mirror of
https://github.com/yquake2/yquake2remaster.git
synced 2024-11-10 07:12:07 +00:00
models: mdl support grouped frames
This commit is contained in:
Denis Pauk
parent
a9d282e406
commit
dd6ae35d75
1 changed files with 140 additions and 52 deletions
|
|
@ -458,7 +458,7 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
/* local copy of all values */
|
||||
int skinwidth, skinheight, framesize;
|
||||
int num_meshes, num_skins, num_xyz, num_st, num_tris, num_glcmds,
|
||||
num_frames;
|
||||
num_frames, frame_count;
|
||||
int ofs_meshes, ofs_skins, ofs_st, ofs_tris, ofs_frames, ofs_glcmds,
|
||||
ofs_imgbit, ofs_end;
|
||||
|
||||
|
|
@ -488,16 +488,6 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
|
||||
framesize = sizeof(daliasxframe_t) + sizeof(dxtrivertx_t) * (num_xyz - 1);
|
||||
|
||||
ofs_meshes = sizeof(*pheader); // just skip header and go
|
||||
ofs_skins = ofs_meshes + num_meshes * sizeof(dmdxmesh_t);
|
||||
ofs_st = ofs_skins + num_skins * MAX_SKINNAME;
|
||||
ofs_tris = ofs_st + num_st * sizeof(dstvert_t);
|
||||
ofs_glcmds = ofs_tris + num_tris * sizeof(dtriangle_t);
|
||||
ofs_frames = ofs_glcmds + num_glcmds * sizeof(int);
|
||||
ofs_imgbit = ofs_frames + framesize * num_frames;
|
||||
/* one less as single vertx in frame by default */
|
||||
ofs_end = ofs_imgbit + (skinwidth * skinheight * num_skins);
|
||||
|
||||
/* validate */
|
||||
if (num_xyz <= 0)
|
||||
{
|
||||
|
|
@ -526,6 +516,103 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
return NULL;
|
||||
}
|
||||
|
||||
/* get full count of frames */
|
||||
{
|
||||
const byte *curr_pos;
|
||||
int i;
|
||||
|
||||
curr_pos = (byte*)buffer + sizeof(mdl_header_t);
|
||||
|
||||
// register all skins
|
||||
for (i = 0; i < num_skins; ++i)
|
||||
{
|
||||
int skin_type;
|
||||
|
||||
/* skip type / int */
|
||||
/* 0 = simple, !0 = group */
|
||||
/* this program can't read models composed of group frames! */
|
||||
skin_type = LittleLong(((int *)curr_pos)[0]);
|
||||
curr_pos += sizeof(int);
|
||||
if (skin_type)
|
||||
{
|
||||
R_Printf(PRINT_ALL, "%s: model %s has unsupported skin type %d",
|
||||
__func__, mod_name, skin_type);
|
||||
return NULL;
|
||||
}
|
||||
|
||||
curr_pos += skinwidth * skinheight;
|
||||
}
|
||||
|
||||
/* texcoordinates */
|
||||
curr_pos += sizeof(mdl_texcoord_t) * num_st;
|
||||
|
||||
/* triangles */
|
||||
curr_pos += sizeof(mdl_triangle_t) * num_tris;
|
||||
|
||||
frame_count = 0;
|
||||
|
||||
/* register all frames */
|
||||
for (i = 0; i < num_frames; ++i)
|
||||
{
|
||||
int frame_type, frames_skip;
|
||||
|
||||
/* Read frame data */
|
||||
/* skip type / int */
|
||||
/* 0 = simple, !0 = group */
|
||||
/* this program can't read models composed of group frames! */
|
||||
frame_type = LittleLong(((int *)curr_pos)[0]);
|
||||
curr_pos += sizeof(int);
|
||||
|
||||
frames_skip = 1;
|
||||
|
||||
if (frame_type)
|
||||
{
|
||||
frames_skip = LittleLong(((int *)curr_pos)[0]);
|
||||
/* skip count of frames */
|
||||
curr_pos += sizeof(int);
|
||||
/* skip bboxmin, bouding box min */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
/* skip bboxmax, bouding box max */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
