Relintai/pandemonium_engine_minimal
1
0
Fork 0
mirror of https://github.com/Relintai/pandemonium_engine_minimal.git synced 2024-11-10 20:12:10 +01:00
Code Issues Packages Projects Releases Wiki Activity
ed65ffa412
pandemonium_engine_minimal/servers/rendering/rendering_server_canvas.cpp

1873 lines
64 KiB
C++
Raw Blame History

/*************************************************************************/
/* rendering_server_canvas.cpp */
/*************************************************************************/
/* This file is part of: */
/* GODOT ENGINE */
/* https://godotengine.org */
/*************************************************************************/
/* Copyright (c) 2007-2022 Juan Linietsky, Ariel Manzur. */
/* Copyright (c) 2014-2022 Godot Engine contributors (cf. AUTHORS.md). */
/* */
/* Permission is hereby granted, free of charge, to any person obtaining */
/* a copy of this software and associated documentation files (the */
/* "Software"), to deal in the Software without restriction, including */
/* without limitation the rights to use, copy, modify, merge, publish, */
/* distribute, sublicense, and/or sell copies of the Software, and to */
/* permit persons to whom the Software is furnished to do so, subject to */
/* the following conditions: */
/* */
/* The above copyright notice and this permission notice shall be */
/* included in all copies or substantial portions of the Software. */
/* */
/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */
/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */
/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/
/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */
/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */
/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */
/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */
/*************************************************************************/
#include "core/config/project_settings.h"
#include "core/containers/fixed_array.h"
#include "core/math/transform_interpolator.h"
#include "rendering_server_canvas.h"
#include "rendering_server_globals.h"
#include "rendering_server_raster.h"
#include "rendering_server_viewport.h"
static const int z_range = RS::CANVAS_ITEM_Z_MAX - RS::CANVAS_ITEM_Z_MIN + 1;
void RenderingServerCanvas::_render_canvas_item_tree(Item *p_canvas_item, const Transform2D &p_transform, const Rect2 &p_clip_rect, const Color &p_modulate) {
memset(z_list, 0, z_range * sizeof(RasterizerCanvas::Item *));
memset(z_last_list, 0, z_range * sizeof(RasterizerCanvas::Item *));
_current_camera_transform = p_transform;
if (_canvas_cull_mode == CANVAS_CULL_MODE_NODE) {
_prepare_tree_bounds(p_canvas_item);
_render_canvas_item_cull_by_node(p_canvas_item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr, false);
} else {
_render_canvas_item_cull_by_item(p_canvas_item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr);
}
RSG::canvas_render->canvas_render_items_begin(p_modulate, p_transform);
for (int i = 0; i < z_range; i++) {
if (!z_list[i]) {
continue;
}
RSG::canvas_render->canvas_render_items(z_list[i], RS::CANVAS_ITEM_Z_MIN + i, p_modulate, p_transform);
}
RSG::canvas_render->canvas_render_items_end();
}
void _collect_ysort_children(RenderingServerCanvas::Item *p_canvas_item, Transform2D p_transform, RenderingServerCanvas::Item *p_material_owner, const Color p_modulate, RenderingServerCanvas::Item **r_items, int &r_index) {
int child_item_count = p_canvas_item->child_items.size();
RenderingServerCanvas::Item **child_items = p_canvas_item->child_items.ptrw();
for (int i = 0; i < child_item_count; i++) {
if (child_items[i]->visible) {
if (r_items) {
r_items[r_index] = child_items[i];
child_items[i]->ysort_modulate = p_modulate;
child_items[i]->ysort_xform = p_transform;
child_items[i]->ysort_pos = p_transform.xform(child_items[i]->xform_curr.columns[2]);
child_items[i]->material_owner = child_items[i]->use_parent_material ? p_material_owner : nullptr;
child_items[i]->ysort_index = r_index;
}
r_index++;
if (child_items[i]->sort_y) {
_collect_ysort_children(child_items[i],
p_transform * child_items[i]->xform_curr,
child_items[i]->use_parent_material ? p_material_owner : child_items[i],
p_modulate * child_items[i]->modulate,
r_items, r_index);
}
}
}
}
void _mark_ysort_dirty(RenderingServerCanvas::Item *ysort_owner, RID_Owner<RenderingServerCanvas::Item> &canvas_item_owner) {
do {
ysort_owner->ysort_children_count = -1;
ysort_owner = canvas_item_owner.owns(ysort_owner->parent) ? canvas_item_owner.getornull(ysort_owner->parent) : nullptr;
} while (ysort_owner && ysort_owner->sort_y);
}
void RenderingServerCanvas::_make_bound_dirty_reparent(Item *p_item) {
MutexLock lock(_bound_mutex);
DEV_ASSERT(p_item);
Item *p_orig_item = p_item;
// propagate up
while (p_item) {
// Don't worry about invisible objects
if (!p_item->visible) {
return;
}
if (!p_item->bound_dirty) {
p_item->bound_dirty = true;
if (canvas_item_owner.owns(p_item->parent)) {
p_item = canvas_item_owner.get(p_item->parent);
} else {
break;
}
} else {
break;
}
}
// propagate down
_make_bound_dirty_down(p_orig_item);
}
void RenderingServerCanvas::_make_bound_dirty(Item *p_item, bool p_changing_visibility) {
if (_canvas_cull_mode != CANVAS_CULL_MODE_NODE) {
return;
}
MutexLock lock(_bound_mutex);
DEV_ASSERT(p_item);
if (!p_changing_visibility) {
_check_bound_integrity(p_item);
}
if (!p_changing_visibility) {
// Traverse up the tree, making each item bound dirty until
// we reach an item that is already dirty (as by definition, if this happens,
// the tree should already be dirty up until the root).
while (p_item) {
// Don't worry about invisible objects
if (!p_item->visible) {
return;
}
if (!p_item->bound_dirty) {
p_item->bound_dirty = true;
if (canvas_item_owner.owns(p_item->parent)) {
p_item = canvas_item_owner.get(p_item->parent);
} else {
break;
}
} else {
break;
}
}
} else {
// special case for visibility changes.
// if hiding, we propagate upwards.
while (p_item) {
if (!p_item->bound_dirty) {
p_item->bound_dirty = true;
}
if (canvas_item_owner.owns(p_item->parent)) {
p_item = canvas_item_owner.get(p_item->parent);
} else {
break;
}
}
// if showing we propagate upwards AND downwards
if (p_item->visible) {
_make_bound_dirty_down(p_item);
}
}
}
void RenderingServerCanvas::_make_bound_dirty_down(Item *p_item) {
// Bounds below an item that is being made visible may be out of date,
// so we make them all dirty.
if (!p_item->visible) {
return;
}
p_item->bound_dirty = true;
int child_item_count = p_item->child_items.size();
Item **child_items = p_item->child_items.ptrw();
for (int i = 0; i < child_item_count; i++) {
_make_bound_dirty_down(child_items[i]);
}
}
void RenderingServerCanvas::_prepare_tree_bounds(Item *p_root) {
Rect2 root_bound;
_calculate_canvas_item_bound(p_root, &root_bound);
}
// This function provides an alternative means of recursively calculating canvas item
// bounds through a branch, leading to an identical (hopefully) result to that
// calculated for the bound in _render_canvas_item().
// The reason for this function's existence is that there are some conditions which
// prevent further drawing in the tree (such as alpha nearing 0.0), in which we
// *still* need to calculate the bounds for lower branches and use them for culling,
// just in case alpha increases above the threshold in a later frame.
void RenderingServerCanvas::_calculate_canvas_item_bound(Item *p_canvas_item, Rect2 *r_branch_bound) {
// TODO - this higher level technique may be able to be optimized better,
// to perhaps only recalculate this on "reappearance" of the child branch, in a
// similar manner to how visibility is handled.
