;
dist = -trmaxs[j];
plane->dist = -plmins[j];
dist = plane->dist - dist;
d1 = -p1[j] - dist;
d2 = -p2[j] - dist;
}
if (d2 > 0)
getout = true; // endpoint is not in solid
if (d1 > 0)
startout = true;
// if completely in front of face, no intersection
if (d1 > 0 && d2 >= d1)
return;
if (d1 <= 0 && d2 <= 0)
continue;
// crosses face
if (d1 > d2)
{ // enter
f = (d1) / (d1-d2);
if (f > enterfrac)
{
enterfrac = f;
nearfrac = (d1-DIST_EPSILON) / (d1-d2);
clipplane = plane;
// leadside = side;
}
}
else
{ // leave
f = (d1) / (d1-d2);
if (f < leavefrac)
leavefrac = f;
}
}
if (!startout)
{ // original point was inside brush
trace->startsolid = true;
if (!getout)
trace->allsolid = true;
return;
}
if (enterfrac <= leavefrac)
{
if (enterfrac > -1 && enterfrac <= trace_truefraction)
{
if (enterfrac < 0)
enterfrac = 0;
trace_nearfraction = nearfrac;
trace_truefraction = enterfrac;
trace->plane.dist = clipplane->dist;
VectorCopy(clipplane->normal, trace->plane.normal);
trace->surface = surf;
trace->contents = surf->value;
}
}
}
static void Mod_Trace_Trisoup_(vecV_t *posedata, index_t *indexes, size_t numindexes, vec3_t start, vec3_t end, vec3_t mins, vec3_t maxs, trace_t *trace, q2csurface_t *surf)
{
size_t i;
int j;
float *p1, *p2, *p3;
vec3_t edge1, edge2, edge3;
mplane_t planes[5];
vec3_t tmins, tmaxs;
for (i = 0; i < numindexes; i+=3)
{
p1 = posedata[indexes[i+0]];
p2 = posedata[indexes[i+1]];
p3 = posedata[indexes[i+2]];
//determine the triangle extents, and skip the triangle if we're completely out of bounds
for (j = 0; j < 3; j++)
{
tmins[j] = p1[j];
if (tmins[j] > p2[j])
tmins[j] = p2[j];
if (tmins[j] > p3[j])
tmins[j] = p3[j];
if (trace_absmaxs[j]+(1/8.f) < tmins[j])
break;
tmaxs[j] = p1[j];
if (tmaxs[j] < p2[j])
tmaxs[j] = p2[j];
if (tmaxs[j] < p3[j])
tmaxs[j] = p3[j];
if (trace_absmins[j]-(1/8.f) > tmaxs[j])
break;
}
//skip any triangles which are completely outside the trace bounds
if (j < 3)
continue;
VectorSubtract(p1, p2, edge1);
VectorSubtract(p3, p2, edge2);
VectorSubtract(p1, p3, edge3);
CrossProduct(edge1, edge2, planes[0].normal);
VectorNormalize(planes[0].normal);
planes[0].dist = DotProduct(p1, planes[0].normal);
VectorNegate(planes[0].normal, planes[1].normal);
planes[1].dist = -planes[0].dist + 4;
//determine edges
//FIXME: use adjacency info
CrossProduct(edge1, planes[0].normal, planes[2].normal);
VectorNormalize(planes[2].normal);
planes[2].dist = DotProduct(p2, planes[2].normal);
CrossProduct(planes[0].normal, edge2, planes[3].normal);
VectorNormalize(planes[3].normal);
planes[3].dist = DotProduct(p3, planes[3].normal);
CrossProduct(planes[0].normal, edge3, planes[4].normal);
VectorNormalize(planes[4].normal);
planes[4].dist = DotProduct(p1, planes[4].normal);
CM_ClipBoxToPlanes(mins, maxs, start, end, trace, tmins, tmaxs, planes, 5, surf);
}
}
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
/*
static void CM_ClipBoxToMesh (vec3_t mins, vec3_t maxs, vec3_t p1, vec3_t p2, trace_t *trace, mesh_t *mesh)
{
trace_truefraction = trace->truefraction;
trace_nearfraction = trace->fraction;
Mod_Trace_Trisoup_(mesh->xyz_array, mesh->indexes, mesh->numindexes, p1, p2, mins, maxs, trace, &nullsurface.c);
trace->truefraction = trace_truefraction;
trace->fraction = trace_nearfraction;
}
fix colormod added frag message filter, and dedicated frag tracker. added 'windowed consoles' for social-type stuff without depending upon csqc mods for it. added in_deviceids command which allows listing/renumbering device ids. slider widgets now support inverted ranges, so gamma selection isn't so weird. fix top/bottom colour selection bug. software banding feature is now part of the 'software' preset (now that it supports mipmaps). support for loading .maps, and editing their brushes etc (with appropriate qc mod). 'map mymap.map' to use. expect problems with missing wads and replacement textures overriding them and messing up texture scales. snd_inactive is now default. fix threading issue with wavs, no more error from 0-sample-but-otherwise-valid wavs. added -makeinstaller option to embed a manifest inside the exe (and icon). the resulting program will insist on installing the game if its run from outside a valid basedir. framegroup support for q1mdl. textures are now loaded on multiple worker threads, for reduced load times. moo har har. netgraph shows packet+byte rates too. added r_lightstylescale, pretty similar to contrast, but doesn't impose any framerate cost, but may have overbrighting issues. r_softwarebanding now works on q2bsp too. fixed crepuscular lights. gzip transfer encoding is performed while downloading, instead of inducing stalls. FINALLY fix ezquake download compat issue (dimman found the issue). git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@4851 fc73d0e0-1445-4013-8a0c-d673dee63da5
2015-04-14 23:12:17 +00:00
*/
static void CM_ClipBoxToPatch (vec3_t mins, vec3_t maxs, vec3_t p1, vec3_t p2,
trace_t *trace, q2cbrush_t *brush)
{
int i, j;
mplane_t *plane, *clipplane;
float enterfrac, leavefrac, nearfrac = 0;
vec3_t ofs;
float d1, d2;
float dist;
qboolean startout;
float f;
q2cbrushside_t *side, *leadside;
if (!brush->numsides)
return;
enterfrac = -1;
leavefrac = 2;
clipplane = NULL;
startout = false;
leadside = NULL;
for (i=0 ; i<brush->numsides ; i++)
{
side = brush->brushside+i;
plane = side->plane;
// push the plane out apropriately for mins/maxs
switch(trace_shape)
{
default:
case shape_isbox: // general box case
// FIXME: use signbits into 8 way lookup for each mins/maxs
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = maxs[j];
else
ofs[j] = mins[j];
}
dist = DotProduct (ofs, plane->normal);
dist = plane->dist - dist;
break;
capsuledist(dist,plane,mins,maxs)
case shape_ispoint: // special point case
dist = plane->dist;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
d2 = DotProduct (p2, plane->normal) - dist;
// if completely in front of face, no intersection
if (d1 > 0 && d2 >= d1)
return;
if (d1 > 0)
startout = true;
if (d1 <= 0 && d2 <= 0)
continue;
// crosses face
if (d1 > d2)
{ // enter
f = (d1) / (d1-d2);
if (f > enterfrac)
{
enterfrac = f;
nearfrac = (d1-DIST_EPSILON) / (d1-d2);
clipplane = plane;
leadside = side;
}
}
else
{ // leave
f = (d1) / (d1-d2);
if (f < leavefrac)
leavefrac = f;
}
}
if (!startout)
{
trace->startsolid = true;
return; // original point is inside the patch
}
if (nearfrac <= leavefrac)
{
if (leadside && leadside->surface
&& enterfrac <= trace_truefraction)
{
if (enterfrac < 0)
enterfrac = 0;
trace_truefraction = enterfrac;
trace_nearfraction = nearfrac;
trace->plane.dist = clipplane->dist;
VectorCopy(clipplane->normal, trace->plane.normal);
trace->surface = &leadside->surface->c;
trace->contents = brush->contents;
}
else if (enterfrac < trace_truefraction)
leavefrac=0;
}
}
#endif
/*
================
CM_TestBoxInBrush
================
*/
static void CM_TestBoxInBrush (vec3_t mins, vec3_t maxs, vec3_t p1,