|
||||
/* skip intervals */
|
||||
curr_pos += frames_skip * sizeof(float);
|
||||
}
|
||||
|
||||
/* skip bboxmin, bouding box min */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
/* skip bboxmax, bouding box max */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
/* name */
|
||||
curr_pos += sizeof(char) * 16;
|
||||
/* first vertex list */
|
||||
curr_pos += sizeof(dtrivertx_t) * num_xyz;
|
||||
|
||||
/* next frames in frame group is unsupported */
|
||||
curr_pos += (frames_skip - 1) * (
|
||||
/* bouding box */
|
||||
sizeof(dtrivertx_t) * 2 +
|
||||
/* name */
|
||||
16 +
|
||||
/* verts */
|
||||
sizeof(dtrivertx_t) * num_xyz
|
||||
);
|
||||
|
||||
frame_count += frames_skip;
|
||||
}
|
||||
}
|
||||
|
||||
/* generate offsets */
|
||||
ofs_meshes = sizeof(*pheader); // just skip header and go
|
||||
ofs_skins = ofs_meshes + num_meshes * sizeof(dmdxmesh_t);
|
||||
ofs_st = ofs_skins + num_skins * MAX_SKINNAME;
|
||||
ofs_tris = ofs_st + num_st * sizeof(dstvert_t);
|
||||
ofs_glcmds = ofs_tris + num_tris * sizeof(dtriangle_t);
|
||||
ofs_frames = ofs_glcmds + num_glcmds * sizeof(int);
|
||||
ofs_imgbit = ofs_frames + framesize * frame_count;
|
||||
/* one less as single vertx in frame by default */
|
||||
ofs_end = ofs_imgbit + (skinwidth * skinheight * num_skins);
|
||||
|
||||
*numskins = num_skins;
|
||||
extradata = Hunk_Begin(ofs_end + Q_max(*numskins, MAX_MD2SKINS) * sizeof(struct image_s *));
|
||||
pheader = Hunk_Alloc(ofs_end);
|
||||
|
|
@ -544,7 +631,7 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
pheader->num_tris = num_tris;
|
||||
pheader->num_glcmds = num_glcmds;
|
||||
pheader->num_imgbit = 8;
|
||||
pheader->num_frames = num_frames;
|
||||
pheader->num_frames = frame_count;
|
||||
|
||||
pheader->ofs_meshes = ofs_meshes;
|
||||
pheader->ofs_skins = ofs_skins;
|
||||
|
|
@ -637,27 +724,28 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
}
|
||||
}
|
||||
|
||||
frame_count = 0;
|
||||
|
||||
/* register all frames */
|
||||
for (i = 0; i < num_frames; ++i)
|
||||
{
|
||||
int frame_type, j, frames_skip;
|
||||
int frame_type, j, k, frames_skip;
|
||||
daliasframe_t *frame;
|
||||
dxtrivertx_t* poutvertx;
|
||||
dtrivertx_t *pinvertx;
|
||||
vec3_t scale, translate;
|
||||
|
||||
frame = (daliasframe_t *) ((byte *)pheader + ofs_frames + i * framesize);
|
||||
frame->scale[0] = LittleFloat(pinmodel->scale[0]) / 0xFF;
|
||||
frame->scale[1] = LittleFloat(pinmodel->scale[1]) / 0xFF;
|
||||
frame->scale[2] = LittleFloat(pinmodel->scale[2]) / 0xFF;
|
||||
scale[0] = LittleFloat(pinmodel->scale[0]) / 0xFF;
|
||||
scale[1] = LittleFloat(pinmodel->scale[1]) / 0xFF;
|
||||
scale[2] = LittleFloat(pinmodel->scale[2]) / 0xFF;
|
||||
|
||||
frame->translate[0] = LittleFloat(pinmodel->translate[0]);
|
||||
frame->translate[1] = LittleFloat(pinmodel->translate[1]);
|
||||
frame->translate[2] = LittleFloat(pinmodel->translate[2]);
|
||||
translate[0] = LittleFloat(pinmodel->translate[0]);
|
||||
translate[1] = LittleFloat(pinmodel->translate[1]);
|
||||
translate[2] = LittleFloat(pinmodel->translate[2]);
|
||||
|
||||
/* Read frame data */
|
||||
/* skip type / int */
|
||||
/* 0 = simple, !0 = group */
