This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
blender-archive/source/blender/blenkernel/intern/displist.c

1869 lines
47 KiB
C
Raw Normal View History

2011-10-10 09:38:02 +00:00
/*
* ***** BEGIN GPL LICENSE BLOCK *****
2002-10-12 11:37:38 +00:00
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
2002-10-12 11:37:38 +00:00
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
2010-02-12 13:34:04 +00:00
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
2002-10-12 11:37:38 +00:00
*
* The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
* All rights reserved.
*
* The Original Code is: all of this file.
*
* Contributor(s): none yet.
*
* ***** END GPL LICENSE BLOCK *****
2002-10-12 11:37:38 +00:00
*/
2011-02-27 20:40:57 +00:00
/** \file blender/blenkernel/intern/displist.c
* \ingroup bke
*/
2002-10-12 11:37:38 +00:00
#include <math.h>
#include <stdio.h>
#include <string.h>
#include "MEM_guardedalloc.h"
#include "DNA_curve_types.h"
#include "DNA_scene_types.h"
#include "DNA_object_types.h"
#include "DNA_vfont_types.h"
2002-10-12 11:37:38 +00:00
#include "BLI_blenlib.h"
#include "BLI_memarena.h"
#include "BLI_math.h"
#include "BLI_scanfill.h"
#include "BLI_utildefines.h"
2002-10-12 11:37:38 +00:00
#include "BKE_global.h"
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
#include "BKE_depsgraph.h"
2002-10-12 11:37:38 +00:00
#include "BKE_displist.h"
#include "BKE_cdderivedmesh.h"
2002-10-12 11:37:38 +00:00
#include "BKE_object.h"
#include "BKE_mball.h"
#include "BKE_mball_tessellate.h"
2002-10-12 11:37:38 +00:00
#include "BKE_curve.h"
#include "BKE_key.h"
2002-10-12 11:37:38 +00:00
#include "BKE_anim.h"
#include "BKE_font.h"
#include "BKE_lattice.h"
#include "BKE_modifier.h"
2002-10-12 11:37:38 +00:00
#include "BLI_sys_types.h" // for intptr_t support
static void boundbox_displist_object(Object *ob);
2012-05-07 06:58:03 +00:00
void BKE_displist_elem_free(DispList *dl)
2002-10-12 11:37:38 +00:00
{
if (dl) {
if (dl->verts) MEM_freeN(dl->verts);
if (dl->nors) MEM_freeN(dl->nors);
if (dl->index) MEM_freeN(dl->index);
if (dl->bevel_split) MEM_freeN(dl->bevel_split);
2002-10-12 11:37:38 +00:00
MEM_freeN(dl);
}
}
2012-05-07 06:58:03 +00:00
void BKE_displist_free(ListBase *lb)
2002-10-12 11:37:38 +00:00
{
DispList *dl;
while ((dl = BLI_pophead(lb))) {
2012-05-07 06:58:03 +00:00
BKE_displist_elem_free(dl);
2002-10-12 11:37:38 +00:00
}
}
2012-05-07 06:58:03 +00:00
DispList *BKE_displist_find_or_create(ListBase *lb, int type)
2002-10-12 11:37:38 +00:00
{
DispList *dl;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = lb->first;
while (dl) {
2012-05-07 06:58:03 +00:00
if (dl->type == type)
return dl;
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "find_disp");
dl->type = type;
2002-10-12 11:37:38 +00:00
BLI_addtail(lb, dl);
return dl;
}
2012-05-07 06:58:03 +00:00
DispList *BKE_displist_find(ListBase *lb, int type)
2002-10-12 11:37:38 +00:00
{
DispList *dl;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = lb->first;
while (dl) {
2012-05-07 06:58:03 +00:00
if (dl->type == type)
return dl;
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
return NULL;
2002-10-12 11:37:38 +00:00
}
bool BKE_displist_has_faces(ListBase *lb)
{
DispList *dl;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
for (dl = lb->first; dl; dl = dl->next) {
if (ELEM(dl->type, DL_INDEX3, DL_INDEX4, DL_SURF)) {
return true;
}
}
2012-05-07 06:58:03 +00:00
return false;
}
2012-05-07 06:58:03 +00:00
void BKE_displist_copy(ListBase *lbn, ListBase *lb)
2002-10-12 11:37:38 +00:00
{
DispList *dln, *dl;
2012-05-07 06:58:03 +00:00