Item *ci = p_canvas_item;
if (!ci->visible) {
return;
}
// easy case, not dirty
if (!ci->bound_dirty) {
_merge_local_bound_to_branch(ci, r_branch_bound);
return;
}
// recalculate the local bound only if out of date
Rect2 *local_bound = nullptr;
if (ci->bound_dirty) {
local_bound = &ci->local_bound;
*local_bound = Rect2();
ci->bound_dirty = false;
}
int child_item_count = ci->child_items.size();
Item **child_items = ci->child_items.ptrw();
for (int i = 0; i < child_item_count; i++) {
// if (ci->sort_y)
// NYI do we need to apply the child_items[i]->ysort_xform? TEST
// See the _render_canvas_item for how to apply.
_calculate_canvas_item_bound(child_items[i], local_bound);
}
_finalize_and_merge_local_bound_to_branch(ci, r_branch_bound);
// If we are interpolating, we want to modify the local_bound (combined)
// to include both the previous AND current bounds.
if (local_bound && _interpolation_data.interpolation_enabled && ci->interpolated) {
Rect2 bound_prev = ci->local_bound_prev;
// Keep track of the previously assigned exact bound for the next tick.
ci->local_bound_prev = ci->local_bound;
// The combined bound is the exact current bound merged with the previous exact bound.
ci->local_bound = ci->local_bound.merge(bound_prev);
// This can overflow, it's no problem, it is just rough to detect when items stop
// having local bounds updated, so we can set prev to curr.
ci->local_bound_last_update_tick = Engine::get_singleton()->get_physics_frames();
// Detect special case of overflow.
// This is omitted but included for reference.
// It is such a rare possibility, and even if it did occur
// so it should just result in slightly larger culling bounds
// probably for one tick (and no visual errors).
// Would occur once every 828.5 days at 60 ticks per second
// with uint32_t counter.
#if 0
if (!ci->local_bound_last_update_tick) {
// Prevents it being treated as non-dirty.
// Just has an increased delay of one tick in this very rare occurrence.
ci->local_bound_last_update_tick = 1;
}
#endif
}
}
Transform2D RenderingServerCanvas::_calculate_item_global_xform(const Item *p_canvas_item) {
// If we use more than the maximum scene tree depth, we are out of luck.
// But that would be super inefficient anyway.
FixedArray<const Transform2D *, 64> transforms;
while (p_canvas_item) {
// Should only happen if scene tree depth too high.
if (transforms.is_full()) {
WARN_PRINT_ONCE("SceneTree depth too high for hierarchical culling.");
break;
}
// Note this is only using the CURRENT transform.
// This may have implications for interpolated bounds - investigate.
transforms.push_back(&p_canvas_item->xform_curr);
if (canvas_item_owner.owns(p_canvas_item->parent)) {
p_canvas_item = canvas_item_owner.get(p_canvas_item->parent);
} else {
p_canvas_item = nullptr;
}
}
Transform2D tr;
for (int n = (int)transforms.size() - 1; n >= 0; n--) {
tr *= *transforms[n];
}
return tr;
}
void RenderingServerCanvas::_finalize_and_merge_local_bound_to_branch(Item *p_canvas_item, Rect2 *r_branch_bound) {
if (r_branch_bound) {
Rect2 this_rect = p_canvas_item->get_rect();
// Special case .. if the canvas_item has use_identity_xform,
// we need to transform the rect from global space to local space,
// because the hierarchical culling expects local space.
if (p_canvas_item->use_identity_xform) {
// This is incredibly inefficient, but should only occur for e.g. CPUParticles2D,
// and is difficult to avoid because global transform is not usually kept track of
// in VisualServer (only final transform which is combinated with camera, and that
// is only calculated on render, so is no use for culling purposes).
Transform2D global_xform = _calculate_item_global_xform(p_canvas_item);
this_rect = global_xform.affine_inverse().xform(this_rect);
// Note that the efficiency will depend linearly on the scene tree depth of the
// identity transform item.
// So e.g. interpolated global CPUParticles2D may run faster at lower depths
// in extreme circumstances.
}
// If this item has a bound...
if (!p_canvas_item->local_bound.has_no_area()) {
// If the rect has an area...
if (!this_rect.has_no_area()) {
p_canvas_item->local_bound = p_canvas_item->local_bound.merge(this_rect);
} else {
// The local bound is set by the children, but is not affected by the canvas item rect.
// So pass through and merge the local bound to the parent.
}
} else {
p_canvas_item->local_bound = this_rect;
// don't merge zero area, as it may expand the branch bound
// unnecessarily.
if (p_canvas_item->local_bound.has_no_area()) {
return;
}
}
// Merge the local bound to the parent.
_merge_local_bound_to_branch(p_canvas_item, r_branch_bound);
}
}
void RenderingServerCanvas::_merge_local_bound_to_branch(Item *p_canvas_item, Rect2 *r_branch_bound) {
if (!r_branch_bound) {
return;
}
Rect2 this_item_total_local_bound = p_canvas_item->xform_curr.xform(p_canvas_item->local_bound);
if (!r_branch_bound->has_no_area()) {
*r_branch_bound = r_branch_bound->merge(this_item_total_local_bound);
} else {
*r_branch_bound = this_item_total_local_bound;
}
}
void RenderingServerCanvas::_render_canvas_item_cull_by_item(Item *p_canvas_item, const Transform2D &p_transform, const Rect2 &p_clip_rect, const Color &p_modulate, int p_z, RasterizerCanvas::Item **z_list, RasterizerCanvas::Item **z_last_list, Item *p_canvas_clip, Item *p_material_owner) {
Item *ci = p_canvas_item;
if (!ci->visible) {
return;
}
if (ci->children_order_dirty) {
ci->child_items.sort_custom<ItemIndexSort>();
ci->children_order_dirty = false;
}
Rect2 rect = ci->get_rect();
Transform2D final_xform;
if (!_interpolation_data.interpolation_enabled || !ci->interpolated) {
final_xform = ci->xform_curr;
} else {
real_t f = Engine::get_singleton()->get_physics_interpolation_fraction();
TransformInterpolator::interpolate_transform_2d(ci->xform_prev, ci->xform_curr, final_xform, f);
}
// Always calculate final transform as if not using identity xform.
// This is so the expected transform is passed to children.
// However, if use_identity_xform is set,
// we can override the transform for rendering purposes for this item only.
