qf/quakeforge
0
0
Fork 0
mirror of https://git.code.sf.net/p/quake/quakeforge synced 2024-12-02 17:12:55 +00:00
Code Issues Projects Releases Packages Wiki Activity
quakeforge/libs/scene/test/test-hierarchy.c

730 lines
26 KiB
C
Raw Normal View History

[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
#include <stdio.h>
#include <string.h>
#include <unistd.h>
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
#include "QF/scene/component.h"
[scene] Rename libQFentity to libQFscene And start working on scene management.
2021-07-24 05:19:52 +00:00
#include "QF/scene/hierarchy.h"
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
#include "QF/scene/scene.h"
[scene] Rename libQFentity to libQFscene And start working on scene management.
2021-07-24 05:19:52 +00:00
#include "QF/scene/transform.h"
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
scene_t *scene;
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
// NOTE: these are the columns of the matrix! (not that it matters for a
// symmetrical matrix, but...)
mat4f_t identity = {
{ 1, 0, 0, 0 },
{ 0, 1, 0, 0 },
{ 0, 0, 1, 0 },
{ 0, 0, 0, 1 },
};
vec4f_t one = { 1, 1, 1, 1 };
static int
vec4_equal (vec4f_t a, vec4f_t b)
{
vec4i_t res = a != b;
return !(res[0] || res[1] || res[2] || res[3]);
}
static int
mat4_equal (const mat4f_t a, const mat4f_t b)
{
vec4i_t res = {};
for (int i = 0; i < 4; i++) {
res |= a[i] != b[i];
}
return !(res[0] || res[1] || res[2] || res[3]);
}
static int
check_hierarchy_size (hierarchy_t *h, uint32_t size)
{
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (h->num_objects != size) {
printf ("hierarchy does not have exactly %u transform\n", size);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
char **name = h->components[transform_type_name];
for (uint32_t i = 0; i < h->num_objects; i++) {
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (h->ref[i]->hierarchy != h) {
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
printf ("transform %d (%s) does not point to hierarchy\n",
i, name[i]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
}
}
return 1;
}
static void
dump_hierarchy (hierarchy_t *h)
{
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
char **name = h->components[transform_type_name];
for (uint32_t i = 0; i < h->num_objects; i++) {
printf ("%2d: %5s %2u %2u %2u %2u\n", i, name[i],
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
h->ref[i]->index, h->parentIndex[i],
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
h->childIndex[i], h->childCount[i]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
}
puts ("");
}
static int
check_indices (transform_t *transform, uint32_t index, uint32_t parentIndex,
uint32_t childIndex, uint32_t childCount)
{
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
__auto_type ref = &transform->ref;
hierarchy_t *h = ref->hierarchy;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
char **name = h->components[transform_type_name];
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (ref->index != index) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("%s/%s index incorrect: expect %u got %u\n",
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
name[ref->index], name[index],
index, ref->index);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (h->parentIndex[index] != parentIndex) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("%s parent index incorrect: expect %u got %u\n",
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
name[index], parentIndex, h->parentIndex[index]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (h->childIndex[index] != childIndex) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("%s child index incorrect: expect %u got %u\n",
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
name[index], childIndex, h->childIndex[index]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (h->childCount[index] != childCount) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("%s child count incorrect: expect %u got %u\n",
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
name[index], childCount, h->childCount[index]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
return 1;
}
static int
test_single_transform (void)
{
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *transform = Transform_New (scene, 0);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
hierarchy_t *h;
if (!transform) {
printf ("Transform_New returned null\n");
return 1;
}