trace_t *trace, q2cbrush_t *brush)
{
int i, j;
mplane_t *plane;
float dist;
vec3_t ofs;
float d1;
q2cbrushside_t *side;
if (!brush->numsides)
return;
for (i=0 ; i<brush->numsides ; i++)
{
side = brush->brushside+i;
plane = side->plane;
switch(trace_shape)
{
default:
case shape_isbox: // general box case
// push the plane out apropriately for mins/maxs
// FIXME: use signbits into 8 way lookup for each mins/maxs
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = maxs[j];
else
ofs[j] = mins[j];
}
dist = DotProduct (ofs, plane->normal);
dist = plane->dist - dist;
break;
capsuledist(dist,plane,mins,maxs)
case shape_ispoint:
dist = plane->dist;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
// if completely in front of face, no intersection
if (d1 > 0)
return;
}
// inside this brush
trace->startsolid = trace->allsolid = true;
trace->contents |= brush->contents;
}
#ifdef Q3BSPS
static void CM_TestBoxInPatch (vec3_t mins, vec3_t maxs, vec3_t p1,
trace_t *trace, q2cbrush_t *brush)
{
int i, j;
mplane_t *plane;
vec3_t ofs, ofs2;
float dist, thickness;
float d1;
q2cbrushside_t *side;
if (!brush->numsides)
return;
i = 0; //front plane
{
side = brush->brushside+i;
plane = side->plane;
switch(trace_shape)
{
default:
case shape_isbox:
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = maxs[j], ofs2[j] = mins[j];
else
ofs[j] = mins[j], ofs2[j] = maxs[j];
}
dist = DotProduct (ofs, plane->normal);
thickness = DotProduct (ofs2, plane->normal)-dist;
dist = plane->dist - dist;
break;
case shape_iscapsule:
dist = DotProduct(trace_up, plane->normal);
thickness = dist*(trace_capsulesize[(dist<0)?2:1]) + trace_capsulesize[0]*2;
dist = dist*(trace_capsulesize[(dist<0)?1:2]) - trace_capsulesize[0];
dist = plane->dist - dist;
break;
case shape_ispoint:
dist = plane->dist;
thickness = 0;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
// if completely in front of face, no intersection
if (d1 > 0)
return;
//point is behind the front plane, so no real intersection.
if (thickness < 0.25)
thickness = 0.25; //FIXME: patches should probably be infinitely thin, but that makes stuff messy.
if (d1 < -thickness)
return;
}
for (i=1 ; i<brush->numsides ; i++)
{
side = brush->brushside+i;
plane = side->plane;
switch(trace_shape)
{
default:
case shape_isbox:
// general box case
// push the plane out apropriately for mins/maxs
// FIXME: use signbits into 8 way lookup for each mins/maxs
for (j=0 ; j<3 ; j++)
{
if (plane->normal[j] < 0)
ofs[j] = maxs[j];
else
ofs[j] = mins[j];
}
dist = DotProduct (ofs, plane->normal);
dist = plane->dist - dist;
break;
capsuledist(dist,plane,mins,maxs)
case shape_ispoint:
dist = plane->dist;
break;
}
d1 = DotProduct (p1, plane->normal) - dist;
// if completely in front of face, no intersection
if (d1 > 0)
return;
}
// inside this patch
trace->startsolid = trace->allsolid = true;
trace->contents = brush->contents;
}
#endif
/*
================
CM_TraceToLeaf
================
*/
static void CM_TraceToLeaf (cminfo_t *prv, mleaf_t *leaf)
{
int k;
q2cbrush_t *b;
#ifdef Q3BSPS
int patchnum, j;
q3cpatch_t *patch;
q3cmesh_t *cmesh;
#endif
if ( !(leaf->contents & trace_contents))
return;
// trace line against all brushes in the leaf
for (k=0 ; k<leaf->numleafbrushes ; k++)
{
b = prv->leafbrushes[leaf->firstleafbrush+k];
if (b->checkcount == checkcount)
continue; // already checked this brush in another leaf
b->checkcount = checkcount;
if ( !(b->contents & trace_contents))
continue;
if (!BoundsIntersect(b->absmins, b->absmaxs, trace_absmins, trace_absmaxs))
continue;
CM_ClipBoxToBrush (trace_mins, trace_maxs, trace_start, trace_end, &trace_trace, b);
if (trace_nearfraction <= 0)
return;
}
#ifdef Q3BSPS
if (!prv->mapisq3 || map_noCurves.value)
return;
// trace line against all patches in the leaf
for (k = 0; k < leaf->numleafpatches; k++)
{
patchnum = prv->leafpatches[leaf->firstleafpatch+k];
patch = &prv->patches[patchnum];
if (patch->checkcount == checkcount)
continue; // already checked this patch in another leaf
patch->checkcount = checkcount;
if ( !(patch->surface->c.value & trace_contents) )
continue;
if ( !BoundsIntersect(patch->absmins, patch->absmaxs, trace_absmins, trace_absmaxs) )
continue;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
for (j = 0; j < patch->numfacets; j++)
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
CM_ClipBoxToPatch (trace_mins, trace_maxs, trace_start, trace_end, &trace_trace, &patch->facets[j]);
if (trace_nearfraction<=0)
return;
}
}
for (k = 0; k < leaf->numleafcmeshes; k++)
{
patchnum = prv->leafcmeshes[leaf->firstleafcmesh+k];
cmesh = &prv->cmeshes[patchnum];
if (cmesh->checkcount == checkcount)
continue; // already checked this patch in another leaf
cmesh->checkcount = checkcount;
if ( !(cmesh->surface->c.value & trace_contents) )
continue;
if ( !BoundsIntersect(cmesh->absmins, cmesh->absmaxs, trace_absmins, trace_absmaxs) )
continue;
Mod_Trace_Trisoup_(cmesh->xyz_array, cmesh->indicies, cmesh->numincidies, trace_start, trace_end, trace_mins, trace_maxs, &trace_trace, &cmesh->surface->c);
if (trace_nearfraction<=0)
return;
}
#endif
}
/*
================
CM_TestInLeaf
================
*/
static void CM_TestInLeaf (cminfo_t *prv, mleaf_t *leaf)
{
int k;
q2cbrush_t *b;
#ifdef Q3BSPS
int patchnum, j;
q3cmesh_t *cmesh;
q3cpatch_t *patch;
#endif
if ( !(leaf->contents & trace_contents))
return;
// trace line against all brushes in the leaf
for (k=0 ; k<leaf->numleafbrushes ; k++)
{
b = prv->leafbrushes[leaf->firstleafbrush+k];
if (b->checkcount == checkcount)
continue; // already checked this brush in another leaf
b->checkcount = checkcount;
if (!(b->contents & trace_contents))
continue;
if (!BoundsIntersect(b->absmins, b->absmaxs, trace_absmins, trace_absmaxs))
continue;
CM_TestBoxInBrush (trace_mins, trace_maxs, trace_start, &trace_trace, b);
if (!trace_trace.fraction)
return;
}
#ifdef Q3BSPS
if (!prv->mapisq3 || map_noCurves.value)
return;
// trace line against all patches in the leaf
for (k = 0; k < leaf->numleafpatches; k++)
{
patchnum = prv->leafpatches[leaf->firstleafpatch+k];
patch = &prv->patches[patchnum];
if (patch->checkcount == checkcount)
continue; // already checked this patch in another leaf
patch->checkcount = checkcount;
if ( !(patch->surface->c.value & trace_contents) )
continue;
if ( !BoundsIntersect(patch->absmins, patch->absmaxs, trace_absmins, trace_absmaxs) )
continue;
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
for (j = 0; j < patch->numfacets; j++)
{