final_xform = p_transform * final_xform;
Rect2 global_rect;
if (!p_canvas_item->use_identity_xform) {
global_rect = final_xform.xform(rect);
} else {
global_rect = _current_camera_transform.xform(rect);
}
global_rect.position += p_clip_rect.position;
if (ci->use_parent_material && p_material_owner) {
ci->material_owner = p_material_owner;
} else {
p_material_owner = ci;
ci->material_owner = nullptr;
}
Color modulate(ci->modulate.r * p_modulate.r, ci->modulate.g * p_modulate.g, ci->modulate.b * p_modulate.b, ci->modulate.a * p_modulate.a);
if (modulate.a < 0.007) {
return;
}
int child_item_count = ci->child_items.size();
Item **child_items = ci->child_items.ptrw();
if (ci->clip) {
if (p_canvas_clip != nullptr) {
ci->final_clip_rect = p_canvas_clip->final_clip_rect.clip(global_rect);
} else {
ci->final_clip_rect = global_rect;
}
ci->final_clip_rect.position = ci->final_clip_rect.position.round();
ci->final_clip_rect.size = ci->final_clip_rect.size.round();
ci->final_clip_owner = ci;
} else {
ci->final_clip_owner = p_canvas_clip;
}
if (ci->sort_y) {
if (ci->ysort_children_count == -1) {
ci->ysort_children_count = 0;
_collect_ysort_children(ci, Transform2D(), p_material_owner, Color(1, 1, 1, 1), nullptr, ci->ysort_children_count);
}
child_item_count = ci->ysort_children_count;
child_items = (Item **)alloca(child_item_count * sizeof(Item *));
int i = 0;
_collect_ysort_children(ci, Transform2D(), p_material_owner, Color(1, 1, 1, 1), child_items, i);
SortArray<Item *, ItemPtrSort> sorter;
sorter.sort(child_items, child_item_count);
}
if (ci->z_relative) {
p_z = CLAMP(p_z + ci->z_index, RS::CANVAS_ITEM_Z_MIN, RS::CANVAS_ITEM_Z_MAX);
} else {
p_z = ci->z_index;
}
for (int i = 0; i < child_item_count; i++) {
if (!child_items[i]->behind || (ci->sort_y && child_items[i]->sort_y)) {
continue;
}
if (ci->sort_y) {
_render_canvas_item_cull_by_item(child_items[i], final_xform * child_items[i]->ysort_xform, p_clip_rect, modulate * child_items[i]->ysort_modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, (Item *)child_items[i]->material_owner);
} else {
_render_canvas_item_cull_by_item(child_items[i], final_xform, p_clip_rect, modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, p_material_owner);
}
}
if (ci->copy_back_buffer) {
ci->copy_back_buffer->screen_rect = final_xform.xform(ci->copy_back_buffer->rect).clip(p_clip_rect);
}
if (ci->update_when_visible) {
RenderingServerRaster::redraw_request(false);
}
if ((!ci->commands.empty() && p_clip_rect.intersects(global_rect, true)) || ci->vp_render || ci->copy_back_buffer) {
//something to draw?
ci->final_transform = !p_canvas_item->use_identity_xform ? final_xform : _current_camera_transform;
ci->final_modulate = Color(modulate.r * ci->self_modulate.r, modulate.g * ci->self_modulate.g, modulate.b * ci->self_modulate.b, modulate.a * ci->self_modulate.a);
ci->global_rect_cache = global_rect;
ci->global_rect_cache.position -= p_clip_rect.position;
ci->light_masked = false;
int zidx = p_z - RS::CANVAS_ITEM_Z_MIN;
if (z_last_list[zidx]) {
z_last_list[zidx]->next = ci;
z_last_list[zidx] = ci;
} else {
z_list[zidx] = ci;
z_last_list[zidx] = ci;
}
ci->next = nullptr;
}
for (int i = 0; i < child_item_count; i++) {
if (child_items[i]->behind || (ci->sort_y && child_items[i]->sort_y)) {
continue;
}
if (ci->sort_y) {
_render_canvas_item_cull_by_item(child_items[i], final_xform * child_items[i]->ysort_xform, p_clip_rect, modulate * child_items[i]->ysort_modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, (Item *)child_items[i]->material_owner);
} else {
_render_canvas_item_cull_by_item(child_items[i], final_xform, p_clip_rect, modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, p_material_owner);
}
}
}
void RenderingServerCanvas::_render_canvas_item_cull_by_node(Item *p_canvas_item, const Transform2D &p_transform, const Rect2 &p_clip_rect, const Color &p_modulate, int p_z, RasterizerCanvas::Item **z_list, RasterizerCanvas::Item **z_last_list, Item *p_canvas_clip, Item *p_material_owner, bool p_enclosed) {
Item *ci = p_canvas_item;
if (!ci->visible) {
return;
}
// This should have been calculated as a pre-process.
DEV_ASSERT(!ci->bound_dirty);
// If we are interpolating, and the updates have stopped, we can reduce the local bound.
if (ci->local_bound_last_update_tick && (ci->local_bound_last_update_tick != Engine::get_singleton()->get_physics_frames())) {
// The combined bound is reduced to the last calculated exact bound.
ci->local_bound = ci->local_bound_prev;
ci->local_bound_last_update_tick = 0;
}
Rect2 rect = ci->get_rect();
Transform2D final_xform;
if (!_interpolation_data.interpolation_enabled || !ci->interpolated) {
final_xform = ci->xform_curr;
} else {
real_t f = Engine::get_singleton()->get_physics_interpolation_fraction();
TransformInterpolator::interpolate_transform_2d(ci->xform_prev, ci->xform_curr, final_xform, f);
}
// Always calculate final transform as if not using identity xform.
// This is so the expected transform is passed to children.
// However, if use_identity_xform is set,
// we can override the transform for rendering purposes for this item only.
final_xform = p_transform * final_xform;
Rect2 global_rect;
if (!p_canvas_item->use_identity_xform) {
global_rect = final_xform.xform(rect);
} else {
global_rect = _current_camera_transform.xform(rect);
}
ci->global_rect_cache = global_rect;
ci->final_transform = !p_canvas_item->use_identity_xform ? final_xform : _current_camera_transform;
global_rect.position += p_clip_rect.position;
int child_item_count = ci->child_items.size();
// If there are children, we can maybe cull them all out if the cached bound has not changed.
if (!p_enclosed && child_item_count) {
// Get the bound in final space.
Rect2 bound = final_xform.xform(ci->local_bound);
bound.position += p_clip_rect.position;
if (!ci->vp_render && !ci->copy_back_buffer) {
// Cull out ALL children in one step.
if (!p_clip_rect.intersects(bound, true)) {
return;
}
}
// can we combine with earlier check?
// if we enclose the bound completely, no need to check further children
p_enclosed = p_clip_rect.encloses(bound);
}
// if we are culled, and no children, no more needs doing
bool item_is_visible = ((!ci->commands.empty() && (p_enclosed ? true : p_clip_rect.intersects(global_rect, true))) || ci->vp_render || ci->copy_back_buffer);
if (!item_is_visible && !child_item_count) {
return;
}
if (ci->use_parent_material && p_material_owner) {
ci->material_owner = p_material_owner;
} else {
p_material_owner = ci;
ci->material_owner = nullptr;
}
Color modulate(ci->modulate.r * p_modulate.r, ci->modulate.g * p_modulate.g, ci->modulate.b * p_modulate.b, ci->modulate.a * p_modulate.a);
if (modulate.a < 0.007) {
return;
}
if (ci->children_order_dirty) {
ci->child_items.sort_custom<ItemIndexSort>();
ci->children_order_dirty = false;
}
Item **child_items = ci->child_items.ptrw();
if (ci->clip) {
if (p_canvas_clip != nullptr) {
ci->final_clip_rect = p_canvas_clip->final_clip_rect.clip(global_rect);
} else {
ci->final_clip_rect = global_rect;
}
ci->final_clip_rect.position = ci->final_clip_rect.position.round();
ci->final_clip_rect.size = ci->final_clip_rect.size.round();
ci->final_clip_owner = ci;
} else {
ci->final_clip_owner = p_canvas_clip;
}
if (ci->sort_y) {
if (ci->ysort_children_count == -1) {
ci->ysort_children_count = 0;
_collect_ysort_children(ci, Transform2D(), p_material_owner, Color(1, 1, 1, 1), nullptr, ci->ysort_children_count);
}
child_item_count = ci->ysort_children_count;
// NOTE : Use of alloca here in a recursive function could make it susceptible to stack overflow.