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!(h = transform->ref.hierarchy)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("New transform has no hierarchy\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mat4f_t *localMatrix = h->components[transform_type_localMatrix];
mat4f_t *localInverse = h->components[transform_type_localInverse];
mat4f_t *worldMatrix = h->components[transform_type_worldMatrix];
mat4f_t *worldInverse = h->components[transform_type_worldInverse];
vec4f_t *localRotation = h->components[transform_type_localRotation];
vec4f_t *localScale = h->components[transform_type_localScale];
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_hierarchy_size (h, 1)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (transform, 0, nullent, 1, 0)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!mat4_equal (localMatrix[0], identity)
|| !mat4_equal (localInverse[0], identity)
|| !mat4_equal (worldMatrix[0], identity)
|| !mat4_equal (worldInverse[0], identity)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("New transform matrices not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!vec4_equal (localRotation[0], identity[3])
|| !vec4_equal (localScale[0], one)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("New transform rotation or scale not identity\n");
return 1;
}
// Delete the hierarchy directly as setparent isn't fully tested
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Hierarchy_Delete (transform->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
static int
test_parent_child_init (void)
{
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *parent = Transform_New (scene, 0);
transform_t *child = Transform_New (scene, parent);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (parent->ref.hierarchy != child->ref.hierarchy) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("parent and child transforms have separate hierarchies\n");
return 1;
}
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (parent->ref.hierarchy, 2)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (parent, 0, nullent, 1, 1)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices (child, 1, 0, 2, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
hierarchy_t *h = parent->ref.hierarchy;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mat4f_t *localMatrix = h->components[transform_type_localMatrix];
mat4f_t *localInverse = h->components[transform_type_localInverse];
mat4f_t *worldMatrix = h->components[transform_type_worldMatrix];
mat4f_t *worldInverse = h->components[transform_type_worldInverse];
vec4f_t *localRotation = h->components[transform_type_localRotation];
vec4f_t *localScale = h->components[transform_type_localScale];
if (!mat4_equal (localMatrix[0], identity)
|| !mat4_equal (localInverse[0], identity)
|| !mat4_equal (worldMatrix[0], identity)
|| !mat4_equal (worldInverse[0], identity)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Parent transform matrices not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!vec4_equal (localRotation[0], identity[3])
|| !vec4_equal (localScale[0], one)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Parent transform rotation or scale not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!mat4_equal (localMatrix[1], identity)
|| !mat4_equal (localInverse[1], identity)
|| !mat4_equal (worldMatrix[1], identity)
|| !mat4_equal (worldInverse[1], identity)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Child transform matrices not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!vec4_equal (localRotation[1], identity[3])
|| !vec4_equal (localScale[1], one)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Child transform rotation or scale not identity\n");
return 1;
}
// Delete the hierarchy directly as setparent isn't fully tested
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Hierarchy_Delete (parent->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
static int
test_parent_child_setparent (void)
{
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *parent = Transform_New (scene, 0);
transform_t *child = Transform_New (scene, 0);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
Transform_SetName (parent, "parent");
Transform_SetName (child, "child");
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (parent, 0, nullent, 1, 0)) { return 1; }
if (!check_indices (child, 0, nullent, 1, 0)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (parent->ref.hierarchy == child->ref.hierarchy) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("parent and child transforms have same hierarchy before"