BKE_displist_free(lbn);
2012-05-06 17:22:54 +00:00
dl = lb->first;
while (dl) {
2012-05-06 17:22:54 +00:00
dln = MEM_dupallocN(dl);
2002-10-12 11:37:38 +00:00
BLI_addtail(lbn, dln);
2012-05-06 17:22:54 +00:00
dln->verts = MEM_dupallocN(dl->verts);
dln->nors = MEM_dupallocN(dl->nors);
dln->index = MEM_dupallocN(dl->index);
if (dl->bevel_split) {
dln->bevel_split = MEM_dupallocN(dl->bevel_split);
}
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
}
2012-05-07 06:58:03 +00:00
void BKE_displist_normals_add(ListBase *lb)
2002-10-12 11:37:38 +00:00
{
DispList *dl = NULL;
float *vdata, *ndata, nor[3];
float *v1, *v2, *v3, *v4;
float *n1, *n2, *n3, *n4;
int a, b, p1, p2, p3, p4;
2012-05-06 17:22:54 +00:00
dl = lb->first;
2012-05-07 06:58:03 +00:00
while (dl) {
2012-05-06 17:22:54 +00:00
if (dl->type == DL_INDEX3) {
if (dl->nors == NULL) {
dl->nors = MEM_callocN(sizeof(float) * 3, "dlnors");
2012-05-07 06:58:03 +00:00
if (dl->verts[2] < 0.0f)
dl->nors[2] = -1.0f;
else
dl->nors[2] = 1.0f;
2002-10-12 11:37:38 +00:00
}
}
2012-05-06 17:22:54 +00:00
else if (dl->type == DL_SURF) {
if (dl->nors == NULL) {
dl->nors = MEM_callocN(sizeof(float) * 3 * dl->nr * dl->parts, "dlnors");
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
vdata = dl->verts;
ndata = dl->nors;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
for (a = 0; a < dl->parts; a++) {
2012-05-07 06:58:03 +00:00
if (BKE_displist_surfindex_get(dl, a, &b, &p1, &p2, &p3, &p4) == 0)
break;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
v1 = vdata + 3 * p1;
n1 = ndata + 3 * p1;
v2 = vdata + 3 * p2;
n2 = ndata + 3 * p2;
v3 = vdata + 3 * p3;
n3 = ndata + 3 * p3;
v4 = vdata + 3 * p4;
n4 = ndata + 3 * p4;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
for (; b < dl->nr; b++) {
2012-04-29 15:47:02 +00:00
normal_quad_v3(nor, v1, v3, v4, v2);
2012-05-07 06:58:03 +00:00
add_v3_v3(n1, nor);
add_v3_v3(n2, nor);
add_v3_v3(n3, nor);
add_v3_v3(n4, nor);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
v2 = v1; v1 += 3;
v4 = v3; v3 += 3;
n2 = n1; n1 += 3;
n4 = n3; n3 += 3;
2002-10-12 11:37:38 +00:00
}
}
2012-05-06 17:22:54 +00:00
a = dl->parts * dl->nr;
v1 = ndata;
while (a--) {
normalize_v3(v1);
2012-05-06 17:22:54 +00:00
v1 += 3;
2002-10-12 11:37:38 +00:00
}
}
}
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
}
void BKE_displist_count(ListBase *lb, int *totvert, int *totface, int *tottri)
{
DispList *dl;
2012-05-07 06:58:03 +00:00
for (dl = lb->first; dl; dl = dl->next) {
int vert_tot = 0;
int face_tot = 0;
int tri_tot = 0;
switch (dl->type) {
case DL_SURF:
{
vert_tot = dl->nr * dl->parts;
face_tot = (dl->nr - 1) * (dl->parts - 1);
tri_tot = face_tot * 2;
break;
}
case DL_INDEX3:
{
vert_tot = dl->nr;
face_tot = dl->parts;
tri_tot = face_tot;
break;
}
case DL_INDEX4:
{
vert_tot = dl->nr;
face_tot = dl->parts;
tri_tot = face_tot * 2;
break;
}
case DL_POLY:
case DL_SEGM:
{
vert_tot = dl->nr * dl->parts;
2012-05-07 06:58:03 +00:00
break;
}
}
2012-05-07 06:58:03 +00:00
*totvert += vert_tot;
*totface += face_tot;
*tottri += tri_tot;
}
}
bool BKE_displist_surfindex_get(DispList *dl, int a, int *b, int *p1, int *p2, int *p3, int *p4)
{
2012-05-06 17:22:54 +00:00
if ((dl->flag & DL_CYCL_V) == 0 && a == (dl->parts) - 1) {
return false;
}
2012-05-07 06:58:03 +00:00
if (dl->flag & DL_CYCL_U) {
2012-05-06 17:22:54 +00:00
(*p1) = dl->nr * a;
(*p2) = (*p1) + dl->nr - 1;
(*p3) = (*p1) + dl->nr;
(*p4) = (*p2) + dl->nr;
(*b) = 0;
}
else {
2012-05-06 17:22:54 +00:00
(*p2) = dl->nr * a;
(*p1) = (*p2) + 1;
(*p4) = (*p2) + dl->nr;
(*p3) = (*p1) + dl->nr;