// This was present in the original Item code. Consider changing to make safer.
child_items = (Item **)alloca(child_item_count * sizeof(Item *));
int i = 0;
_collect_ysort_children(ci, Transform2D(), p_material_owner, Color(1, 1, 1, 1), child_items, i);
SortArray<Item *, ItemPtrSort> sorter;
sorter.sort(child_items, child_item_count);
}
if (ci->z_relative) {
p_z = CLAMP(p_z + ci->z_index, RS::CANVAS_ITEM_Z_MIN, RS::CANVAS_ITEM_Z_MAX);
} else {
p_z = ci->z_index;
}
for (int i = 0; i < child_item_count; i++) {
if (!child_items[i]->behind || (ci->sort_y && child_items[i]->sort_y)) {
continue;
}
if (ci->sort_y) {
_render_canvas_item_cull_by_node(child_items[i], final_xform * child_items[i]->ysort_xform, p_clip_rect, modulate * child_items[i]->ysort_modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, (Item *)child_items[i]->material_owner, p_enclosed);
} else {
_render_canvas_item_cull_by_node(child_items[i], final_xform, p_clip_rect, modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, p_material_owner, p_enclosed);
}
}
if (ci->copy_back_buffer) {
ci->copy_back_buffer->screen_rect = final_xform.xform(ci->copy_back_buffer->rect).clip(p_clip_rect);
}
// something to draw?
if (item_is_visible) {
// Note : This has been moved to inside the (item_is_visible) check.
// It was OUTSIDE in the item culled code, which I suspect was incorrect.
// A redraw should not be issued if an item is not on screen?
// Even so, watch for regressions here.
if (ci->update_when_visible) {
RenderingServerRaster::redraw_request(false);
}
// Note we have already stored ci->final_transform
// and ci->global_rect_cache, and made sure these are up to date.
ci->final_modulate = Color(modulate.r * ci->self_modulate.r, modulate.g * ci->self_modulate.g, modulate.b * ci->self_modulate.b, modulate.a * ci->self_modulate.a);
ci->light_masked = false;
int zidx = p_z - RS::CANVAS_ITEM_Z_MIN;
if (z_last_list[zidx]) {
z_last_list[zidx]->next = ci;
z_last_list[zidx] = ci;
} else {
z_list[zidx] = ci;
z_last_list[zidx] = ci;
}
ci->next = nullptr;
}
for (int i = 0; i < child_item_count; i++) {
if (child_items[i]->behind || (ci->sort_y && child_items[i]->sort_y)) {
continue;
}
if (ci->sort_y) {
_render_canvas_item_cull_by_node(child_items[i], final_xform * child_items[i]->ysort_xform, p_clip_rect, modulate * child_items[i]->ysort_modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, (Item *)child_items[i]->material_owner, p_enclosed);
} else {
_render_canvas_item_cull_by_node(child_items[i], final_xform, p_clip_rect, modulate, p_z, z_list, z_last_list, (Item *)ci->final_clip_owner, p_material_owner, p_enclosed);
}
}
}
void RenderingServerCanvas::render_canvas(Canvas *p_canvas, const Transform2D &p_transform, const Rect2 &p_clip_rect, int p_canvas_layer_id) {
RSG::canvas_render->canvas_begin();
if (p_canvas->children_order_dirty) {
p_canvas->child_items.sort();
p_canvas->children_order_dirty = false;
}
int l = p_canvas->child_items.size();
Canvas::ChildItem *ci = p_canvas->child_items.ptrw();
bool has_mirror = false;
for (int i = 0; i < l; i++) {
if (ci[i].mirror.x || ci[i].mirror.y) {
has_mirror = true;
break;
}
}
if (!has_mirror) {
static const int z_range = RS::CANVAS_ITEM_Z_MAX - RS::CANVAS_ITEM_Z_MIN + 1;
RasterizerCanvas::Item *z_list[z_range];
RasterizerCanvas::Item *z_last_list[z_range];
memset(z_list, 0, z_range * sizeof(RasterizerCanvas::Item *));
memset(z_last_list, 0, z_range * sizeof(RasterizerCanvas::Item *));
_current_camera_transform = p_transform;
#ifdef RENDERING_SERVER_CANVAS_TIME_NODE_CULLING
bool measure = (Engine::get_singleton()->get_frames_drawn() % 100) == 0;
measure &= !Engine::get_singleton()->is_editor_hint();
if (measure) {
uint64_t totalA = 0;
uint64_t totalB = 0;
for (int i = 0; i < l; i++) {
uint64_t beforeB = OS::get_singleton()->get_ticks_usec();
_render_canvas_item_cull_by_item(ci[i].item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr);
uint64_t afterB = OS::get_singleton()->get_ticks_usec();
uint64_t beforeA = OS::get_singleton()->get_ticks_usec();
_prepare_tree_bounds(ci[i].item);
_render_canvas_item_cull_by_node(ci[i].item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr, false);
uint64_t afterA = OS::get_singleton()->get_ticks_usec();
totalA += afterA - beforeA;
totalB += afterB - beforeB;
} // for i
print_line("old : " + itos(totalB) + ", new : " + itos(totalA));
} // if measure
else {
#else
{
#endif
if (_canvas_cull_mode == CANVAS_CULL_MODE_NODE) {
for (int i = 0; i < l; i++) {
_prepare_tree_bounds(ci[i].item);
_render_canvas_item_cull_by_node(ci[i].item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr, false);
}
} else {
for (int i = 0; i < l; i++) {
_render_canvas_item_cull_by_item(ci[i].item, p_transform, p_clip_rect, Color(1, 1, 1, 1), 0, z_list, z_last_list, nullptr, nullptr);
}
}
} // if not measure
RSG::canvas_render->canvas_render_items_begin(p_canvas->modulate, p_transform);
for (int i = 0; i < z_range; i++) {
if (!z_list[i]) {
continue;
}
RSG::canvas_render->canvas_render_items(z_list[i], RS::CANVAS_ITEM_Z_MIN + i, p_canvas->modulate, p_transform);
}
RSG::canvas_render->canvas_render_items_end();
} else {
for (int i = 0; i < l; i++) {
const Canvas::ChildItem &ci2 = p_canvas->child_items[i];
_render_canvas_item_tree(ci2.item, p_transform, p_clip_rect, p_canvas->modulate);
//mirroring (useful for scrolling backgrounds)
if (ci2.mirror.x != 0) {
Transform2D xform2 = p_transform * Transform2D(0, Vector2(ci2.mirror.x, 0));
_render_canvas_item_tree(ci2.item, xform2, p_clip_rect, p_canvas->modulate);
}
if (ci2.mirror.y != 0) {
Transform2D xform2 = p_transform * Transform2D(0, Vector2(0, ci2.mirror.y));
_render_canvas_item_tree(ci2.item, xform2, p_clip_rect, p_canvas->modulate);
}
if (ci2.mirror.y != 0 && ci2.mirror.x != 0) {
Transform2D xform2 = p_transform * Transform2D(0, ci2.mirror);
_render_canvas_item_tree(ci2.item, xform2, p_clip_rect, p_canvas->modulate);
}
}
}
RSG::canvas_render->canvas_end();
}
RID RenderingServerCanvas::canvas_create() {
Canvas *canvas = memnew(Canvas);
ERR_FAIL_COND_V(!canvas, RID());
RID rid = canvas_owner.make_rid(canvas);
return rid;
}
void RenderingServerCanvas::canvas_set_item_mirroring(RID p_canvas, RID p_item, const Point2 &p_mirroring) {
Canvas *canvas = canvas_owner.getornull(p_canvas);
ERR_FAIL_COND(!canvas);
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
int idx = canvas->find_item(canvas_item);
ERR_FAIL_COND(idx == -1);
canvas->child_items.write[idx].mirror = p_mirroring;
}
void RenderingServerCanvas::canvas_set_modulate(RID p_canvas, const Color &p_color) {
Canvas *canvas = canvas_owner.get(p_canvas);
ERR_FAIL_COND(!canvas);
canvas->modulate = p_color;
}
void RenderingServerCanvas::canvas_set_disable_scale(bool p_disable) {
disable_scale = p_disable;
}
void RenderingServerCanvas::canvas_set_parent(RID p_canvas, RID p_parent, float p_scale) {
Canvas *canvas = canvas_owner.get(p_canvas);
ERR_FAIL_COND(!canvas);
canvas->parent = p_parent;
canvas->parent_scale = p_scale;
}
RID RenderingServerCanvas::canvas_item_create() {
Item *canvas_item = memnew(Item);
ERR_FAIL_COND_V(!canvas_item, RID());