openxr plugin: tweaked - inputs should be working properly now, and are visible to csqc. subject to further breaking changes, however. _pext_vrinputs: added cvar to enable vr inputs protocol extension allowing vr inputs to be networked to ssqc too. defaults to 0 for now, will be renamed when deemed final. updates menu: the prompt to enable sources is now more explicit instead of expecting the user to have a clue. updates menu: added a v3 sources format, which should be more maintainable. not final. updates menu: try to give reasons why sources might be failing (to help blame ISPs if they try fucking over TTH dns again). presets menu: no longer closes the instant a preset is chosen. some presets have a couple of modifiers listed. force the demo loop in the background to serve as a preview. prompts menus: now does word wrapping. ftemaster: support importing server lists from other master servers (requested by Eukara). server: try to detect when non-reply inbound packets are blocked by firewalls/nats/etc (using ftemaster to do so). qcvm: added pointcontentsmask builtin, allowing it to probe more than just world, with fte's full contentbit range instead of just q1 legacy. qcvm: memfill8 builtin now works on createbuffer() pointers. qcvm: add missing unsigned ops. Fixed double comparison ops. fixed bug with op_store_i64. added missing OP_LOADP_I64 qcc: added '#pragma framerate RATE' for overriding implicit nextthink durations. qcc: fixed '#pragma DONT_COMPILE_THIS_FILE' to not screw up comments. qcc: added __GITURL__ __GITHASH__ __GITDATE__ __GITDATETIME__ __GITDESC__ for any mods that might want to make use of that. qcc: fix up -Fhashonly a little setrenderer: support for vulkan gpu enumeration. rulesets: reworked to support custom rulesets (using hashes to catch haxxors, though still nothing prevents just changing the client to ignore rulesets) bspx: use our BIH code for the bspx BRUSHLIST lump instead of the older less efficient code. (static)iqm+obj: these model formats can now be used for the worldmodel (with a suitable .ent file). Also using BIH for much better collision performance. pmove: tried to optimise PM_NudgePosition, should boost fps in stress tests. wayland: fix a crash on startup. mousegrabs now works better. imagetool: uses sdl for previews. git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5813 fc73d0e0-1445-4013-8a0c-d673dee63da5
2021-04-14 05:21:04 +00:00
CM_TestBoxInPatch (trace_mins, trace_maxs, trace_start, &trace_trace, &patch->facets[j]);
if (!trace_trace.fraction)
return;
}
}
for (k = 0; k < leaf->numleafcmeshes; k++)
{
patchnum = prv->leafcmeshes[leaf->firstleafcmesh+k];
cmesh = &prv->cmeshes[patchnum];
if (cmesh->checkcount == checkcount)
continue; // already checked this patch in another leaf
cmesh->checkcount = checkcount;
if ( !(cmesh->surface->c.value & trace_contents) )
continue;
if ( !BoundsIntersect(cmesh->absmins, cmesh->absmaxs, trace_absmins, trace_absmaxs) )
continue;
Mod_Trace_Trisoup_(cmesh->xyz_array, cmesh->indicies, cmesh->numincidies, trace_start, trace_end, trace_mins, trace_maxs, &trace_trace, &cmesh->surface->c);
if (trace_nearfraction<=0)
return;
}
#endif
}
/*
==================
CM_RecursiveHullCheck
==================
*/
static void CM_RecursiveHullCheck (model_t *mod, int num, float p1f, float p2f, vec3_t p1, vec3_t p2)
{
mnode_t *node;
mplane_t *plane;
float t1, t2, offset;
float frac, frac2;
float idist;
int i;
vec3_t mid;
int side;
float midf;
if (trace_truefraction <= p1f)
return; // already hit something nearer
// if < 0, we are in a leaf node
if (num < 0)
{
CM_TraceToLeaf (mod->meshinfo, &mod->leafs[-1-num]);
return;
}
//
// find the point distances to the seperating plane
// and the offset for the size of the box
//
node = mod->nodes + num;
plane = node->plane;
if (plane->type < 3)
{
t1 = p1[plane->type] - plane->dist;
t2 = p2[plane->type] - plane->dist;
offset = trace_extents[plane->type];
}
else
{
t1 = DotProduct (plane->normal, p1) - plane->dist;
t2 = DotProduct (plane->normal, p2) - plane->dist;
if (trace_shape == shape_ispoint)
offset = 0;
else
offset = fabs(trace_extents[0]*plane->normal[0]) +
fabs(trace_extents[1]*plane->normal[1]) +
fabs(trace_extents[2]*plane->normal[2]);
}
#if 0
CM_RecursiveHullCheck (node->childnum[0], p1f, p2f, p1, p2);
CM_RecursiveHullCheck (node->childnum[1], p1f, p2f, p1, p2);
return;
#endif
// see which sides we need to consider
if (t1 >= offset && t2 >= offset)
{
CM_RecursiveHullCheck (mod, node->childnum[0], p1f, p2f, p1, p2);
return;
}
if (t1 < -offset && t2 < -offset)
{
CM_RecursiveHullCheck (mod, node->childnum[1], p1f, p2f, p1, p2);
return;
}
// put the crosspoint DIST_EPSILON pixels on the near side
if (t1 < t2)
{
idist = 1.0/(t1-t2);
side = 1;
frac2 = (t1 + offset + DIST_EPSILON)*idist;
frac = (t1 - offset + DIST_EPSILON)*idist;
}
else if (t1 > t2)
{
idist = 1.0/(t1-t2);
side = 0;
frac2 = (t1 - offset - DIST_EPSILON)*idist;
frac = (t1 + offset + DIST_EPSILON)*idist;
}
else
{
side = 0;
frac = 1;
frac2 = 0;
}
// move up to the node
if (frac < 0)
frac = 0;
if (frac > 1)
frac = 1;
midf = p1f + (p2f - p1f)*frac;
for (i=0 ; i<3 ; i++)
mid[i] = p1[i] + frac*(p2[i] - p1[i]);
CM_RecursiveHullCheck (mod, node->childnum[side], p1f, midf, p1, mid);
// go past the node
if (frac2 < 0)
frac2 = 0;
if (frac2 > 1)
frac2 = 1;
midf = p1f + (p2f - p1f)*frac2;
for (i=0 ; i<3 ; i++)
mid[i] = p1[i] + frac2*(p2[i] - p1[i]);
CM_RecursiveHullCheck (mod, node->childnum[side^1], midf, p2f, mid, p2);
}
//======================================================================
/*
==================
CM_BoxTrace
==================
*/
static trace_t CM_BoxTrace (model_t *mod, const vec3_t start, const vec3_t end,
const vec3_t mins, const vec3_t maxs, qboolean capsule,
int brushmask)
{
int i;
vec3_t point;
checkcount++; // for multi-check avoidance
// fill in a default trace
memset (&trace_trace, 0, sizeof(trace_trace));
trace_truefraction = 1;
trace_nearfraction = 1;
trace_trace.fraction = 1;
trace_trace.truefraction = 1;
trace_trace.surface = &(nullsurface.c);
if (!mod) // map not loaded
return trace_trace;
trace_contents = brushmask;
VectorCopy (start, trace_start);
VectorCopy (end, trace_end);
VectorCopy (mins, trace_mins);
VectorCopy (maxs, trace_maxs);
if (1)
{
VectorAdd(trace_maxs, trace_mins, point);
VectorScale(point, 0.5, point);
VectorAdd(trace_start, point, trace_start);
VectorAdd(trace_end, point, trace_end);
VectorSubtract(trace_mins, point, trace_mins);
VectorSubtract(trace_maxs, point, trace_maxs);
}
// build a bounding box of the entire move (for patches)
ClearBounds (trace_absmins, trace_absmaxs);
//determine the type of trace that we're going to use, and the max extents
if (trace_mins[0] == 0 && trace_mins[1] == 0 && trace_mins[2] == 0 && trace_maxs[0] == 0 && trace_maxs[1] == 0 && trace_maxs[2] == 0)
{
trace_shape = shape_ispoint;
VectorSet (trace_extents, 1/32.0, 1/32.0, 1/32.0);
//acedemic
AddPointToBounds (trace_start, trace_absmins, trace_absmaxs);
AddPointToBounds (trace_end, trace_absmins, trace_absmaxs);
}
else if (capsule)
{
float ext;
trace_shape = shape_iscapsule;
//determine the capsule sizes
trace_capsulesize[0] = ((trace_maxs[0]-trace_mins[0]) + (trace_maxs[1]-trace_mins[1]))/4.0;
trace_capsulesize[1] = trace_maxs[2];
trace_capsulesize[2] = trace_mins[2];
//make sure the mins_z/maxs_z isn't screwed.