" set paret\n");
return 1;
}
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Transform_SetParent (scene,child, parent);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (parent->ref.hierarchy != child->ref.hierarchy) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("parent and child transforms have separate hierarchies\n");
return 1;
}
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (parent->ref.hierarchy, 2)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (parent, 0, nullent, 1, 1)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices (child, 1, 0, 2, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
hierarchy_t *h = parent->ref.hierarchy;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mat4f_t *localMatrix = h->components[transform_type_localMatrix];
mat4f_t *localInverse = h->components[transform_type_localInverse];
mat4f_t *worldMatrix = h->components[transform_type_worldMatrix];
mat4f_t *worldInverse = h->components[transform_type_worldInverse];
vec4f_t *localRotation = h->components[transform_type_localRotation];
vec4f_t *localScale = h->components[transform_type_localScale];
if (!mat4_equal (localMatrix[0], identity)
|| !mat4_equal (localInverse[0], identity)
|| !mat4_equal (worldMatrix[0], identity)
|| !mat4_equal (worldInverse[0], identity)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Parent transform matrices not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!vec4_equal (localRotation[0], identity[3])
|| !vec4_equal (localScale[0], one)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Parent transform rotation or scale not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!mat4_equal (localMatrix[1], identity)
|| !mat4_equal (localInverse[1], identity)
|| !mat4_equal (worldMatrix[1], identity)
|| !mat4_equal (worldInverse[1], identity)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Child transform matrices not identity\n");
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
if (!vec4_equal (localRotation[1], identity[3])
|| !vec4_equal (localScale[1], one)) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("Child transform rotation or scale not identity\n");
return 1;
}
// Delete the hierarchy directly as setparent isn't fully tested
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Hierarchy_Delete (parent->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
static int
test_build_hierarchy (void)
{
printf ("test_build_hierarchy\n");
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *root = Transform_NewNamed (scene, 0, "root");
transform_t *A = Transform_NewNamed (scene, root, "A");
transform_t *B = Transform_NewNamed (scene, root, "B");
transform_t *C = Transform_NewNamed (scene, root, "C");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices (A, 1, 0, 4, 0)) { return 1; }
if (!check_indices (B, 2, 0, 4, 0)) { return 1; }
if (!check_indices (C, 3, 0, 4, 0)) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *B1 = Transform_NewNamed (scene, B, "B1");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 4, 0)) { return 1; }
if (!check_indices ( B, 2, 0, 4, 1)) { return 1; }
if (!check_indices ( C, 3, 0, 5, 0)) { return 1; }
if (!check_indices (B1, 4, 2, 5, 0)) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *A1 = Transform_NewNamed (scene, A, "A1");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 4, 1)) { return 1; }
if (!check_indices ( B, 2, 0, 5, 1)) { return 1; }
if (!check_indices ( C, 3, 0, 6, 0)) { return 1; }
if (!check_indices (A1, 4, 1, 6, 0)) { return 1; }
if (!check_indices (B1, 5, 2, 6, 0)) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *A1a = Transform_NewNamed (scene, A1, "A1a");
transform_t *B2 = Transform_NewNamed (scene, B, "B2");
transform_t *A2 = Transform_NewNamed (scene, A, "A2");
transform_t *B3 = Transform_NewNamed (scene, B, "B3");
transform_t *B2a = Transform_NewNamed (scene, B2, "B2a");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (root->ref.hierarchy, 11)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 4, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 6, 3)) { return 1; }
if (!check_indices ( C, 3, 0, 9, 0)) { return 1; }
if (!check_indices ( A1, 4, 1, 9, 1)) { return 1; }
if (!check_indices ( A2, 5, 1, 10, 0)) { return 1; }
if (!check_indices ( B1, 6, 2, 10, 0)) { return 1; }
if (!check_indices ( B2, 7, 2, 10, 1)) { return 1; }
if (!check_indices ( B3, 8, 2, 11, 0)) { return 1; }