(*b) = 1;
}
2012-05-07 06:58:03 +00:00
if ((dl->flag & DL_CYCL_V) && a == dl->parts - 1) {
2012-05-06 17:22:54 +00:00
(*p3) -= dl->nr * dl->parts;
(*p4) -= dl->nr * dl->parts;
}
2012-05-07 06:58:03 +00:00
return true;
}
/* ****************** make displists ********************* */
2002-10-12 11:37:38 +00:00
2014-03-20 22:56:28 +11:00
static void curve_to_displist(Curve *cu, ListBase *nubase, ListBase *dispbase,
const bool for_render, const bool use_render_resolution)
2002-10-12 11:37:38 +00:00
{
Nurb *nu;
DispList *dl;
BezTriple *bezt, *prevbezt;
BPoint *bp;
float *data;
int a, len, resolu;
2014-03-20 22:56:28 +11:00
const bool editmode = (!for_render && (cu->editnurb || cu->editfont));
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
nu = nubase->first;
while (nu) {
2014-03-20 22:56:28 +11:00
if (nu->hide == 0 || editmode == false) {
if (use_render_resolution && cu->resolu_ren != 0)
2012-05-06 17:22:54 +00:00
resolu = cu->resolu_ren;
else
2012-05-06 17:22:54 +00:00
resolu = nu->resolu;
2012-05-07 06:58:03 +00:00
2012-10-07 09:48:59 +00:00
if (!BKE_nurb_check_valid_u(nu)) {
/* pass */
}
else if (nu->type == CU_BEZIER) {
/* count */
2012-05-06 17:22:54 +00:00
len = 0;
a = nu->pntsu - 1;
if (nu->flagu & CU_NURB_CYCLIC) a++;
2002-10-12 11:37:38 +00:00
2012-05-06 17:22:54 +00:00
prevbezt = nu->bezt;
bezt = prevbezt + 1;
while (a--) {
2012-05-07 06:58:03 +00:00
if (a == 0 && (nu->flagu & CU_NURB_CYCLIC))
bezt = nu->bezt;
if (prevbezt->h2 == HD_VECT && bezt->h1 == HD_VECT)
len++;
else
len += resolu;
if (a == 0 && (nu->flagu & CU_NURB_CYCLIC) == 0)
len++;
2012-05-06 17:22:54 +00:00
prevbezt = bezt;
2002-10-12 11:37:38 +00:00
bezt++;
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListbez");
/* len+1 because of 'forward_diff_bezier' function */
dl->verts = MEM_mallocN((len + 1) * sizeof(float[3]), "dlverts");
2002-10-12 11:37:38 +00:00
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
dl->parts = 1;
dl->nr = len;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
2002-10-12 11:37:38 +00:00
2012-05-06 17:22:54 +00:00
data = dl->verts;
2002-10-12 11:37:38 +00:00
/* check that (len != 2) so we don't immediately loop back on ourselves */
if (nu->flagu & CU_NURB_CYCLIC && (dl->nr != 2)) {
2012-05-06 17:22:54 +00:00
dl->type = DL_POLY;
a = nu->pntsu;
2002-10-12 11:37:38 +00:00
}
else {
2012-05-06 17:22:54 +00:00
dl->type = DL_SEGM;
a = nu->pntsu - 1;
2002-10-12 11:37:38 +00:00
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
prevbezt = nu->bezt;
bezt = prevbezt + 1;
2012-05-07 06:58:03 +00:00
while (a--) {
2012-05-07 06:58:03 +00:00
if (a == 0 && dl->type == DL_POLY)
bezt = nu->bezt;
2012-05-06 17:22:54 +00:00
if (prevbezt->h2 == HD_VECT && bezt->h1 == HD_VECT) {
2011-11-06 15:17:43 +00:00
copy_v3_v3(data, prevbezt->vec[1]);
2012-05-06 17:22:54 +00:00
data += 3;
2002-10-12 11:37:38 +00:00
}
else {
int j;
2012-05-06 17:22:54 +00:00
for (j = 0; j < 3; j++) {
BKE_curve_forward_diff_bezier(prevbezt->vec[1][j],
prevbezt->vec[2][j],
bezt->vec[0][j],
bezt->vec[1][j],
data + j, resolu, 3 * sizeof(float));
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
data += 3 * resolu;
2002-10-12 11:37:38 +00:00
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
if (a == 0 && dl->type == DL_SEGM) {
2011-11-06 15:17:43 +00:00
copy_v3_v3(data, bezt->vec[1]);
2002-10-12 11:37:38 +00:00
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
prevbezt = bezt;
2002-10-12 11:37:38 +00:00
bezt++;
}
}
else if (nu->type == CU_NURBS) {
2012-05-06 17:22:54 +00:00
len = (resolu * SEGMENTSU(nu));