return canvas_item_owner.make_rid(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_name(RID p_item, String p_name) {
#ifdef RENDERING_SERVER_CANVAS_DEBUG_ITEM_NAMES
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->name = p_name;
#endif
}
void RenderingServerCanvas::canvas_item_set_parent(RID p_item, RID p_parent) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
// dirty the item and any previous parents
_make_bound_dirty(canvas_item);
if (canvas_item->parent.is_valid()) {
if (canvas_owner.owns(canvas_item->parent)) {
Canvas *canvas = canvas_owner.get(canvas_item->parent);
canvas->erase_item(canvas_item);
} else if (canvas_item_owner.owns(canvas_item->parent)) {
Item *item_owner = canvas_item_owner.get(canvas_item->parent);
item_owner->child_items.erase(canvas_item);
if (item_owner->sort_y) {
_mark_ysort_dirty(item_owner, canvas_item_owner);
}
}
canvas_item->parent = RID();
}
if (p_parent.is_valid()) {
if (canvas_owner.owns(p_parent)) {
Canvas *canvas = canvas_owner.get(p_parent);
Canvas::ChildItem ci;
ci.item = canvas_item;
canvas->child_items.push_back(ci);
canvas->children_order_dirty = true;
_make_bound_dirty(canvas_item);
} else if (canvas_item_owner.owns(p_parent)) {
Item *item_owner = canvas_item_owner.get(p_parent);
item_owner->child_items.push_back(canvas_item);
item_owner->children_order_dirty = true;
if (item_owner->sort_y) {
_mark_ysort_dirty(item_owner, canvas_item_owner);
}
// keep the integrity of the bounds when adding to avoid false
// warning flags, by forcing the added child to be dirty
canvas_item->bound_dirty = false;
canvas_item->parent = p_parent;
_make_bound_dirty_reparent(canvas_item);
} else {
ERR_FAIL_MSG("Invalid parent.");
}
}
canvas_item->parent = p_parent;
_check_bound_integrity(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_visible(RID p_item, bool p_visible) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
// check for noop
if (p_visible != canvas_item->visible) {
canvas_item->visible = p_visible;
_make_bound_dirty(canvas_item, true);
}
// could this be enclosed in the noop? not sure
_mark_ysort_dirty(canvas_item, canvas_item_owner);
}
void RenderingServerCanvas::canvas_item_set_light_mask(RID p_item, int p_mask) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->light_mask = p_mask;
_check_bound_integrity(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_transform(RID p_item, const Transform2D &p_transform) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
if (_interpolation_data.interpolation_enabled && canvas_item->interpolated) {
if (!canvas_item->on_interpolate_transform_list) {
_interpolation_data.canvas_item_transform_update_list_curr->push_back(p_item);
canvas_item->on_interpolate_transform_list = true;
} else {
DEV_ASSERT(_interpolation_data.canvas_item_transform_update_list_curr->size());
}
}
if (_interpolation_data.interpolation_enabled && canvas_item->interpolated) {
if (!canvas_item->on_interpolate_transform_list) {
_interpolation_data.canvas_item_transform_update_list_curr->push_back(p_item);
canvas_item->on_interpolate_transform_list = true;
} else {
DEV_ASSERT(_interpolation_data.canvas_item_transform_update_list_curr->size());
}
}
canvas_item->xform_curr = p_transform;
// Special case!
// Modifying the transform DOES NOT affect the local bound.
// It only affects the local bound of the PARENT node (if there is one).
if (canvas_item_owner.owns(canvas_item->parent)) {
Item *canvas_item_parent = canvas_item_owner.get(canvas_item->parent);
_make_bound_dirty(canvas_item_parent);
}
}
void RenderingServerCanvas::canvas_item_set_clip(RID p_item, bool p_clip) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->clip = p_clip;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_distance_field_mode(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->distance_field = p_enable;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_custom_rect(RID p_item, bool p_custom_rect, const Rect2 &p_rect) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->custom_rect = p_custom_rect;
canvas_item->rect = p_rect;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_modulate(RID p_item, const Color &p_color) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->modulate = p_color;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_self_modulate(RID p_item, const Color &p_color) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->self_modulate = p_color;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_draw_behind_parent(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->behind = p_enable;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_use_identity_transform(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->use_identity_xform = p_enable;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_update_when_visible(RID p_item, bool p_update) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->update_when_visible = p_update;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_line(RID p_item, const Point2 &p_from, const Point2 &p_to, const Color &p_color, float p_width, bool p_antialiased) {
// Try drawing as a poly, because polys are batched and thus should run faster than thick lines,
// which run extremely slowly.
if (!p_antialiased && (p_width > 1.0)) {
// use poly drawing, as it is faster as it can use batching
static Vector<Point2> points;
static Vector<Color> colors;
static Vector<Point2> uvs;
if (points.size() != 4) {
// this should only be done once at runtime due to use of a static
points.resize(4);
colors.resize(4);
uvs.resize(4);
}
Vector2 side = p_to - p_from;
real_t length = side.length();
if (length == 0.0) {
// Not sure yet whether zero length is a noop operation later on,
// watch for visual errors. If there are visual errors, pass through
// to the line drawing routine below.
return;
}
// normalize
side /= length;
// 90 degrees
side = Vector2(-side.y, side.x);
side *= p_width * 0.5;
points.set(0, p_from + side);
points.set(1, p_from - side);
points.set(2, p_to - side);
points.set(3, p_to + side);
for (int n = 0; n < 4; n++) {
colors.set(n, p_color);
}
canvas_item_add_polygon(p_item, points, colors, uvs, RID(), RID(), false);
return;
}
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandLine *line = memnew(Item::CommandLine);
ERR_FAIL_COND(!line);
line->color = p_color;
line->from = p_from;
line->to = p_to;
line->width = p_width;
line->antialiased = p_antialiased;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(line);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_polyline(RID p_item, const Vector<Point2> &p_points, const Vector<Color> &p_colors, float p_width, bool p_antialiased) {
ERR_FAIL_COND(p_points.size() < 2);
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandPolyLine *pline = memnew(Item::CommandPolyLine);
ERR_FAIL_COND(!pline);
pline->antialiased = p_antialiased;
pline->multiline = false;
if (p_width <= 1) {
pline->lines = p_points;
pline->line_colors = p_colors;
if (pline->line_colors.size() == 0) {
pline->line_colors.push_back(Color(1, 1, 1, 1));
} else if (pline->line_colors.size() > 1 && pline->line_colors.size() != pline->lines.size()) {
pline->line_colors.resize(1);
}
} else {
//make a trianglestrip for drawing the line...