if (!check_indices (A1a, 9, 4, 11, 0)) { return 1; }
if (!check_indices (B2a, 10, 7, 11, 0)) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *D = Transform_NewNamed (scene, root, "D");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (root->ref.hierarchy, 12)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 4)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 5, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 7, 3)) { return 1; }
if (!check_indices ( C, 3, 0, 10, 0)) { return 1; }
if (!check_indices ( D, 4, 0, 10, 0)) { return 1; }
if (!check_indices ( A1, 5, 1, 10, 1)) { return 1; }
if (!check_indices ( A2, 6, 1, 11, 0)) { return 1; }
if (!check_indices ( B1, 7, 2, 11, 0)) { return 1; }
if (!check_indices ( B2, 8, 2, 11, 1)) { return 1; }
if (!check_indices ( B3, 9, 2, 12, 0)) { return 1; }
if (!check_indices (A1a, 10, 5, 12, 0)) { return 1; }
if (!check_indices (B2a, 11, 8, 12, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
dump_hierarchy (root->ref.hierarchy);
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *C1 = Transform_NewNamed (scene, C, "C1");
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
dump_hierarchy (root->ref.hierarchy);
if (!check_hierarchy_size (root->ref.hierarchy, 13)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 4)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 5, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 7, 3)) { return 1; }
if (!check_indices ( C, 3, 0, 10, 1)) { return 1; }
if (!check_indices ( D, 4, 0, 11, 0)) { return 1; }
if (!check_indices ( A1, 5, 1, 11, 1)) { return 1; }
if (!check_indices ( A2, 6, 1, 12, 0)) { return 1; }
if (!check_indices ( B1, 7, 2, 12, 0)) { return 1; }
if (!check_indices ( B2, 8, 2, 12, 1)) { return 1; }
if (!check_indices ( B3, 9, 2, 13, 0)) { return 1; }
if (!check_indices ( C1, 10, 3, 13, 0)) { return 1; }
if (!check_indices (A1a, 11, 5, 13, 0)) { return 1; }
if (!check_indices (B2a, 12, 8, 13, 0)) { return 1; }
// Delete the hierarchy directly as setparent isn't fully tested
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Hierarchy_Delete (root->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
static int
test_build_hierarchy2 (void)
{
printf ("test_build_hierarchy2\n");
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *root = Transform_NewNamed (scene, 0, "root");
transform_t *A = Transform_NewNamed (scene, root, "A");
transform_t *B = Transform_NewNamed (scene, root, "B");
transform_t *C = Transform_NewNamed (scene, root, "C");
transform_t *B1 = Transform_NewNamed (scene, B, "B1");
transform_t *A1 = Transform_NewNamed (scene, A, "A1");
transform_t *A1a = Transform_NewNamed (scene, A1, "A1a");
transform_t *B2 = Transform_NewNamed (scene, B, "B2");
transform_t *A2 = Transform_NewNamed (scene, A, "A2");
transform_t *B3 = Transform_NewNamed (scene, B, "B3");
transform_t *B2a = Transform_NewNamed (scene, B2, "B2a");
transform_t *D = Transform_NewNamed (scene, root, "D");
transform_t *C1 = Transform_NewNamed (scene, C, "C1");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (root->ref.hierarchy, 13)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 4)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 5, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 7, 3)) { return 1; }
if (!check_indices ( C, 3, 0, 10, 1)) { return 1; }
if (!check_indices ( D, 4, 0, 11, 0)) { return 1; }
if (!check_indices ( A1, 5, 1, 11, 1)) { return 1; }
if (!check_indices ( A2, 6, 1, 12, 0)) { return 1; }
if (!check_indices ( B1, 7, 2, 12, 0)) { return 1; }
if (!check_indices ( B2, 8, 2, 12, 1)) { return 1; }
if (!check_indices ( B3, 9, 2, 13, 0)) { return 1; }
if (!check_indices ( C1, 10, 3, 13, 0)) { return 1; }
if (!check_indices (A1a, 11, 5, 13, 0)) { return 1; }
if (!check_indices (B2a, 12, 8, 13, 0)) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *T = Transform_NewNamed (scene, 0, "T");
transform_t *X = Transform_NewNamed (scene, T, "X");
transform_t *Y = Transform_NewNamed (scene, T, "Y");
transform_t *Z = Transform_NewNamed (scene, T, "Z");
transform_t *Y1 = Transform_NewNamed (scene, Y, "Y1");
transform_t *X1 = Transform_NewNamed (scene, X, "X1");
transform_t *X1a = Transform_NewNamed (scene, X1, "X1a");
transform_t *Y2 = Transform_NewNamed (scene, Y, "Y2");
transform_t *X2 = Transform_NewNamed (scene, X, "X2");
transform_t *Y3 = Transform_NewNamed (scene, Y, "Y3");
transform_t *Y2a = Transform_NewNamed (scene, Y2, "Y2a");