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListsurf");
dl->verts = MEM_mallocN(len * sizeof(float[3]), "dlverts");
2002-10-12 11:37:38 +00:00
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
dl->parts = 1;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl->nr = len;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
2002-10-12 11:37:38 +00:00
2012-05-06 17:22:54 +00:00
data = dl->verts;
2012-05-07 06:58:03 +00:00
if (nu->flagu & CU_NURB_CYCLIC)
dl->type = DL_POLY;
2012-05-06 17:22:54 +00:00
else dl->type = DL_SEGM;
BKE_nurb_makeCurve(nu, data, NULL, NULL, NULL, resolu, 3 * sizeof(float));
2002-10-12 11:37:38 +00:00
}
else if (nu->type == CU_POLY) {
2012-05-06 17:22:54 +00:00
len = nu->pntsu;
dl = MEM_callocN(sizeof(DispList), "makeDispListpoly");
dl->verts = MEM_mallocN(len * sizeof(float[3]), "dlverts");
2002-10-12 11:37:38 +00:00
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
dl->parts = 1;
dl->nr = len;
dl->col = nu->mat_nr;
More text object fancyness, and fixes: - "Flush" is now split into two seperate Alignment modes "Flush" and "Justify": - Justify does exactly the same as a normal word processor's justify function does, and in addition, it uses *whitespace* instead of *character spacing* (kerning) to fill lines. Much more readable. - Flush is pretty much the old Blender "Flush" mode - and as such it uses character spacing to fill lines. Just as Justify, this only works with at least one textframe. - Underlining for text objects. Not a lot to explain. New button "U" in the editbuttons, and CTRL-U as hotkey toggle underlining for newly entered characters or for the selection, just like CTRL-B/CTRL-I do for bold/italic. Underline height (thickness) and Underline position (vertical) can be set in the editbuttons. Implemented as CU_POLY polygon curves. - The B, U and i buttons (and the corresponding CTRL-B/U/I keystrokes) have been fixed to only affect *one* attribute at a time. Formerly, hitting CTRL-B when no other style was active, on a text portion with italics text, for example, would kill the italics and just apply bold. Now, these attributes always add or substract only, but do not replace the style. - In the past, there were bugs with material indices uninitialized, and thus crashes in the renderer with illegal material indices. Even though I assume they have been fixed, I've put in a check that checks (hah) if the material index of a character is illegal (bigger than ob->totcol), and then sets it to zero, and spits out a warning on stderr. If you see such warnings, please report and link to the .blend. - Bugfix: All alignment modes only worked if there were at least *two* lines of text in the text object. Fixed There's now a regression test file for text objects, please add to the corresponding repository: http://blender.instinctive.de/downloads/release/demo/text-regression.blend.gz
2005-08-29 12:46:07 +00:00
dl->charidx = nu->charidx;
2002-10-12 11:37:38 +00:00
2012-05-06 17:22:54 +00:00
data = dl->verts;
if ((nu->flagu & CU_NURB_CYCLIC) && (dl->nr != 2)) {
dl->type = DL_POLY;
}
else {
dl->type = DL_SEGM;
}
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
a = len;
bp = nu->bp;
while (a--) {
2011-11-06 15:17:43 +00:00
copy_v3_v3(data, bp->vec);
2002-10-12 11:37:38 +00:00
bp++;
2012-05-06 17:22:54 +00:00
data += 3;
2002-10-12 11:37:38 +00:00
}
}
}
2012-05-06 17:22:54 +00:00
nu = nu->next;
2002-10-12 11:37:38 +00:00
}
}
/**
* \param normal_proj Optional normal thats used to project the scanfill verts into 2d coords.
* Pass this along if known since it saves time calculating the normal.