Vector2 prev_t;
pline->triangles.resize(p_points.size() * 2);
if (p_antialiased) {
pline->lines.resize(p_points.size() * 2);
}
if (p_colors.size() == 0) {
pline->triangle_colors.push_back(Color(1, 1, 1, 1));
if (p_antialiased) {
pline->line_colors.push_back(Color(1, 1, 1, 1));
}
} else if (p_colors.size() == 1) {
pline->triangle_colors = p_colors;
pline->line_colors = p_colors;
} else {
if (p_colors.size() != p_points.size()) {
pline->triangle_colors.push_back(p_colors[0]);
pline->line_colors.push_back(p_colors[0]);
} else {
pline->triangle_colors.resize(pline->triangles.size());
pline->line_colors.resize(pline->lines.size());
}
}
for (int i = 0; i < p_points.size(); i++) {
Vector2 t;
if (i == p_points.size() - 1) {
t = prev_t;
} else {
t = (p_points[i + 1] - p_points[i]).normalized().tangent();
if (i == 0) {
prev_t = t;
}
}
Vector2 tangent = ((t + prev_t).normalized()) * p_width * 0.5;
if (p_antialiased) {
pline->lines.write[i] = p_points[i] + tangent;
pline->lines.write[p_points.size() * 2 - i - 1] = p_points[i] - tangent;
if (pline->line_colors.size() > 1) {
pline->line_colors.write[i] = p_colors[i];
pline->line_colors.write[p_points.size() * 2 - i - 1] = p_colors[i];
}
}
pline->triangles.write[i * 2 + 0] = p_points[i] + tangent;
pline->triangles.write[i * 2 + 1] = p_points[i] - tangent;
if (pline->triangle_colors.size() > 1) {
pline->triangle_colors.write[i * 2 + 0] = p_colors[i];
pline->triangle_colors.write[i * 2 + 1] = p_colors[i];
}
prev_t = t;
}
}
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(pline);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_multiline(RID p_item, const Vector<Point2> &p_points, const Vector<Color> &p_colors, float p_width, bool p_antialiased) {
ERR_FAIL_COND(p_points.size() < 2);
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandPolyLine *pline = memnew(Item::CommandPolyLine);
ERR_FAIL_COND(!pline);
pline->antialiased = false; //todo
pline->multiline = true;
pline->lines = p_points;
pline->line_colors = p_colors;
if (pline->line_colors.size() == 0) {
pline->line_colors.push_back(Color(1, 1, 1, 1));
} else if (pline->line_colors.size() > 1 && pline->line_colors.size() != pline->lines.size()) {
pline->line_colors.resize(1);
}
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(pline);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_rect(RID p_item, const Rect2 &p_rect, const Color &p_color) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandRect *rect = memnew(Item::CommandRect);
ERR_FAIL_COND(!rect);
rect->modulate = p_color;
rect->rect = p_rect;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(rect);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_circle(RID p_item, const Point2 &p_pos, float p_radius, const Color &p_color) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandCircle *circle = memnew(Item::CommandCircle);
ERR_FAIL_COND(!circle);
circle->color = p_color;
circle->pos = p_pos;
circle->radius = p_radius;
canvas_item->commands.push_back(circle);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_texture_rect(RID p_item, const Rect2 &p_rect, RID p_texture, bool p_tile, const Color &p_modulate, bool p_transpose, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandRect *rect = memnew(Item::CommandRect);
ERR_FAIL_COND(!rect);
rect->modulate = p_modulate;
rect->rect = p_rect;
rect->flags = 0;
if (p_tile) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_TILE;
rect->flags |= RasterizerCanvas::CANVAS_RECT_REGION;
rect->source = Rect2(0, 0, fabsf(p_rect.size.width), fabsf(p_rect.size.height));
}
if (p_rect.size.x < 0) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_FLIP_H;
rect->rect.size.x = -rect->rect.size.x;
}
if (p_rect.size.y < 0) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_FLIP_V;
rect->rect.size.y = -rect->rect.size.y;
}
if (p_transpose) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_TRANSPOSE;
SWAP(rect->rect.size.x, rect->rect.size.y);
}
rect->texture = p_texture;
rect->normal_map = p_normal_map;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(rect);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_texture_multirect_region(RID p_item, const Vector<Rect2> &p_rects, RID p_texture, const Vector<Rect2> &p_src_rects, const Color &p_modulate, uint32_t p_canvas_rect_flags, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
ERR_FAIL_COND(p_rects.size() != p_src_rects.size());
ERR_FAIL_COND(!p_rects.size());
Item::CommandMultiRect *rect = memnew(Item::CommandMultiRect);
ERR_FAIL_COND(!rect);
rect->modulate = p_modulate;
rect->texture = p_texture;
rect->normal_map = p_normal_map;
// Rects should have flips and transposes pre-applied, and the relevant
// flags added to p_canvas_rect_flags.
// A single Multirect should contain rects ALL of the same flag type.
// The idea is to simplify the renderer as much as possible, and push the complexity
// to the one off creation code.
rect->flags = p_canvas_rect_flags | RasterizerCanvas::CANVAS_RECT_REGION;
rect->rects = p_rects;
rect->sources = p_src_rects;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(rect);
}
void RenderingServerCanvas::canvas_item_add_texture_rect_region(RID p_item, const Rect2 &p_rect, RID p_texture, const Rect2 &p_src_rect, const Color &p_modulate, bool p_transpose, RID p_normal_map, bool p_clip_uv) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandRect *rect = memnew(Item::CommandRect);
ERR_FAIL_COND(!rect);
rect->modulate = p_modulate;
rect->rect = p_rect;
rect->texture = p_texture;
rect->normal_map = p_normal_map;
rect->source = p_src_rect;
rect->flags = RasterizerCanvas::CANVAS_RECT_REGION;
if (p_rect.size.x < 0) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_FLIP_H;
rect->rect.size.x = -rect->rect.size.x;
}
if (p_src_rect.size.x < 0) {
rect->flags ^= RasterizerCanvas::CANVAS_RECT_FLIP_H;
rect->source.size.x = -rect->source.size.x;
}
if (p_rect.size.y < 0) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_FLIP_V;
rect->rect.size.y = -rect->rect.size.y;
}
if (p_src_rect.size.y < 0) {
rect->flags ^= RasterizerCanvas::CANVAS_RECT_FLIP_V;
rect->source.size.y = -rect->source.size.y;
}
if (p_transpose) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_TRANSPOSE;
SWAP(rect->rect.size.x, rect->rect.size.y);
}
if (p_clip_uv) {
rect->flags |= RasterizerCanvas::CANVAS_RECT_CLIP_UV;
}
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(rect);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_nine_patch(RID p_item, const Rect2 &p_rect, const Rect2 &p_source, RID p_texture, const Vector2 &p_topleft, const Vector2 &p_bottomright, RS::NinePatchAxisMode p_x_axis_mode, RS::NinePatchAxisMode p_y_axis_mode, bool p_draw_center, const Color &p_modulate, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandNinePatch *style = memnew(Item::CommandNinePatch);
ERR_FAIL_COND(!style);
style->texture = p_texture;
style->normal_map = p_normal_map;
style->rect = p_rect;
style->source = p_source;
style->draw_center = p_draw_center;
style->color = p_modulate;
style->margin[MARGIN_LEFT] = p_topleft.x;
style->margin[MARGIN_TOP] = p_topleft.y;
style->margin[MARGIN_RIGHT] = p_bottomright.x;
style->margin[MARGIN_BOTTOM] = p_bottomright.y;
style->axis_x = p_x_axis_mode;