transform_t *Z1 = Transform_NewNamed (scene, Z, "Z1");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
dump_hierarchy (T->ref.hierarchy);
if (!check_hierarchy_size (T->ref.hierarchy, 12)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices ( T, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( X, 1, 0, 4, 2)) { return 1; }
if (!check_indices ( Y, 2, 0, 6, 3)) { return 1; }
if (!check_indices ( Z, 3, 0, 9, 1)) { return 1; }
if (!check_indices ( X1, 4, 1, 10, 1)) { return 1; }
if (!check_indices ( X2, 5, 1, 11, 0)) { return 1; }
if (!check_indices ( Y1, 6, 2, 11, 0)) { return 1; }
if (!check_indices ( Y2, 7, 2, 11, 1)) { return 1; }
if (!check_indices ( Y3, 8, 2, 12, 0)) { return 1; }
if (!check_indices ( Z1, 9, 3, 12, 0)) { return 1; }
if (!check_indices (X1a, 10, 4, 12, 0)) { return 1; }
if (!check_indices (Y2a, 11, 7, 12, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Transform_SetParent (scene,T, B);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
dump_hierarchy (root->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_hierarchy_size (root->ref.hierarchy, 25)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 4)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 5, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 7, 4)) { return 1; }
if (!check_indices ( C, 3, 0, 11, 1)) { return 1; }
if (!check_indices ( D, 4, 0, 12, 0)) { return 1; }
if (!check_indices ( A1, 5, 1, 12, 1)) { return 1; }
if (!check_indices ( A2, 6, 1, 13, 0)) { return 1; }
if (!check_indices ( B1, 7, 2, 13, 0)) { return 1; }
if (!check_indices ( B2, 8, 2, 13, 1)) { return 1; }
if (!check_indices ( B3, 9, 2, 14, 0)) { return 1; }
if (!check_indices ( T, 10, 2, 14, 3)) { return 1; }
if (!check_indices ( C1, 11, 3, 17, 0)) { return 1; }
if (!check_indices (A1a, 12, 5, 17, 0)) { return 1; }
if (!check_indices (B2a, 13, 8, 17, 0)) { return 1; }
if (!check_indices ( X, 14, 10, 17, 2)) { return 1; }
if (!check_indices ( Y, 15, 10, 19, 3)) { return 1; }
if (!check_indices ( Z, 16, 10, 22, 1)) { return 1; }
if (!check_indices ( X1, 17, 14, 23, 1)) { return 1; }
if (!check_indices ( X2, 18, 14, 24, 0)) { return 1; }
if (!check_indices ( Y1, 19, 15, 24, 0)) { return 1; }
if (!check_indices ( Y2, 20, 15, 24, 1)) { return 1; }
if (!check_indices ( Y3, 21, 15, 25, 0)) { return 1; }
if (!check_indices ( Z1, 22, 16, 25, 0)) { return 1; }
if (!check_indices (X1a, 23, 17, 25, 0)) { return 1; }
if (!check_indices (Y2a, 24, 20, 25, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Transform_SetParent (scene,Y, 0);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
dump_hierarchy (root->ref.hierarchy);
dump_hierarchy (Y->ref.hierarchy);
if (!check_hierarchy_size (root->ref.hierarchy, 20)) { return 1; }
if (!check_hierarchy_size (Y->ref.hierarchy, 5)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices (root, 0, nullent, 1, 4)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( A, 1, 0, 5, 2)) { return 1; }
if (!check_indices ( B, 2, 0, 7, 4)) { return 1; }
if (!check_indices ( C, 3, 0, 11, 1)) { return 1; }
if (!check_indices ( D, 4, 0, 12, 0)) { return 1; }
if (!check_indices ( A1, 5, 1, 12, 1)) { return 1; }
if (!check_indices ( A2, 6, 1, 13, 0)) { return 1; }
if (!check_indices ( B1, 7, 2, 13, 0)) { return 1; }
if (!check_indices ( B2, 8, 2, 13, 1)) { return 1; }
if (!check_indices ( B3, 9, 2, 14, 0)) { return 1; }
if (!check_indices ( T, 10, 2, 14, 3)) { return 1; }
if (!check_indices ( C1, 11, 3, 17, 0)) { return 1; }
if (!check_indices (A1a, 12, 5, 17, 0)) { return 1; }
if (!check_indices (B2a, 13, 8, 17, 0)) { return 1; }
if (!check_indices ( X, 14, 10, 16, 2)) { return 1; }
if (!check_indices ( Z, 15, 10, 18, 1)) { return 1; }
if (!check_indices ( X1, 16, 14, 19, 1)) { return 1; }
if (!check_indices ( X2, 17, 14, 20, 0)) { return 1; }
if (!check_indices ( Z1, 18, 15, 20, 0)) { return 1; }
if (!check_indices (X1a, 19, 16, 20, 0)) { return 1; }
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
if (!check_indices ( Y, 0, nullent, 1, 3)) { return 1; }
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!check_indices ( Y1, 1, 0, 4, 0)) { return 1; }
if (!check_indices ( Y2, 2, 0, 4, 1)) { return 1; }
if (!check_indices ( Y3, 3, 0, 5, 0)) { return 1; }
if (!check_indices (Y2a, 4, 2, 5, 0)) { return 1; }
// Delete the hierarchy directly as setparent isn't fully tested
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Hierarchy_Delete (root->ref.hierarchy);
Hierarchy_Delete (Y->ref.hierarchy);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