* \param flipnormal Flip the normal (same as passing \a normal_proj negated)
*/
void BKE_displist_fill(ListBase *dispbase, ListBase *to, const float normal_proj[3], const bool flipnormal)
2002-10-12 11:37:38 +00:00
{
ScanFillContext sf_ctx;
ScanFillVert *sf_vert, *sf_vert_new, *sf_vert_last;
ScanFillFace *sf_tri;
MemArena *sf_arena;
2012-05-06 17:22:54 +00:00
DispList *dlnew = NULL, *dl;
2002-10-12 11:37:38 +00:00
float *f1;
2012-05-06 17:22:54 +00:00
int colnr = 0, charidx = 0, cont = 1, tot, a, *index, nextcol = 0;
int totvert;
const int scanfill_flag = BLI_SCANFILL_CALC_REMOVE_DOUBLES | BLI_SCANFILL_CALC_POLYS | BLI_SCANFILL_CALC_HOLES;
2012-05-07 06:58:03 +00:00
if (dispbase == NULL)
return;
if (BLI_listbase_is_empty(dispbase))
2012-05-07 06:58:03 +00:00
return;
2002-10-12 11:37:38 +00:00
sf_arena = BLI_memarena_new(BLI_SCANFILL_ARENA_SIZE, __func__);
while (cont) {
2012-05-06 17:22:54 +00:00
cont = 0;
totvert = 0;
nextcol = 0;
2012-05-07 06:58:03 +00:00
BLI_scanfill_begin_arena(&sf_ctx, sf_arena);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = dispbase->first;
while (dl) {
2012-05-06 17:22:54 +00:00
if (dl->type == DL_POLY) {
2012-05-07 06:58:03 +00:00
if (charidx < dl->charidx)
cont = 1;
2012-05-06 17:22:54 +00:00
else if (charidx == dl->charidx) { /* character with needed index */
if (colnr == dl->col) {
sf_ctx.poly_nr++;
/* make editverts and edges */
2012-05-06 17:22:54 +00:00
f1 = dl->verts;
a = dl->nr;
sf_vert = sf_vert_new = NULL;
2012-05-07 06:58:03 +00:00
while (a--) {
sf_vert_last = sf_vert;
sf_vert = BLI_scanfill_vert_add(&sf_ctx, f1);
totvert++;
if (sf_vert_last == NULL)
sf_vert_new = sf_vert;
else {
BLI_scanfill_edge_add(&sf_ctx, sf_vert_last, sf_vert);
}
2012-05-06 17:22:54 +00:00
f1 += 3;
2002-10-12 11:37:38 +00:00
}
if (sf_vert != NULL && sf_vert_new != NULL) {
BLI_scanfill_edge_add(&sf_ctx, sf_vert, sf_vert_new);
}
}
2012-05-06 17:22:54 +00:00
else if (colnr < dl->col) {
/* got poly with next material at current char */
2012-05-06 17:22:54 +00:00
cont = 1;
nextcol = 1;
2002-10-12 11:37:38 +00:00
}
}
}
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
2012-05-07 06:58:03 +00:00
2015-06-17 07:06:59 +10:00
/* XXX (obedit && obedit->actcol) ? (obedit->actcol - 1) : 0)) { */
if (totvert && (tot = BLI_scanfill_calc_ex(&sf_ctx,
scanfill_flag,
normal_proj)))
{
if (tot) {
2012-05-06 17:22:54 +00:00
dlnew = MEM_callocN(sizeof(DispList), "filldisplist");
dlnew->type = DL_INDEX3;
dlnew->col = colnr;
dlnew->nr = totvert;
dlnew->parts = tot;
dlnew->index = MEM_mallocN(tot * 3 * sizeof(int), "dlindex");
dlnew->verts = MEM_mallocN(totvert * 3 * sizeof(float), "dlverts");
2012-05-07 06:58:03 +00:00
/* vert data */
2012-05-06 17:22:54 +00:00
f1 = dlnew->verts;
totvert = 0;
for (sf_vert = sf_ctx.fillvertbase.first; sf_vert; sf_vert = sf_vert->next) {
copy_v3_v3(f1, sf_vert->co);
2012-05-06 17:22:54 +00:00
f1 += 3;
2012-05-07 06:58:03 +00:00
/* index number */
sf_vert->tmp.i = totvert;
2002-10-12 11:37:38 +00:00
totvert++;
}
2012-05-07 06:58:03 +00:00
/* index data */
2012-05-06 17:22:54 +00:00
index = dlnew->index;
for (sf_tri = sf_ctx.fillfacebase.first; sf_tri; sf_tri = sf_tri->next) {
index[0] = sf_tri->v1->tmp.i;
index[1] = sf_tri->v2->tmp.i;
index[2] = sf_tri->v3->tmp.i;
if (flipnormal)
SWAP(int, index[0], index[2]);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
index += 3;
2002-10-12 11:37:38 +00:00
}
}
BLI_addhead(to, dlnew);
}
BLI_scanfill_end_arena(&sf_ctx, sf_arena);
2002-10-12 11:37:38 +00:00
if (nextcol) {
/* stay at current char but fill polys with next material */
colnr++;
}
else {
/* switch to next char and start filling from first material */
charidx++;
2012-05-06 17:22:54 +00:00
colnr = 0;
}
2002-10-12 11:37:38 +00:00
}
2012-05-07 06:58:03 +00:00
BLI_memarena_free(sf_arena);
/* do not free polys, needed for wireframe display */
2002-10-12 11:37:38 +00:00
}
static void bevels_to_filledpoly(Curve *cu, ListBase *dispbase)
{
const float z_up[3] = {0.0f, 0.0f, 1.0f};
2002-10-12 11:37:38 +00:00
ListBase front, back;
DispList *dl, *dlnew;
float *fp, *fp1;
int a, dpoly;
2012-05-07 06:58:03 +00:00
BLI_listbase_clear(&front);
BLI_listbase_clear(&back);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = dispbase->first;
while (dl) {
2012-05-06 17:22:54 +00:00
if (dl->type == DL_SURF) {
2012-05-07 06:58:03 +00:00
if ((dl->flag & DL_CYCL_V) && (dl->flag & DL_CYCL_U) == 0) {
2014-03-17 21:48:13 +11:00
if ((cu->flag & CU_BACK) && (dl->flag & DL_BACK_CURVE)) {