style->axis_y = p_y_axis_mode;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(style);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_primitive(RID p_item, const Vector<Point2> &p_points, const Vector<Color> &p_colors, const Vector<Point2> &p_uvs, RID p_texture, float p_width, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandPrimitive *prim = memnew(Item::CommandPrimitive);
ERR_FAIL_COND(!prim);
prim->texture = p_texture;
prim->normal_map = p_normal_map;
prim->points = p_points;
prim->uvs = p_uvs;
prim->colors = p_colors;
prim->width = p_width;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(prim);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_polygon(RID p_item, const Vector<Point2> &p_points, const Vector<Color> &p_colors, const Vector<Point2> &p_uvs, RID p_texture, RID p_normal_map, bool p_antialiased) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
#ifdef DEBUG_ENABLED
int pointcount = p_points.size();
ERR_FAIL_COND(pointcount < 3);
int color_size = p_colors.size();
int uv_size = p_uvs.size();
ERR_FAIL_COND(color_size != 0 && color_size != 1 && color_size != pointcount);
ERR_FAIL_COND(uv_size != 0 && (uv_size != pointcount));
#endif
Vector<int> indices = Geometry::triangulate_polygon(p_points);
ERR_FAIL_COND_MSG(indices.empty(), "Invalid polygon data, triangulation failed.");
Item::CommandPolygon *polygon = memnew(Item::CommandPolygon);
ERR_FAIL_COND(!polygon);
polygon->texture = p_texture;
polygon->normal_map = p_normal_map;
polygon->points = p_points;
polygon->uvs = p_uvs;
polygon->colors = p_colors;
polygon->indices = indices;
polygon->count = indices.size();
polygon->antialiased = p_antialiased;
polygon->antialiasing_use_indices = false;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(polygon);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_triangle_array(RID p_item, const Vector<int> &p_indices, const Vector<Point2> &p_points, const Vector<Color> &p_colors, const Vector<Point2> &p_uvs, const Vector<int> &p_bones, const Vector<float> &p_weights, RID p_texture, int p_count, RID p_normal_map, bool p_antialiased, bool p_antialiasing_use_indices) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
int vertex_count = p_points.size();
ERR_FAIL_COND(vertex_count == 0);
ERR_FAIL_COND(!p_colors.empty() && p_colors.size() != vertex_count && p_colors.size() != 1);
ERR_FAIL_COND(!p_uvs.empty() && p_uvs.size() != vertex_count);
ERR_FAIL_COND(!p_bones.empty() && p_bones.size() != vertex_count * 4);
ERR_FAIL_COND(!p_weights.empty() && p_weights.size() != vertex_count * 4);
const Vector<int> &indices = p_indices;
int count = p_count * 3;
if (indices.empty()) {
ERR_FAIL_COND(vertex_count % 3 != 0);
if (p_count == -1) {
count = vertex_count;
}
} else {
ERR_FAIL_COND(indices.size() % 3 != 0);
if (p_count == -1) {
count = indices.size();
}
}
Item::CommandPolygon *polygon = memnew(Item::CommandPolygon);
ERR_FAIL_COND(!polygon);
polygon->texture = p_texture;
polygon->normal_map = p_normal_map;
polygon->points = p_points;
polygon->uvs = p_uvs;
polygon->colors = p_colors;
polygon->bones = p_bones;
polygon->weights = p_weights;
polygon->indices = indices;
polygon->count = count;
polygon->antialiased = p_antialiased;
polygon->antialiasing_use_indices = p_antialiasing_use_indices;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(polygon);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_set_transform(RID p_item, const Transform2D &p_transform) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandTransform *tr = memnew(Item::CommandTransform);
ERR_FAIL_COND(!tr);
tr->xform = p_transform;
canvas_item->commands.push_back(tr);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_mesh(RID p_item, const RID &p_mesh, const Transform2D &p_transform, const Color &p_modulate, RID p_texture, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandMesh *m = memnew(Item::CommandMesh);
ERR_FAIL_COND(!m);
m->mesh = p_mesh;
m->texture = p_texture;
m->normal_map = p_normal_map;
m->transform = p_transform;
m->modulate = p_modulate;
canvas_item->commands.push_back(m);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_add_multimesh(RID p_item, RID p_mesh, RID p_texture, RID p_normal_map) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandMultiMesh *mm = memnew(Item::CommandMultiMesh);
ERR_FAIL_COND(!mm);
mm->multimesh = p_mesh;
mm->texture = p_texture;
mm->normal_map = p_normal_map;
mm->canvas_item = p_item;
canvas_item->rect_dirty = true;
canvas_item->commands.push_back(mm);
_make_bound_dirty(canvas_item);
// Attach to multimesh a backlink to enable updating
// the canvas item local bound when the multimesh changes.
if (p_mesh.is_valid()) {
RSG::storage->multimesh_attach_canvas_item(p_mesh, p_item, true);
}
}
void RenderingServerCanvas::canvas_item_add_clip_ignore(RID p_item, bool p_ignore) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
Item::CommandClipIgnore *ci = memnew(Item::CommandClipIgnore);
ERR_FAIL_COND(!ci);
ci->ignore = p_ignore;
canvas_item->commands.push_back(ci);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_sort_children_by_y(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->sort_y = p_enable;
_mark_ysort_dirty(canvas_item, canvas_item_owner);
_check_bound_integrity(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_z_index(RID p_item, int p_z) {
ERR_FAIL_COND(p_z < RS::CANVAS_ITEM_Z_MIN || p_z > RS::CANVAS_ITEM_Z_MAX);
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->z_index = p_z;
_check_bound_integrity(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_z_as_relative_to_parent(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->z_relative = p_enable;
_check_bound_integrity(canvas_item);
}
Rect2 RenderingServerCanvas::_debug_canvas_item_get_rect(RID p_item) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND_V(!canvas_item, Rect2());
return canvas_item->get_rect();
}
Rect2 RenderingServerCanvas::_debug_canvas_item_get_local_bound(RID p_item) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND_V(!canvas_item, Rect2());
return canvas_item->local_bound;
}
void RenderingServerCanvas::canvas_item_set_skeleton_relative_xform(RID p_item, Transform2D p_relative_xform) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
if (!canvas_item->skinning_data) {
canvas_item->skinning_data = memnew(Item::SkinningData);
}
canvas_item->skinning_data->skeleton_relative_xform = p_relative_xform;
canvas_item->skinning_data->skeleton_relative_xform_inv = p_relative_xform.affine_inverse();
// Set any Polygon2Ds pre-calced bone bounds to dirty.
for (int n = 0; n < canvas_item->commands.size(); n++) {
Item::Command *c = canvas_item->commands[n];
if (c->type == Item::Command::TYPE_POLYGON) {
Item::CommandPolygon *polygon = static_cast<Item::CommandPolygon *>(c);
// Make sure skinning data is present.
if (!polygon->skinning_data) {
polygon->skinning_data = memnew(Item::CommandPolygon::SkinningData);
}
polygon->skinning_data->dirty = true;
}
}
}
void RenderingServerCanvas::canvas_item_attach_skeleton(RID p_item, RID p_skeleton) {
}
// Canvas items may contain references to other resources (such as MultiMesh).
// If the resources are deleted first, and the canvas_item retains references, it
// will crash / error when it tries to access these.