// if (trace_capsulesize[1]-trace_capsulesize[2] < trace_capsulesize[0])
// trace_capsulesize[1] = trace_capsulesize[0]+trace_capsulesize[2];
ext = (trace_capsulesize[1] > -trace_capsulesize[2])?trace_capsulesize[1]:-trace_capsulesize[2];
trace_capsulesize[1] -= trace_capsulesize[0];
trace_capsulesize[2] += trace_capsulesize[0];
trace_extents[0] = ext+1;
trace_extents[1] = ext+1;
trace_extents[2] = ext+1;
//determine the total range
VectorSubtract (trace_start, trace_extents, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorAdd (trace_start, trace_extents, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorSubtract (trace_end, trace_extents, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorAdd (trace_end, trace_extents, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
}
else
{
VectorAdd (trace_start, trace_mins, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorAdd (trace_start, trace_maxs, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorAdd (trace_end, trace_mins, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
VectorAdd (trace_end, trace_maxs, point);
AddPointToBounds (point, trace_absmins, trace_absmaxs);
trace_shape = shape_isbox;
trace_extents[0] = ((-trace_mins[0] > trace_maxs[0]) ? -trace_mins[0] : trace_maxs[0])+1;
trace_extents[1] = ((-trace_mins[1] > trace_maxs[1]) ? -trace_mins[1] : trace_maxs[1])+1;
trace_extents[2] = ((-trace_mins[2] > trace_maxs[2]) ? -trace_mins[2] : trace_maxs[2])+1;
}
trace_absmins[0] -= 1.0;
trace_absmins[1] -= 1.0;
trace_absmins[2] -= 1.0;
trace_absmaxs[0] += 1.0;
trace_absmaxs[1] += 1.0;
trace_absmaxs[2] += 1.0;
#if 0
if (0)
{ //treat *ALL* tests against the actual geometry instead of using any brushes.
//also ignores the bsp etc. not fast. testing only.
trace_ispoint = trace_mins[0] == 0 && trace_mins[1] == 0 && trace_mins[2] == 0
&& trace_maxs[0] == 0 && trace_maxs[1] == 0 && trace_maxs[2] == 0;
for (i = 0; i < mod->numsurfaces; i++)
{
CM_ClipBoxToMesh(trace_mins, trace_maxs, trace_start, trace_end, &trace_trace, mod->surfaces[i].mesh);
}
}
else
if (0)
{
trace_ispoint = trace_mins[0] == 0 && trace_mins[1] == 0 && trace_mins[2] == 0
&& trace_maxs[0] == 0 && trace_maxs[1] == 0 && trace_maxs[2] == 0;
for (i = 0; i < mod->numleafs; i++)
CM_TraceToLeaf(&mod->leafs[i]);
}
else
#endif
//
// check for position test special case
//
if (start[0] == end[0] && start[1] == end[1] && start[2] == end[2])
{
int leafs[1024];
int i, numleafs;
int topnode;
numleafs = CM_BoxLeafnums_headnode (mod, trace_absmins, trace_absmaxs, leafs, sizeof(leafs)/sizeof(leafs[0]), mod->hulls[0].firstclipnode, &topnode);
for (i=0 ; i<numleafs ; i++)
{
CM_TestInLeaf (mod->meshinfo, &mod->leafs[leafs[i]]);
if (trace_trace.allsolid)
break;
}
VectorCopy (start, trace_trace.endpos);
return trace_trace;
}
//
// general aabb trace
//
else
{
CM_RecursiveHullCheck (mod, mod->hulls[0].firstclipnode, 0, 1, trace_start, trace_end);
}
if (trace_nearfraction == 1)
{
trace_trace.fraction = 1;
VectorCopy (end, trace_trace.endpos);
}
else
{
if (trace_nearfraction<0)
trace_nearfraction=0;
trace_trace.fraction = trace_nearfraction;
for (i=0 ; i<3 ; i++)
trace_trace.endpos[i] = start[i] + trace_trace.fraction * (end[i] - start[i]);
}
return trace_trace;
}
static qboolean BM_NativeTrace(model_t *model, int forcehullnum, const framestate_t *framestate, const vec3_t axis[3], const vec3_t start, const vec3_t end, const vec3_t mins, const vec3_t maxs, qboolean capsule, unsigned int contents, trace_t *trace)
{
int i;
memset (trace, 0, sizeof(*trace));
trace_truefraction = 1;
trace_nearfraction = 1;
trace->fraction = 1;
trace->truefraction = 1;
trace->surface = &(nullsurface.c);
if (contents & FTECONTENTS_BODY)
{
trace_contents = contents;
VectorCopy (start, trace_start);
VectorCopy (end, trace_end);
VectorCopy (mins, trace_mins);
VectorCopy (maxs, trace_maxs);
if (trace_mins[0] == 0 && trace_mins[1] == 0 && trace_mins[2] == 0 && trace_maxs[0] == 0 && trace_maxs[1] == 0 && trace_maxs[2] == 0)
trace_shape = shape_ispoint;
else if (capsule)
trace_shape = shape_iscapsule;
else
trace_shape = shape_isbox;
CM_ClipBoxToBrush (trace_mins, trace_maxs, trace_start, trace_end, trace, &box_brush);
}
if (trace_nearfraction == 1)
{
trace->fraction = 1;
VectorCopy (trace_end, trace->endpos);
}
else
{
if (trace_nearfraction<0)
trace_nearfraction=0;
trace->fraction = trace_nearfraction;
trace->truefraction = trace_truefraction;
for (i=0 ; i<3 ; i++)
trace->endpos[i] = trace_start[i] + trace->fraction * (trace_end[i] - trace_start[i]);
}
return trace->fraction != 1;
}
static qboolean CM_NativeTrace(model_t *model, int forcehullnum, const framestate_t *framestate, const vec3_t axis[3], const vec3_t start, const vec3_t end, const vec3_t mins, const vec3_t maxs, qboolean capsule, unsigned int contents, trace_t *trace)
{
if (axis)
{
vec3_t start_l;
vec3_t end_l;
start_l[0] = DotProduct(start, axis[0]);
start_l[1] = DotProduct(start, axis[1]);
start_l[2] = DotProduct(start, axis[2]);
end_l[0] = DotProduct(end, axis[0]);
end_l[1] = DotProduct(end, axis[1]);
end_l[2] = DotProduct(end, axis[2]);
VectorSet(trace_up, axis[0][2], -axis[1][2], axis[2][2]);
*trace = CM_BoxTrace(model, start_l, end_l, mins, maxs, capsule, contents);
#ifdef TERRAIN
if (model->terrain)
{
trace_t hmt;
Heightmap_Trace(model, forcehullnum, framestate, NULL, start, end, mins, maxs, capsule, contents, &hmt);
if (hmt.fraction < trace->fraction)
*trace = hmt;
}
#endif
if (trace->fraction == 1)
{
VectorCopy (end, trace->endpos);
}
else
{
vec3_t iaxis[3];
vec3_t norm;
Matrix3x3_RM_Invert_Simple((void *)axis, iaxis);
VectorCopy(trace->plane.normal, norm);
trace->plane.normal[0] = DotProduct(norm, iaxis[0]);
trace->plane.normal[1] = DotProduct(norm, iaxis[1]);
trace->plane.normal[2] = DotProduct(norm, iaxis[2]);
/*just interpolate it, its easier than inverse matrix rotations*/
VectorInterpolate(start, trace->fraction, end, trace->endpos);