2012-05-06 17:22:54 +00:00
dlnew = MEM_callocN(sizeof(DispList), "filldisp");
2002-10-12 11:37:38 +00:00
BLI_addtail(&front, dlnew);
2012-05-06 17:22:54 +00:00
dlnew->verts = fp1 = MEM_mallocN(sizeof(float) * 3 * dl->parts, "filldisp1");
dlnew->nr = dl->parts;
dlnew->parts = 1;
dlnew->type = DL_POLY;
dlnew->col = dl->col;
dlnew->charidx = dl->charidx;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
fp = dl->verts;
dpoly = 3 * dl->nr;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
a = dl->parts;
while (a--) {
2011-11-06 15:17:43 +00:00
copy_v3_v3(fp1, fp);
2012-05-06 17:22:54 +00:00
fp1 += 3;
fp += dpoly;
2002-10-12 11:37:38 +00:00
}
}
2012-05-07 06:58:03 +00:00
if ((cu->flag & CU_FRONT) && (dl->flag & DL_FRONT_CURVE)) {
2012-05-06 17:22:54 +00:00
dlnew = MEM_callocN(sizeof(DispList), "filldisp");
2002-10-12 11:37:38 +00:00
BLI_addtail(&back, dlnew);
2012-05-06 17:22:54 +00:00
dlnew->verts = fp1 = MEM_mallocN(sizeof(float) * 3 * dl->parts, "filldisp1");
dlnew->nr = dl->parts;
dlnew->parts = 1;
dlnew->type = DL_POLY;
dlnew->col = dl->col;
dlnew->charidx = dl->charidx;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
fp = dl->verts + 3 * (dl->nr - 1);
dpoly = 3 * dl->nr;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
a = dl->parts;
while (a--) {
2011-11-06 15:17:43 +00:00
copy_v3_v3(fp1, fp);
2012-05-06 17:22:54 +00:00
fp1 += 3;
fp += dpoly;
2002-10-12 11:37:38 +00:00
}
}
}
}
2012-05-06 17:22:54 +00:00
dl = dl->next;
2002-10-12 11:37:38 +00:00
}
BKE_displist_fill(&front, dispbase, z_up, true);
BKE_displist_fill(&back, dispbase, z_up, false);
2012-05-07 06:58:03 +00:00
BKE_displist_free(&front);
BKE_displist_free(&back);
2002-10-12 11:37:38 +00:00
BKE_displist_fill(dispbase, dispbase, z_up, false);
2002-10-12 11:37:38 +00:00
}
static void curve_to_filledpoly(Curve *cu, ListBase *UNUSED(nurb), ListBase *dispbase)
2002-10-12 11:37:38 +00:00
{
if (!CU_DO_2DFILL(cu))
return;
2012-05-06 17:22:54 +00:00
if (dispbase->first && ((DispList *) dispbase->first)->type == DL_SURF) {
bevels_to_filledpoly(cu, dispbase);
}
2002-10-12 11:37:38 +00:00
else {
2014-08-07 09:14:54 +10:00
const float z_up[3] = {0.0f, 0.0f, 1.0f};
BKE_displist_fill(dispbase, dispbase, z_up, false);
2002-10-12 11:37:38 +00:00
}
}
/* taper rules:
2012-03-09 18:28:30 +00:00
* - only 1 curve
* - first point left, last point right
* - based on subdivided points in original curve, not on points in taper curve (still)
*/
static float displist_calc_taper(Scene *scene, Object *taperobj, float fac)
{
DispList *dl;
2012-05-07 06:58:03 +00:00
if (taperobj == NULL || taperobj->type != OB_CURVE)
return 1.0;
dl = taperobj->curve_cache ? taperobj->curve_cache->disp.first : NULL;
2012-05-06 17:22:54 +00:00
if (dl == NULL) {
2012-05-07 06:58:03 +00:00
BKE_displist_make_curveTypes(scene, taperobj, 0);
dl = taperobj->curve_cache->disp.first;
}
if (dl) {
float minx, dx, *fp;
int a;
2012-05-07 06:58:03 +00:00
/* horizontal size */
2012-05-06 17:22:54 +00:00
minx = dl->verts[0];
dx = dl->verts[3 * (dl->nr - 1)] - minx;
if (dx > 0.0f) {
2012-05-06 17:22:54 +00:00
fp = dl->verts;
for (a = 0; a < dl->nr; a++, fp += 3) {
2012-05-07 06:58:03 +00:00
if ((fp[0] - minx) / dx >= fac) {
/* interpolate with prev */
2012-05-06 17:22:54 +00:00
if (a > 0) {
float fac1 = (fp[-3] - minx) / dx;
float fac2 = (fp[0] - minx) / dx;
if (fac1 != fac2)
return fp[1] * (fac1 - fac) / (fac1 - fac2) + fp[-2] * (fac - fac2) / (fac1 - fac2);
}
return fp[1];
}
}
2012-05-06 17:22:54 +00:00
return fp[-2]; // last y coord
}
}
2012-05-07 06:58:03 +00:00
return 1.0;
}
float BKE_displist_calc_taper(Scene *scene, Object *taperobj, int cur, int tot)
{
float fac = ((float)cur) / (float)(tot - 1);
return displist_calc_taper(scene, taperobj, fac);
}
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
void BKE_displist_make_mball(EvaluationContext *eval_ctx, Scene *scene, Object *ob)
{
2012-05-07 06:58:03 +00:00
if (!ob || ob->type != OB_MBALL)
return;
if (ob == BKE_mball_basis_find(scene, ob)) {
if (ob->curve_cache) {
BKE_displist_free(&(ob->curve_cache->disp));
}
else {
ob->curve_cache = MEM_callocN(sizeof(CurveCache), "CurveCache for MBall");
}
BKE_mball_polygonize(eval_ctx, scene, ob, &ob->curve_cache->disp);
BKE_mball_texspace_calc(ob);
object_deform_mball(ob, &ob->curve_cache->disp);
/* NOP for MBALLs anyway... */
boundbox_displist_object(ob);
}
}
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