static int
check_vector (const transform_t *transform,
vec4f_t (*func) (const transform_t *t),
vec4f_t expect, const char *msg)
{
vec4f_t res = func(transform);
if (!vec4_equal (res, expect)) {
printf ("%s %s: expected "VEC4F_FMT" got "VEC4F_FMT"\n",
Transform_GetName (transform), msg,
VEC4_EXP (expect), VEC4_EXP (res));
return 0;
}
return 1;
}
static int
test_frames (void)
{
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
transform_t *root = Transform_NewNamed (scene, 0, "root");
transform_t *A = Transform_NewNamed (scene, root, "A");
transform_t *B = Transform_NewNamed (scene, root, "B");
transform_t *A1 = Transform_NewNamed (scene, A, "A1");
transform_t *B1 = Transform_NewNamed (scene, B, "B1");
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
Transform_SetLocalPosition (root, (vec4f_t) { 0, 0, 1, 1 });
Transform_SetLocalPosition (A, (vec4f_t) { 1, 0, 0, 1 });
Transform_SetLocalRotation (A, (vec4f_t) { 0.5, 0.5, 0.5, 0.5 });
Transform_SetLocalPosition (B, (vec4f_t) { 0, 1, 0, 1 });
Transform_SetLocalRotation (B, (vec4f_t) { 0.5, -0.5, 0.5, 0.5 });
Transform_SetLocalPosition (A1, (vec4f_t) { 1, 0, 0, 1 });
Transform_SetLocalRotation (A1, (vec4f_t) { -0.5, -0.5, -0.5, 0.5 });
Transform_SetLocalPosition (B1, (vec4f_t) { 0, 1, 0, 1 });
Transform_SetLocalRotation (B1, (vec4f_t) { -0.5, 0.5, -0.5, 0.5 });
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
hierarchy_t *h = root->ref.hierarchy;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mat4f_t *localMatrix = h->components[transform_type_localMatrix];
mat4f_t *localInverse = h->components[transform_type_localInverse];
mat4f_t *worldMatrix = h->components[transform_type_worldMatrix];
mat4f_t *worldInverse = h->components[transform_type_worldInverse];
char **name = h->components[transform_type_name];
for (uint32_t i = 0; i < h->num_objects; i++) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
mat4f_t res;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mmulf (res, localMatrix[i], localInverse[i]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!mat4_equal (res, identity)) {
printf ("%s: localInverse not inverse of localMatrix\n",
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
name[i]);
printf ("l: " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 3));
printf ("i: " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 3));
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("r: " VEC4F_FMT "\n", MAT4_ROW(res, 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 3));
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
puts (name[i]);
printf ("l: " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localMatrix[i], 3));
printf ("i: " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(localInverse[i], 3));
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
for (uint32_t i = 0; i < h->num_objects; i++) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
mat4f_t res;
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
mmulf (res, worldMatrix[i], worldInverse[i]);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (!mat4_equal (res, identity)) {
printf ("%s: worldInverse not inverse of worldMatrix\n",
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
name[i]);
printf ("l: " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 3));
printf ("i: " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 3));
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
printf ("r: " VEC4F_FMT "\n", MAT4_ROW(res, 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(res, 3));
return 1;
}
[scene] Move hierarchies to an ECS-based system This is the beginning of adding ECS to QF. While the previous iteration of hierarchies was a start in the direction towards ECS, this pulls most of the 3d-specific transform stuff out of the hierarchy "objects", making all the matrices and vectors/quaternions actual components (in the ECS sense). There's more work to be done with respect to the transform and entity members of hierarchy_t (entity should probably go away entirely, and transform should become hierref_t (or whatever its final name becomes), but I wanted to get things working sooner than later. The motivation for the effort was to allow views to use hierarchy_t, which should be possible once I get entity and transform sorted out. I am really glad I already had automated tests for hierarchies, as things proved to be a little tricky to get working due to forgetting why certain things were there.