}
}
else
{
VectorSet(trace_up, 0, 0, 1);
*trace = CM_BoxTrace(model, start, end, mins, maxs, capsule, contents);
#ifdef TERRAIN
if (model->terrain)
{
trace_t hmt;
Heightmap_Trace(model, forcehullnum, framestate, NULL, start, end, mins, maxs, capsule, contents, &hmt);
if (hmt.fraction < trace->fraction)
*trace = hmt;
}
#endif
}
return trace->fraction != 1;
}
/*
===============================================================================
PVS / PHS
===============================================================================
*/
/*
===================
CM_DecompressVis
===================
*/
/*
qbyte *Mod_Q2DecompressVis (qbyte *in, model_t *model)
{
static qbyte decompressed[MAX_MAP_LEAFS/8];
int c;
qbyte *out;
int row;
row = (model->vis->numclusters+7)>>3;
out = decompressed;
if (!in)
{ // no vis info, so make all visible
while (row)
{
*out++ = 0xff;
row--;
}
return decompressed;
}
do
{
if (*in)
{
*out++ = *in++;
continue;
}
c = in[1];
in += 2;
while (c)
{
*out++ = 0;
c--;
}
} while (out - decompressed < row);
return decompressed;
}
#define DVIS_PVS 0
#define DVIS_PHS 1
qbyte *Mod_ClusterPVS (int cluster, model_t *model)
{
if (cluster == -1 || !model->vis)
return mod_novis;
return Mod_Q2DecompressVis ( (qbyte *)model->vis + model->vis->bitofs[cluster][DVIS_PVS],
model);
}
*/
static void CM_DecompressVis (model_t *mod, qbyte *in, qbyte *out, qboolean merge)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int c;
qbyte *out_p;
int row;
row = (mod->numclusters+7)>>3;
out_p = out;
if (!in || !prv->numvisibility)
{ // no vis info, so make all visible
while (row)
{
*out_p++ = 0xff;
row--;
}
return;
}
if (merge)
{
do
{
if (*in)
{
*out_p++ |= *in++;
continue;
}
out_p += in[1];
in += 2;
} while (out_p - out < row);
}
else
{
do
{
if (*in)
{
*out_p++ = *in++;
continue;
}
c = in[1];
in += 2;
if ((out_p - out) + c > row)
{
c = row - (out_p - out);
Con_DPrintf ("warning: Vis decompression overrun\n");
}
while (c)
{
*out_p++ = 0;
c--;
}
} while (out_p - out < row);
}
}
static pvsbuffer_t pvsrow;
static pvsbuffer_t phsrow;
static qbyte *CM_ClusterPVS (model_t *mod, int cluster, pvsbuffer_t *buffer, pvsmerge_t merge)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
if (!buffer)
buffer = &pvsrow;
if (buffer->buffersize < mod->pvsbytes)
buffer->buffer = BZ_Realloc(buffer->buffer, buffer->buffersize=mod->pvsbytes);
if (mod->fromgame == fg_quake2)
{
if (cluster == -1)
memset (buffer->buffer, 0, (mod->numclusters+7)>>3);
else
CM_DecompressVis (mod, ((qbyte*)prv->q2vis) + prv->q2vis->bitofs[cluster][DVIS_PVS], buffer->buffer, merge==PVM_MERGE);
return buffer->buffer;
}
else
{
if (cluster != -1 && prv->q3pvs->numclusters)
{
if (merge == PVM_FAST)
return (qbyte *)prv->q3pvs->data + cluster * prv->q3pvs->rowsize;
else if (merge == PVM_REPLACE)
memcpy(buffer->buffer, prv->q3pvs->data + cluster * prv->q3pvs->rowsize, mod->pvsbytes);
else
{
int c;
char *in = prv->q3pvs->data + cluster * prv->q3pvs->rowsize;
for (c = 0; c < mod->pvsbytes; c+=4)
*(int*)&buffer->buffer[c] |= *(int*)&in[c];
}
}
else
{
if (merge != PVM_MERGE)
memset (buffer->buffer, 0, (mod->numclusters+7)>>3);
}
return buffer->buffer;
}
}
static qbyte *CM_ClusterPHS (model_t *mod, int cluster, pvsbuffer_t *buffer)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
if (!buffer)
buffer = &phsrow;
if (buffer->buffersize < mod->pvsbytes)
buffer->buffer = BZ_Realloc(buffer->buffer, buffer->buffersize=mod->pvsbytes);
#ifdef Q3BSPS
if (mod->fromgame != fg_quake2)
{
if (cluster != -1 && prv->q3phs->numclusters)
{
if (prv->phscalced && !(prv->phscalced[cluster>>3] & (1<<(cluster&7))))
CalcClusterPHS(prv, cluster);
return (qbyte *)prv->q3phs->data + cluster * prv->q3phs->rowsize;
}
else
{
memset (buffer->buffer, 0, (mod->numclusters+7)>>3);
return buffer->buffer;
}
}
#endif
if (cluster == -1)
memset (buffer->buffer, 0, (mod->numclusters+7)>>3);
else
CM_DecompressVis (mod, ((qbyte*)prv->q2vis) + prv->q2vis->bitofs[cluster][DVIS_PHS], buffer->buffer, false);
return buffer->buffer;
}
2023-03-27 17:22:00 +00:00
#ifdef HAVE_SERVER
static unsigned int SV_Q2BSP_FatPVS (model_t *mod, const vec3_t org, pvsbuffer_t *result, qboolean merge)
{
int leafs[64];
int i, j, count;
vec3_t mins, maxs;
for (i=0 ; i<3 ; i++)
{
mins[i] = org[i] - 8;
maxs[i] = org[i] + 8;
}
count = CM_BoxLeafnums (mod, mins, maxs, leafs, countof(leafs), NULL);
if (count < 1)
Sys_Error ("SV_Q2FatPVS: count < 1");
// convert leafs to clusters
for (i=0 ; i<count ; i++)
leafs[i] = CM_LeafCluster(mod, leafs[i]);
//grow the buffer if needed
if (result->buffersize < mod->pvsbytes)
result->buffer = BZ_Realloc(result->buffer, result->buffersize=mod->pvsbytes);
if (count == 1 && leafs[0] == -1)
{ //if the only leaf is the outside then broadcast it.
memset(result->buffer, 0xff, mod->pvsbytes);
i = count;
}
else
{
i = 0;
if (!merge)
CM_ClusterPVS(mod, leafs[i++], result, PVM_REPLACE);
// or in all the other leaf bits
for ( ; i<count ; i++)
{
for (j=0 ; j<i ; j++)
if (leafs[i] == leafs[j])
break;
if (j != i)
continue; // already have the cluster we want
CM_ClusterPVS(mod, leafs[i], result, PVM_MERGE);
}
}
return mod->pvsbytes;
}
static int clientarea;
static unsigned int Q23BSP_FatPVS(model_t *mod, const vec3_t org, pvsbuffer_t *fte_restrict buffer, qboolean merge)
{//fixme: this doesn't add areas
int leafnum;
leafnum = CM_PointLeafnum (mod, org);
clientarea = CM_LeafArea (mod, leafnum);
return SV_Q2BSP_FatPVS (mod, org, buffer, merge);
}
static qboolean Q23BSP_EdictInFatPVS(model_t *mod, const pvscache_t *ent, const qbyte *pvs, const int *areas)
{
int i,l;
int nullarea = (mod->fromgame == fg_quake2)?0:-1;
if (areas)
{
for (i = 1; ; i++)
{
if (i > areas[0])
return false; //none of the camera's areas could see the entity
if (areas[i] == ent->areanum)
{
if (areas[i] != nullarea)
break;
//else entity is fully outside the world, invisible to all...