|
||||
/* this program can't read models composed of group frames! */
|
||||
frame_type = LittleLong(((int *)curr_pos)[0]);
|
||||
curr_pos += sizeof(int);
|
||||
|
||||
|
|
@ -675,39 +763,38 @@ Mod_LoadModel_MDL(const char *mod_name, const void *buffer, int modfilelen,
|
|||
|
||||
/* skip intervals */
|
||||
curr_pos += frames_skip * sizeof(float);
|
||||
|
||||
R_Printf(PRINT_DEVELOPER,
|
||||
"%s: model %s have used first one of %d subframes\n",
|
||||
__func__, mod_name, frames_skip);
|
||||
}
|
||||
|
||||
/* skip bboxmin, bouding box min */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
/* skip bboxmax, bouding box max */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
|
||||
memcpy(&frame->name, curr_pos, sizeof(char) * 16);
|
||||
curr_pos += sizeof(char) * 16;
|
||||
|
||||
poutvertx = (dxtrivertx_t*)&frame->verts[0];
|
||||
pinvertx = (dtrivertx_t*)curr_pos;
|
||||
// verts are all 8 bit, so no swapping needed
|
||||
for (j=0; j < num_xyz; j ++)
|
||||
for (k = 0; k < frames_skip; k ++)
|
||||
{
|
||||
Mod_LoadFrames_VertMD2(poutvertx + j, pinvertx[j].v);
|
||||
poutvertx[j].lightnormalindex = pinvertx[j].lightnormalindex;
|
||||
}
|
||||
curr_pos += sizeof(dtrivertx_t) * num_xyz;
|
||||
frame = (daliasframe_t *) ((byte *)pheader + ofs_frames + frame_count * framesize);
|
||||
frame->scale[0] = scale[0];
|
||||
frame->scale[1] = scale[1];
|
||||
frame->scale[2] = scale[2];
|
||||
|
||||
/* next frames in frame group is unsupported */
|
||||
curr_pos += (frames_skip - 1) * (
|
||||
/* bouding box */
|
||||
sizeof(dtrivertx_t) * 2 +
|
||||
/* name */
|
||||
16 +
|
||||
/* verts */
|
||||
sizeof(dtrivertx_t) * num_xyz
|
||||
);
|
||||
frame->translate[0] = translate[0];
|
||||
frame->translate[1] = translate[1];
|
||||
frame->translate[2] = translate[2];
|
||||
|
||||
/* skip bboxmin, bouding box min */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
/* skip bboxmax, bouding box max */
|
||||
curr_pos += sizeof(dtrivertx_t);
|
||||
|
||||
memcpy(&frame->name, curr_pos, sizeof(char) * 16);
|
||||
curr_pos += sizeof(char) * 16;
|
||||
|
||||
poutvertx = (dxtrivertx_t*)&frame->verts[0];
|
||||
pinvertx = (dtrivertx_t*)curr_pos;
|
||||
// verts are all 8 bit, so no swapping needed
|
||||
for (j=0; j < num_xyz; j ++)
|
||||
{
|
||||
Mod_LoadFrames_VertMD2(poutvertx + j, pinvertx[j].v);
|
||||
poutvertx[j].lightnormalindex = pinvertx[j].lightnormalindex;
|
||||
}
|
||||
curr_pos += sizeof(dtrivertx_t) * num_xyz;
|
||||
frame_count++;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
|
@ -2251,8 +2338,9 @@ Mod_LoadLimits(const char *mod_name, void *extradata, modtype_t type)
|
|||
num_glcmds += mesh_nodes[i].num_glcmds;
|
||||
}
|
||||
R_Printf(PRINT_DEVELOPER,
|
||||
"%s: model %s num tris %d / num vert %d / commands %d of %d\n",
|
||||
__func__, mod_name, pheader->num_tris, pheader->num_xyz, num_glcmds, pheader->num_glcmds);
|
||||
"%s: model %s num tris %d / num vert %d / commands %d of %d / frames %d\n",
|
||||
__func__, mod_name, pheader->num_tris, pheader->num_xyz, num_glcmds,
|
||||
pheader->num_glcmds, pheader->num_frames);
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Reference in a new issue