void BKE_displist_make_mball_forRender(EvaluationContext *eval_ctx, Scene *scene, Object *ob, ListBase *dispbase)
{
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
BKE_mball_polygonize(eval_ctx, scene, ob, dispbase);
BKE_mball_texspace_calc(ob);
2012-05-07 06:58:03 +00:00
object_deform_mball(ob, dispbase);
}
2014-03-20 22:56:28 +11:00
static ModifierData *curve_get_tessellate_point(Scene *scene, Object *ob,
const bool use_render_resolution, const bool editmode)
{
VirtualModifierData virtualModifierData;
ModifierData *md = modifiers_getVirtualModifierList(ob, &virtualModifierData);
ModifierData *pretessellatePoint;
int required_mode;
2014-03-20 22:56:28 +11:00
if (use_render_resolution)
2012-05-07 06:58:03 +00:00
required_mode = eModifierMode_Render;
else
required_mode = eModifierMode_Realtime;
2012-05-07 06:58:03 +00:00
if (editmode)
required_mode |= eModifierMode_Editmode;
pretessellatePoint = NULL;
2012-05-06 17:22:54 +00:00
for (; md; md = md->next) {
2015-03-30 21:17:07 +11:00
const ModifierTypeInfo *mti = modifierType_getInfo(md->type);
2012-05-07 06:58:03 +00:00
if (!modifier_isEnabled(scene, md, required_mode))
continue;
if (mti->type == eModifierTypeType_Constructive)
return pretessellatePoint;
if (ELEM(md->type, eModifierType_Hook, eModifierType_Softbody, eModifierType_MeshDeform)) {
pretessellatePoint = md;
/* this modifiers are moving point of tessellation automatically
2012-03-18 07:38:51 +00:00
* (some of them even can't be applied on tessellated curve), set flag
* for information button in modifier's header
2012-05-07 06:58:03 +00:00
*/
md->mode |= eModifierMode_ApplyOnSpline;
}
2012-05-06 17:22:54 +00:00
else if (md->mode & eModifierMode_ApplyOnSpline) {
pretessellatePoint = md;
}
}
return pretessellatePoint;
}
static void curve_calc_modifiers_pre(Scene *scene, Object *ob, ListBase *nurb,
2014-03-20 22:56:28 +11:00
const bool for_render, const bool use_render_resolution)
{
VirtualModifierData virtualModifierData;
ModifierData *md = modifiers_getVirtualModifierList(ob, &virtualModifierData);
ModifierData *pretessellatePoint;
2012-05-06 17:22:54 +00:00
Curve *cu = ob->data;
int numVerts = 0;
2014-03-20 22:56:28 +11:00
const bool editmode = (!for_render && (cu->editnurb || cu->editfont));
ModifierApplyFlag app_flag = 0;
float (*deformedVerts)[3] = NULL;
2012-05-06 17:22:54 +00:00
float *keyVerts = NULL;
int required_mode;
modifiers_clearErrors(ob);
if (editmode)
app_flag |= MOD_APPLY_USECACHE;
2014-03-20 22:56:28 +11:00
if (use_render_resolution) {
app_flag |= MOD_APPLY_RENDER;
2012-05-07 06:58:03 +00:00
required_mode = eModifierMode_Render;
}
2012-05-07 06:58:03 +00:00
else
required_mode = eModifierMode_Realtime;
2014-03-20 22:56:28 +11:00
pretessellatePoint = curve_get_tessellate_point(scene, ob, use_render_resolution, editmode);
2012-05-07 06:58:03 +00:00
if (editmode)
required_mode |= eModifierMode_Editmode;
Fix T50745: Shape key editing on bezier objects broken with Rendered Viewport Shading So... Curve+shapekey was even more broken than it looked, this report was actually a nice crasher (immediate crash in an ASAN build when trying to edit a curve shapekey with some viewport rendering enabled). There were actually two different issues here. I) The less critical: rB6f1493f68fe was not fully fixing issues from T50614. More specifically, if you updated obdata from editnurb *without* freeing editnurb afterwards, you had a 'restored' (to original curve) editnurb, without the edited shapekey modifications anymore. This was fixed by tweaking again `calc_shapeKeys()` behavior in `ED_curve_editnurb_load()`. II) The crasher: in `ED_curve_editnurb_make()`, the call to `init_editNurb_keyIndex()` was directly storing pointers of obdata nurbs. Since those get freed every time `ED_curve_editnurb_load()` is executed, it easily ended up being pointers to freed memory. This was fixed by copying those data, which implied more complex handling code for editnurbs->keyindex, and some reshuffling of a few functions to avoid duplicating things between editor's editcurve.c and BKE's curve.c Note that the separation of functions between editors and BKE area for curve could use a serious update, it's currently messy to say the least. Then again, that area is due to rework since a long time now... :/ Finally, aligned 'for_render' curve evaluation to mesh one - now editing a shapekey will show in rendered viewports, if it does have some weight (exactly as with shapekeys of meshes).