}
else if (CM_AreasConnected (mod, areas[i], ent->areanum))
break;
// doors can legally straddle two areas, so
// we may need to check another one
else if (ent->areanum2 != nullarea && CM_AreasConnected (mod, areas[i], ent->areanum2))
break;
}
}
if (ent->num_leafs == -1)
{ // too many leafs for individual check, go by headnode
if (!CM_HeadnodeVisible (mod, ent->headnode, pvs))
return false;
}
else
{ // check individual leafs
for (i=0 ; i < ent->num_leafs ; i++)
{
l = ent->leafnums[i];
if (pvs[l >> 3] & (1 << (l&7) ))
break;
}
if (i == ent->num_leafs)
return false; // not visible
}
return true;
}
static void Q23BSP_FindTouchedLeafs(model_t *model, struct pvscache_s *ent, const float *mins, const float *maxs)
{
#define MAX_TOTAL_ENT_LEAFS 128
int leafs[MAX_TOTAL_ENT_LEAFS];
int clusters[MAX_TOTAL_ENT_LEAFS];
int num_leafs;
int topnode;
int i, j;
int area;
int nullarea = (model->fromgame == fg_quake2)?0:-1;
//ent->num_leafs == q2's ent->num_clusters
ent->num_leafs = 0;
ent->areanum = nullarea;
ent->areanum2 = nullarea;
if (!mins || !maxs)
return;
//get all leafs, including solids
num_leafs = CM_BoxLeafnums (model, mins, maxs,
leafs, MAX_TOTAL_ENT_LEAFS, &topnode);
// set areas
for (i=0 ; i<num_leafs ; i++)
{
clusters[i] = CM_LeafCluster (model, leafs[i]);
area = CM_LeafArea (model, leafs[i]);
if (area != nullarea)
{ // doors may legally straggle two areas,
// but nothing should ever need more than that
if (ent->areanum != nullarea && ent->areanum != area)
ent->areanum2 = area;
else
ent->areanum = area;
}
}
if (num_leafs >= MAX_TOTAL_ENT_LEAFS)
{ // assume we missed some leafs, and mark by headnode
ent->num_leafs = -1;
ent->headnode = topnode;
}
else
{
ent->num_leafs = 0;
for (i=0 ; i<num_leafs ; i++)
{
if (clusters[i] == -1)
continue; // not a visible leaf
for (j=0 ; j<i ; j++)
if (clusters[j] == clusters[i])
break;
if (j == i)
{
if (ent->num_leafs == MAX_ENT_LEAFS)
{ // assume we missed some leafs, and mark by headnode
ent->num_leafs = -1;
ent->headnode = topnode;
break;
}
ent->leafnums[ent->num_leafs++] = clusters[i];
}
}
}
}
2023-03-27 17:22:00 +00:00
#endif
/*
===============================================================================
AREAPORTALS
===============================================================================
*/
static void FloodArea_r (cminfo_t *prv, size_t areaidx, int floodnum)
{
size_t i;
careaflood_t *flood = &prv->areaflood[areaidx];
if (flood->floodvalid == prv->floodvalid)
{
if (flood->floodnum == floodnum)
return;
Con_Printf ("FloodArea_r: reflooded\n");
return;
}
flood->floodnum = floodnum;
flood->floodvalid = prv->floodvalid;
switch(prv->mapisq3)
{
case true:
#ifdef Q3BSPS
for (i=0 ; i<prv->numareas ; i++)
{
if (prv->q3areas[areaidx].numareaportals[i]>0)
FloodArea_r (prv, i, floodnum);
}
#endif
break;
case false:
#ifdef Q2BSPS
{
q2carea_t *area = &prv->q2areas[areaidx];
q2dareaportal_t *p = &prv->q2areaportals[area->firstareaportal];
for (i=0 ; i<area->numareaportals ; i++, p++)
{
if (prv->q2portalopen[p->portalnum])
FloodArea_r (prv, p->otherarea, floodnum);
}
}
#endif
break;
}
}
/*
====================
FloodAreaConnections
====================
*/
static void FloodAreaConnections (cminfo_t *prv)
{
size_t i;
int floodnum;
// all current floods are now invalid
prv->floodvalid++;
floodnum = 0;
// area 0 is not used
for (i=0 ; i<prv->numareas ; i++)
{
if (prv->areaflood[i].floodvalid == prv->floodvalid)
continue; // already flooded into
floodnum++;
FloodArea_r (prv, i, floodnum);
}
}
static void CM_SetAreaPortalState (model_t *mod, unsigned int portalnum, unsigned int area1, unsigned int area2, qboolean open)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
switch(prv->mapisq3)
{
#ifdef Q3BSPS
case true:
if (area1 >= prv->numareas || area2 >= prv->numareas || area1==area2)
return;
if (open)
{
prv->q3areas[area1].numareaportals[area2]++;
prv->q3areas[area2].numareaportals[area1]++;
}
else
{
if (!prv->q3areas[area1].numareaportals[area2])
{
Con_Printf(CON_WARNING"CM_SetAreaPortalState: Areaportal closed more than opened...\n");
return;
}
prv->q3areas[area1].numareaportals[area2]--;
prv->q3areas[area2].numareaportals[area1]--;
}
break;
#endif
#ifdef Q2BSPS
case false:
if (portalnum > prv->numq2areaportals)
return;
if (prv->q2portalopen[portalnum] == open)
return;
prv->q2portalopen[portalnum] = open;
break;
#endif
default: break;
}
FloodAreaConnections (prv);
}
static qboolean VARGS CM_AreasConnected (model_t *mod, unsigned int area1, unsigned int area2)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
if (map_noareas.value)
return true;
if (area1 == ~0 || area2 == ~0)
return area1 == area2;
if (area1 > prv->numareas || area2 > prv->numareas)
Host_Error ("area > numareas");
if (prv->areaflood[area1].floodnum == prv->areaflood[area2].floodnum)
return true;
return false;
}
/*
=================
CM_WriteAreaBits
Writes a length qbyte followed by a bit vector of all the areas
that area in the same flood as the area parameter
This is used by the client refreshes to cull visibility
=================
*/
static size_t CM_WriteAreaBits (model_t *mod, qbyte *buffer, size_t buffersize, int area, qboolean merge)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
int i;
int floodnum;
int bytes;
bytes = (prv->numareas+7)>>3;
if (bytes > buffersize)
bytes = buffersize;
if (map_noareas.value || (area < 0 && !merge))
{ // for debugging, send everything
if (!merge)
memset (buffer, 255, bytes);
}
else
{
if (!merge)
memset (buffer, 0, bytes);
floodnum = prv->areaflood[area].floodnum;
for (i=0 ; i<prv->numareas ; i++)
{
if (prv->areaflood[i].floodnum == floodnum || !area)
buffer[i>>3] |= 1<<(i&7);
}
}
return bytes;
}
/*
===================
CM_WritePortalState
Returns a size+pointer to the data that needs to be written into a saved game.