2022-10-09 17:00:33 +00:00
puts (name[i]);
printf ("l: " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldMatrix[i], 3));
printf ("i: " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 0));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 1));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 2));
printf (" " VEC4F_FMT "\n", MAT4_ROW(worldInverse[i], 3));
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
}
if (!check_vector (root, Transform_GetLocalPosition,
(vec4f_t) { 0, 0, 1, 1 }, "local position")) {
return 1;
}
if (!check_vector (root, Transform_GetWorldPosition,
(vec4f_t) { 0, 0, 1, 1 }, "world position")) {
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (root, Transform_Forward, (vec4f_t) { 1, 0, 0, 0 },
"forward")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (root, Transform_Right, (vec4f_t) { 0, -1, 0, 0 },
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
"right")) {
return 1;
}
if (!check_vector (root, Transform_Up, (vec4f_t) { 0, 0, 1, 0 },
"up")) {
return 1;
}
if (!check_vector (A, Transform_GetLocalPosition, (vec4f_t) { 1, 0, 0, 1 },
"local position")) {
return 1;
}
if (!check_vector (A, Transform_GetWorldPosition, (vec4f_t) { 1, 0, 1, 1 },
"world position")) {
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (A, Transform_Forward, (vec4f_t) { 0, 1, 0, 0 },
"forward")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (A, Transform_Right, (vec4f_t) { 0, 0, -1, 0 },
"right")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
if (!check_vector (A, Transform_Up, (vec4f_t) { 1, 0, 0, 0 },
"up")) {
return 1;
}
if (!check_vector (A1, Transform_GetLocalPosition, (vec4f_t) { 1, 0, 0, 1 },
"local position")) {
return 1;
}
if (!check_vector (A1, Transform_GetWorldPosition, (vec4f_t) { 1, 1, 1, 1 },
"world position")) {
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (A1, Transform_Forward, (vec4f_t) { 1, 0, 0, 0 },
"forward")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (A1, Transform_Right, (vec4f_t) { 0, -1, 0, 0 },
"right")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
if (!check_vector (A1, Transform_Up, (vec4f_t) { 0, 0, 1, 0 },
"up")) {
return 1;
}
if (!check_vector (B, Transform_GetLocalPosition, (vec4f_t) { 0, 1, 0, 1 },
"local position")) {
return 1;
}
if (!check_vector (B, Transform_GetWorldPosition, (vec4f_t) { 0, 1, 1, 1 },
"world position")) {
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (B, Transform_Forward, (vec4f_t) { 0, 0, 1, 0 },
"forward")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (B, Transform_Right, (vec4f_t) { 1, 0, 0, 0 },
"right")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
if (!check_vector (B, Transform_Up, (vec4f_t) { 0,-1, 0, 0 },
"up")) {
return 1;
}
if (!check_vector (B1, Transform_GetLocalPosition, (vec4f_t) { 0, 1, 0, 1 },
"local position")) {
return 1;
}
if (!check_vector (B1, Transform_GetWorldPosition, (vec4f_t) {-1, 1, 1, 1 },
"world position")) {
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (B1, Transform_Forward, (vec4f_t) { 1, 0, 0, 0 },
"forward")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
[entity] Fix tests for changed forward/right I had forgotten that Quake has forward as +X and right as -Y (still a right-handed system, just that with X forward, Y points left).
2021-03-21 11:45:43 +00:00
if (!check_vector (B1, Transform_Right, (vec4f_t) { 0, -1, 0, 0 },
"right")) {
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 1;
}
if (!check_vector (B1, Transform_Up, (vec4f_t) { 0, 0, 1, 0 },
"up")) {
return 1;
}
[scene] Use hierref_t instead of transform_t in hierarchy Hierarchies are now much closer to being more general in that they are not tied to 3d transforms. This is a major step to moving the whole entity/transform system into an ECS.
2022-10-22 04:04:07 +00:00
Transform_Delete (scene,root);
[entity] Free all memory in the unit test Now I know that deleting transforms works properly.
2021-03-21 13:33:57 +00:00
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
int
main (void)
{
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
scene = Scene_NewScene ();
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
if (test_single_transform ()) { return 1; }
if (test_parent_child_init ()) { return 1; }
if (test_parent_child_setparent ()) { return 1; }
if (test_build_hierarchy ()) { return 1; }
if (test_build_hierarchy2 ()) { return 1; }
if (test_frames ()) { return 1; }
[scene] Make transforms owned by the scene This actually has at least two benefits: the transform id is managed by the scene and thus does not need separate management by the Ruamoko wrapper functions, and better memory handling of the transform objects. Another benefit that isn't realized yet is that this is a step towards breaking the renderers free of quake and quakeworld: although the clients don't actually use the scene yet, it will be a good place to store the rendering information (functions to run, etc).
2022-02-14 11:01:36 +00:00
Scene_DeleteScene (scene);
[entity] Start work on a new entity library The plan is to have a fully component based entity system. This adds hierarchical transforms. Not particularly useful for quake itself at this stage, but it will allow for much more flexibility later on, especially when QuakeForge becomes more general-purpose.
2021-03-09 02:39:41 +00:00
return 0;
}
Reference in a new issue Copy permalink