void RenderingServerCanvas::_canvas_item_remove_references(RID p_item, RID p_rid) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->remove_references(p_rid);
}
void RenderingServerCanvas::_canvas_item_invalidate_local_bound(RID p_item) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_copy_to_backbuffer(RID p_item, bool p_enable, const Rect2 &p_rect) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
if (bool(canvas_item->copy_back_buffer != nullptr) != p_enable) {
if (p_enable) {
canvas_item->copy_back_buffer = memnew(RasterizerCanvas::Item::CopyBackBuffer);
} else {
memdelete(canvas_item->copy_back_buffer);
canvas_item->copy_back_buffer = nullptr;
}
}
if (p_enable) {
canvas_item->copy_back_buffer->rect = p_rect;
canvas_item->copy_back_buffer->full = p_rect == Rect2();
}
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_clear(RID p_item) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
_make_bound_dirty(canvas_item);
canvas_item->clear();
}
void RenderingServerCanvas::canvas_item_set_draw_index(RID p_item, int p_index) {
Item *canvas_item = canvas_item_owner.getornull(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->index = p_index;
if (canvas_item_owner.owns(canvas_item->parent)) {
Item *canvas_item_parent = canvas_item_owner.getornull(canvas_item->parent);
canvas_item_parent->children_order_dirty = true;
return;
}
Canvas *canvas = canvas_owner.getornull(canvas_item->parent);
if (canvas) {
canvas->children_order_dirty = true;
return;
}
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_material(RID p_item, RID p_material) {
Item *canvas_item = canvas_item_owner.get(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->material = p_material;
_make_bound_dirty(canvas_item);
}
void RenderingServerCanvas::canvas_item_set_use_parent_material(RID p_item, bool p_enable) {
Item *canvas_item = canvas_item_owner.get(p_item);
ERR_FAIL_COND(!canvas_item);
canvas_item->use_parent_material = p_enable;
_make_bound_dirty(canvas_item);
}
bool RenderingServerCanvas::free(RID p_rid) {
if (canvas_owner.owns(p_rid)) {
Canvas *canvas = canvas_owner.get(p_rid);
ERR_FAIL_COND_V(!canvas, false);
while (canvas->viewports.size()) {
RenderingServerViewport::Viewport *vp = RSG::viewport->viewport_owner.get(canvas->viewports.front()->get());
ERR_FAIL_COND_V(!vp, true);
RBMap<RID, RenderingServerViewport::Viewport::CanvasData>::Element *E = vp->canvas_map.find(p_rid);
ERR_FAIL_COND_V(!E, true);
vp->canvas_map.erase(p_rid);
canvas->viewports.erase(canvas->viewports.front());
}
for (int i = 0; i < canvas->child_items.size(); i++) {
canvas->child_items[i].item->parent = RID();
}
canvas_owner.free(p_rid);
memdelete(canvas);
} else if (canvas_item_owner.owns(p_rid)) {
Item *canvas_item = canvas_item_owner.get(p_rid);
ERR_FAIL_COND_V(!canvas_item, true);
_make_bound_dirty(canvas_item);
if (canvas_item->parent.is_valid()) {
if (canvas_owner.owns(canvas_item->parent)) {
Canvas *canvas = canvas_owner.get(canvas_item->parent);
canvas->erase_item(canvas_item);
} else if (canvas_item_owner.owns(canvas_item->parent)) {
Item *item_owner = canvas_item_owner.get(canvas_item->parent);
item_owner->child_items.erase(canvas_item);
if (item_owner->sort_y) {
_mark_ysort_dirty(item_owner, canvas_item_owner);
}
_check_bound_integrity(item_owner);
}
}
for (int i = 0; i < canvas_item->child_items.size(); i++) {
canvas_item->child_items[i]->parent = RID();
}
/*
if (canvas_item->material) {
canvas_item->material->owners.erase(canvas_item);
}
*/
canvas_item_owner.free(p_rid);
memdelete(canvas_item);
} else {
return false;
}
return true;
}
#ifdef RENDERING_SERVER_CANVAS_CHECK_BOUNDS
// Debugging function to check that the bound dirty flags in the tree make sense.
// Any item that has is dirty, all parents should be dirty up to the root
// (except in hidden branches, which are not kept track of for performance reasons).
bool RenderingServerCanvas::_check_bound_integrity(const Item *p_item) {
while (p_item) {
if (canvas_item_owner.owns(p_item->parent)) {
p_item = canvas_item_owner.get(p_item->parent);
} else {
return _check_bound_integrity_down(p_item, p_item->bound_dirty);
}
}
return true;
}
bool RenderingServerCanvas::_check_bound_integrity_down(const Item *p_item, bool p_bound_dirty) {
// don't care about integrity into invisible branches
if (!p_item->visible) {
return true;
}
if (p_item->bound_dirty) {
if (!p_bound_dirty) {
_print_tree(p_item);
ERR_PRINT("bound integrity check failed");
return false;
}
}
// go through children
int child_item_count = p_item->child_items.size();
Item *const *child_items = p_item->child_items.ptr();
for (int n = 0; n < child_item_count; n++) {
if (!_check_bound_integrity_down(child_items[n], p_bound_dirty)) {
return false;
}
}
return true;
}
void RenderingServerCanvas::_print_tree(const Item *p_item) {
const Item *highlight = p_item;
while (p_item) {
if (canvas_item_owner.owns(p_item->parent)) {
p_item = canvas_item_owner.get(p_item->parent);
} else {
_print_tree_down(0, 0, p_item, highlight);
return;
}
}
}
void RenderingServerCanvas::_print_tree_down(int p_child_id, int p_depth, const Item *p_item, const Item *p_highlight, bool p_hidden) {
String sz;
for (int n = 0; n < p_depth; n++) {
sz += "\t";
}
if (p_item == p_highlight) {
sz += "* ";
}
sz += itos(p_child_id) + " ";
#ifdef RENDERING_SERVER_CANVAS_DEBUG_ITEM_NAMES
sz += p_item->name + "\t";
#endif
sz += String(Variant(p_item->global_rect_cache)) + " ";
if (!p_item->visible) {
sz += "(H) ";
p_hidden = true;
} else if (p_hidden) {
sz += "(HI) ";
}
if (p_item->bound_dirty) {
sz += "(dirty) ";
}
if (p_item->parent == RID()) {
sz += "(parent NULL) ";
}
print_line(sz);
// go through children
int child_item_count = p_item->child_items.size();
Item *const *child_items = p_item->child_items.ptr();
for (int n = 0; n < child_item_count; n++) {
_print_tree_down(n, p_depth + 1, child_items[n], p_highlight, p_hidden);
}
}
#endif
void RenderingServerCanvas::tick() {
if (_interpolation_data.interpolation_enabled) {
update_interpolation_tick(true);
}
}
void RenderingServerCanvas::update_interpolation_tick(bool p_process) {
#define PANDEMONIUM_UPDATE_INTERPOLATION_TICK(LIST_PREV, LIST_CURR, TYPE, OWNER_LIST) \
/* Detect any that were on the previous transform list that are no longer active. */ \
for (unsigned int n = 0; n < _interpolation_data.LIST_PREV->size(); n++) { \
const RID &rid = (*_interpolation_data.LIST_PREV)[n]; \
TYPE *item = OWNER_LIST.getornull(rid); \
/* no longer active? (either the instance deleted or no longer being transformed) */ \
if (item && !item->on_interpolate_transform_list) { \
item->xform_prev = item->xform_curr; \
} \
} \
/* and now for any in the transform list (being actively interpolated), */ \
/* keep the previous transform value up to date and ready for next tick */ \
if (p_process) { \
for (unsigned int n = 0; n < _interpolation_data.LIST_CURR->size(); n++) { \
const RID &rid = (*_interpolation_data.LIST_CURR)[n]; \
TYPE *item = OWNER_LIST.getornull(rid); \
if (item) { \
item->xform_prev = item->xform_curr; \
item->on_interpolate_transform_list = false; \
} \
} \
} \
SWAP(_interpolation_data.LIST_CURR, _interpolation_data.LIST_PREV); \
_interpolation_data.LIST_CURR->clear();
PANDEMONIUM_UPDATE_INTERPOLATION_TICK(canvas_item_transform_update_list_prev, canvas_item_transform_update_list_curr, Item, canvas_item_owner);
#undef PANDEMONIUM_UPDATE_INTERPOLATION_TICK
}
RenderingServerCanvas::RenderingServerCanvas() {
_canvas_cull_mode = CANVAS_CULL_MODE_NODE;
z_list = (RasterizerCanvas::Item **)memalloc(z_range * sizeof(RasterizerCanvas::Item *));
z_last_list = (RasterizerCanvas::Item **)memalloc(z_range * sizeof(RasterizerCanvas::Item *));
disable_scale = false;
int mode = GLOBAL_DEF("rendering/2d/options/culling_mode", 1);
ProjectSettings::get_singleton()->set_custom_property_info("rendering/2d/options/culling_mode", PropertyInfo(Variant::INT, "rendering/2d/options/culling_mode", PROPERTY_HINT_ENUM, "Item,Node"));
switch (mode) {
default: {
_canvas_cull_mode = CANVAS_CULL_MODE_NODE;
} break;
case 0: {
_canvas_cull_mode = CANVAS_CULL_MODE_ITEM;
} break;
}
}
RenderingServerCanvas::~RenderingServerCanvas() {
memfree(z_list);
memfree(z_last_list);
}
Reference in New Issue View Git Blame Copy Permalink