===================
*/
static size_t CM_SaveAreaPortalBlob (model_t *mod, void **data)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
if (mod->type == mod_brush && (mod->fromgame == fg_quake2 || mod->fromgame == fg_quake3))
{
switch(prv->mapisq3)
{
#ifdef Q3BSPS
case true:
//endian issues. oh well.
*data = prv->q3areas;
return sizeof(prv->q3areas);
#endif
#ifdef Q2BSPS
case false:
*data = prv->q2portalopen;
return sizeof(prv->q2portalopen);
#endif
default: break;
}
}
*data = NULL;
return 0;
}
/*
===================
CM_ReadPortalState
Reads the portal state from a savegame file
and recalculates the area connections
===================
*/
static size_t CM_LoadAreaPortalBlob (model_t *mod, void *ptr, size_t ptrsize)
{
cminfo_t *prv = (cminfo_t*)mod->meshinfo;
if (mod->type == mod_brush && (mod->fromgame == fg_quake2 || mod->fromgame == fg_quake3))
{
switch(prv->mapisq3)
{
#ifdef Q3BSPS
case 1: //area*area refcounts. byte sizes don't tell us how many areas there were - would need sqrt(ptrsize/4) and I cba.
if (ptrsize != sizeof(prv->q3areas))
{ //don't bother trying to handle graceful expansion/truncation, just reset the entire thing.
size_t x,y;
if (ptrsize)
Con_Printf("CM_ReadPortalState() expected %u, but only %u available\n",(unsigned int)sizeof(prv->q3areas),(unsigned int)ptrsize);
for (x = 0; x < countof(prv->q3areas); x++)
for (y = 0; y < countof(prv->q3areas[x].numareaportals); y++)
prv->q3areas[x].numareaportals[y] = map_autoopenportals.ival;
}
else
memcpy(prv->q3areas, ptr, ptrsize);
FloodAreaConnections (prv);
return sizeof(prv->q3areas);
#endif
#ifdef Q2BSPS
case 0: //per-portal booleans. we can just pad any missing portals.
if (ptrsize && ptrsize != sizeof(prv->q2portalopen))
Con_Printf("CM_ReadPortalState() expected %u, but only %u available\n",(unsigned int)sizeof(prv->q2portalopen),(unsigned int)ptrsize);
if (ptrsize > sizeof(prv->q2portalopen))
ptrsize = sizeof(prv->q2portalopen);
memcpy(prv->q2portalopen, ptr, ptrsize);
memset(prv->q2portalopen+ptrsize, map_autoopenportals.ival, sizeof(prv->q2portalopen)-ptrsize);
FloodAreaConnections (prv);
return sizeof(prv->q2portalopen);
#endif
default: break;
}
}
return 0;
}
2023-03-27 17:22:00 +00:00
#ifdef HAVE_SERVER
/*
=============
CM_HeadnodeVisible
Returns true if any leaf under headnode has a cluster that
is potentially visible
=============
*/
static qboolean CM_HeadnodeVisible (model_t *mod, int nodenum, const qbyte *visbits)
{
int leafnum;
int cluster;
mnode_t *node;
if (nodenum < 0)
{
leafnum = -1-nodenum;
cluster = mod->leafs[leafnum].cluster;
if (cluster == -1)
return false;
if (visbits[cluster>>3] & (1<<(cluster&7)))
return true;
return false;
}
node = &mod->nodes[nodenum];
if (CM_HeadnodeVisible(mod, node->childnum[0], visbits))
return true;
return CM_HeadnodeVisible(mod, node->childnum[1], visbits);
}
2023-03-27 17:22:00 +00:00
#endif
static unsigned int Q2BSP_PointContents(model_t *mod, const vec3_t axis[3], const vec3_t p)
{
int pc;
pc = CM_PointContents (mod, p);
return pc;
}
#ifdef HAVE_CLIENT
static qbyte *frustumvis;
static vec3_t modelorg;
static unsigned int scenesequence;
static unsigned int vissequence;
/*
===============
R_MarkLeaves
===============
*/
#ifdef Q3BSPS
qbyte *R_MarkLeaves_Q3 (model_t *mod, int clusters[2])
{
static pvsbuffer_t curframevis[R_MAX_RECURSE];
qbyte *vis;
int i;
int cluster;
mleaf_t *leaf;
mnode_t *node;
int portal = r_refdef.recurse;
cminfo_t *prv = mod->meshinfo;
if (!portal)
{
if (prv->oldclusters[0] == clusters[0] && !r_novis.value && clusters[0] != -1)
return prv->oldvis;
}
// development aid to let you run around and see exactly where
// the pvs ends
// if (r_lockpvs->value)
// return;
vissequence++;
prv->oldclusters[0] = clusters[0];
if (r_novis.ival || clusters[0] == -1 || !mod->vis )
{
vis = NULL;
// mark everything
for (i=0,leaf=mod->leafs ; i<mod->numleafs ; i++, leaf++)
{
// if (!leaf->nummarksurfaces)
// {
// continue;
// }
#if 1
for (node = (mnode_t*)leaf; node; node = node->parent)
{
if (node->visframe == vissequence)
break;
node->visframe = vissequence;
}
#else
leaf->visframe = vissequence;
leaf->vischain = r_vischain;
r_vischain = leaf;
#endif
}
}
else
{
vis = CM_ClusterPVS (mod, clusters[0], &curframevis[portal], PVM_FAST);
for (i=0,leaf=mod->leafs ; i<mod->numleafs ; i++, leaf++)
{
cluster = leaf->cluster;
if (cluster == -1)// || !leaf->nummarksurfaces)
{
continue;
}
if (vis[cluster>>3] & (1<<(cluster&7)))
{
#if 1
for (node = (mnode_t*)leaf; node; node = node->parent)
{
if (node->visframe == vissequence)
break;
node->visframe = vissequence;
}
#else
leaf->visframe = vissequence;
leaf->vischain = r_vischain;
r_vischain = leaf;
#endif
}
}
}
prv->oldvis = vis;
return vis;
}
static void Surf_RecursiveQ3WorldNode (mnode_t *node, unsigned int clipflags)
{
int c, side, clipped;
mplane_t *plane, *clipplane;
msurface_t *surf, **mark;
mleaf_t *pleaf;
double dot;
start:
if (node->visframe != vissequence)
return;
for (c = 0, clipplane = r_refdef.frustum; c < r_refdef.frustum_numworldplanes; c++, clipplane++)
{
if (!(clipflags & (1 << c)))
continue; // don't need to clip against it
clipped = BOX_ON_PLANE_SIDE (node->minmaxs, node->minmaxs + 3, clipplane);
if (clipped == 2)
return;
else if (clipped == 1)
clipflags -= (1<<c); // node is entirely on screen
}
// if a leaf node, draw stuff
if (node->contents != -1)
{
pleaf = (mleaf_t *)node;
if (! (r_refdef.areabits[pleaf->area>>3] & (1<<(pleaf->area&7)) ) )
return; // not visible
c = pleaf->cluster;
if (c >= 0)
frustumvis[c>>3] |= 1<<(c&7);
mark = pleaf->firstmarksurface;
for (c = pleaf->nummarksurfaces; c; c--)
{
surf = *mark++;
if (surf->visframe == scenesequence)
continue;
surf->visframe = scenesequence;
// if (((dot < 0) ^ !!(surf->flags & SURF_PLANEBACK)))
// continue; // wrong side
surf->sbatch->mesh[surf->sbatch->meshes++] = surf->mesh;
}
return;
}
// node is just a decision point, so go down the apropriate sides
// find which side of the node we are on
plane = node->plane;
switch (plane->type)
{
case PLANE_X:
dot = modelorg[0] - plane->dist;
break;
case PLANE_Y:
dot = modelorg[1] - plane->dist;
break;
case PLANE_Z:
dot = modelorg[2] - plane->dist;
break;
default:
dot = DotProduct (modelorg, plane->normal) - plane->dist;
break;
}
if (dot >= 0)
side = 0;
else
side = 1;
// recurse down the children, front side first
Surf_RecursiveQ3WorldNode (node->children[side], clipflags);
// q3 nodes contain no drawables
// recurse down the back side
//GLR_RecursiveWorldNode (node->children[!side], clipflags);
node = node->children[!side];
goto start;
}
#endif
#ifdef Q2BSPS
qbyte *R_MarkLeaves_Q2 (model_t *mod, int viewclusters[2])
{
static pvsbuffer_t curframevis[R_MAX_RECURSE];
static qbyte *cvis[R_MAX_RECURSE];
mnode_t *node;
int i;
int cluster;
mleaf_t *leaf;
qbyte *vis;
int portal = r_refdef.recurse;
cminfo_t *prv = mod->meshinfo;
if (r_refdef.forcevis)
{
vis = cvis[portal] = r_refdef.forcedvis;
prv->oldclusters[0] = -1;
prv->oldclusters[1] = -1;
}
else
{
vis = cvis[portal];
if (!portal)
{
if (prv->oldclusters[0] == viewclusters[0] && prv->oldclusters[1] == viewclusters[1])
return vis;
prv->oldclusters[0] = viewclusters[0];
prv->oldclusters[1] = viewclusters[1];
}
else
{
prv->oldclusters[0] = -1;
prv->oldclusters[1] = -1;
}
if (r_novis.ival == 2)
return vis;
if (r_novis.ival || r_viewcluster == -1 || !mod->vis)
{
// mark everything
for (i=0 ; i<mod->numleafs ; i++)
mod->leafs[i].visframe = vissequence;
for (i=0 ; i<mod->numnodes ; i++)
mod->nodes[i].visframe = vissequence;
return vis;
}
if (viewclusters[1] != viewclusters[0]) // may have to combine two clusters because of solid water boundaries
{
vis = CM_ClusterPVS (mod, viewclusters[0], &curframevis[portal], PVM_REPLACE);
vis = CM_ClusterPVS (mod, viewclusters[1], &curframevis[portal], PVM_MERGE);
}
else
vis = CM_ClusterPVS (mod, viewclusters[0], &curframevis[portal], PVM_FAST);
cvis[portal] = vis;
}
vissequence++;
for (i=0,leaf=mod->leafs ; i<mod->numleafs ; i++, leaf++)
{
cluster = leaf->cluster;
if (cluster == -1)
continue;
if (vis[cluster>>3] & (1<<(cluster&7)))
{
node = (mnode_t *)leaf;
do
{
if (node->visframe == vissequence)
break;
node->visframe = vissequence;
node = node->parent;
} while (node);
}
}
return vis;
}
static void Surf_RecursiveQ2WorldNode (mnode_t *node)
{
int c, side;
mplane_t *plane;
msurface_t *surf, **mark;
mleaf_t *pleaf;
double dot;
int sidebit;
if (node->contents == Q2CONTENTS_SOLID)
return; // solid
if (node->visframe != vissequence)
return;
if (R_CullBox (node->minmaxs, node->minmaxs+3))
return;
// if a leaf node, draw stuff
if (node->contents != -1)
{
pleaf = (mleaf_t *)node;
// check for door connected areas
if (! (r_refdef.areabits[pleaf->area>>3] & (1<<(pleaf->area&7)) ) )
return; // not visible
c = pleaf->cluster;
if (c >= 0)
frustumvis[c>>3] |= 1<<(c&7);
mark = pleaf->firstmarksurface;
c = pleaf->nummarksurfaces;
if (c)
{
do
{
(*mark)->visframe = scenesequence;
mark++;
} while (--c);
}
return;
}
// node is just a decision point, so go down the apropriate sides
// find which side of the node we are on
plane = node->plane;
switch (plane->type)
{
case PLANE_X:
dot = modelorg[0] - plane->dist;
break;
case PLANE_Y:
dot = modelorg[1] - plane->dist;
break;
case PLANE_Z:
dot = modelorg[2] - plane->dist;
break;
default:
dot = DotProduct (modelorg, plane->normal) - plane->dist;
break;
}
if (dot >= 0)
{
side = 0;
sidebit = 0;
}
else
{
side = 1;
sidebit = SURF_PLANEBACK;
}
// recurse down the children, front side first
Surf_RecursiveQ2WorldNode (node->children[side]);
// draw stuff
for ( c = node->numsurfaces, surf = currentmodel->surfaces + node->firstsurface; c ; c--, surf++)
{
if (surf->visframe != scenesequence)
continue;
if ( (surf->flags & SURF_PLANEBACK) != sidebit )
continue; // wrong side
surf->visframe = 0;//scenesequence+1;//-1;
Surf_RenderDynamicLightmaps (surf);
surf->sbatch->mesh[surf->sbatch->meshes++] = surf->mesh;
}
// recurse down the back side
Surf_RecursiveQ2WorldNode (node->children[!side]);
}
#endif
static void CM_PrepareFrame(model_t *mod, refdef_t *refdef, int area, int viewclusters[2], pvsbuffer_t *vis, qbyte **entvis_out, qbyte **surfvis_out)
{
qbyte *surfvis, *entvis;
// qbyte *frustumvis;
if (vis->buffersize < mod->pvsbytes)
vis->buffer = BZ_Realloc(vis->buffer, vis->buffersize=mod->pvsbytes);
frustumvis = vis->buffer;
memset(frustumvis, 0, mod->pvsbytes);
VectorCopy (r_refdef.vieworg, modelorg);
scenesequence++;
#ifdef Q3BSPS
if (mod->fromgame == fg_quake3)
{
entvis = surfvis = R_MarkLeaves_Q3 (mod, viewclusters);
Surf_RecursiveQ3WorldNode (mod->nodes, (1<<r_refdef.frustum_numworldplanes)-1);
}
else
#endif
#ifdef Q2BSPS
if (mod->fromgame == fg_quake2)
{
entvis = surfvis = R_MarkLeaves_Q2 (mod, viewclusters);
Surf_RecursiveQ2WorldNode (mod->nodes);
}
else
#endif
{
entvis = surfvis = NULL;
}
*surfvis_out = frustumvis;
*entvis_out = entvis;
}
#endif
qboolean QDECL Mod_LoadQ2BrushModel (model_t *mod, void *buffer, size_t fsize)
{
mod->fromgame = fg_quake2;
return CM_LoadMap(mod, buffer, fsize, true) != NULL;
}
void CM_Init(void) //register cvars.
{
#define MAPOPTIONS "Map Cvar Options"
Cvar_Register(&map_noareas, MAPOPTIONS);
Cvar_Register(&map_noCurves, MAPOPTIONS);
Cvar_Register(&map_autoopenportals, MAPOPTIONS);
Cvar_Register(&q3bsp_surf_meshcollision_flag, MAPOPTIONS);
Cvar_Register(&q3bsp_surf_meshcollision_force, MAPOPTIONS);
Cvar_Register(&q3bsp_mergeq3lightmaps, MAPOPTIONS);
Cvar_Register(&q3bsp_ignorestyles, MAPOPTIONS);
Cvar_Register(&q3bsp_bihtraces, MAPOPTIONS);
Cvar_Register(&r_subdivisions, MAPOPTIONS);
CM_InitBoxHull ();
}
#endif