2017-02-22 21:20:50 +01:00
if (!editmode) {
keyVerts = BKE_key_evaluate_object(ob, &numVerts);
if (keyVerts) {
/* split coords from key data, the latter also includes
2012-03-09 18:28:30 +00:00
* tilts, which is passed through in the modifier stack.
* this is also the reason curves do not use a virtual
* shape key modifier yet. */
deformedVerts = BKE_curve_nurbs_keyVertexCos_get(nurb, keyVerts);
BLI_assert(BKE_nurbList_verts_count(nurb) == numVerts);
}
}
2012-05-07 06:58:03 +00:00
if (pretessellatePoint) {
2012-05-06 17:22:54 +00:00
for (; md; md = md->next) {
2015-03-30 21:17:07 +11:00
const ModifierTypeInfo *mti = modifierType_getInfo(md->type);
2012-05-06 17:22:54 +00:00
md->scene = scene;
2012-05-07 06:58:03 +00:00
if (!modifier_isEnabled(scene, md, required_mode))
2012-05-07 06:58:03 +00:00
continue;
if (mti->type != eModifierTypeType_OnlyDeform)
continue;
if (!deformedVerts) {
deformedVerts = BKE_curve_nurbs_vertexCos_get(nurb, &numVerts);
}
mti->deformVerts(md, ob, NULL, deformedVerts, numVerts, app_flag);
2012-05-06 17:22:54 +00:00
if (md == pretessellatePoint)
break;
}
}
if (deformedVerts) {
BK_curve_nurbs_vertexCos_apply(nurb, deformedVerts);
MEM_freeN(deformedVerts);
}
if (keyVerts) /* these are not passed through modifier stack */
BKE_curve_nurbs_keyVertexTilts_apply(nurb, keyVerts);
if (keyVerts)
MEM_freeN(keyVerts);
}
static float (*displist_get_allverts(ListBase *dispbase, int *totvert))[3]
{
DispList *dl;
float (*allverts)[3], *fp;
2012-05-06 17:22:54 +00:00
*totvert = 0;
2012-05-06 17:22:54 +00:00
for (dl = dispbase->first; dl; dl = dl->next)
*totvert += (dl->type == DL_INDEX3) ? dl->nr : dl->parts * dl->nr;
2012-05-06 17:22:54 +00:00
allverts = MEM_mallocN((*totvert) * sizeof(float) * 3, "displist_get_allverts allverts");
fp = (float *)allverts;
for (dl = dispbase->first; dl; dl = dl->next) {
int offs = 3 * ((dl->type == DL_INDEX3) ? dl->nr : dl->parts * dl->nr);
memcpy(fp, dl->verts, sizeof(float) * offs);
2012-05-06 17:22:54 +00:00
fp += offs;
}
return allverts;
}
static void displist_apply_allverts(ListBase *dispbase, float (*allverts)[3])
{
DispList *dl;
const float *fp;
2012-05-06 17:22:54 +00:00
fp = (float *)allverts;
for (dl = dispbase->first; dl; dl = dl->next) {
int offs = 3 * ((dl->type == DL_INDEX3) ? dl->nr : dl->parts * dl->nr);
memcpy(dl->verts, fp, sizeof(float) * offs);
2012-05-06 17:22:54 +00:00
fp += offs;
}
}
static void curve_calc_modifiers_post(Scene *scene, Object *ob, ListBase *nurb,
2014-03-20 22:56:28 +11:00
ListBase *dispbase, DerivedMesh **r_dm_final,
const bool for_render, const bool use_render_resolution)
{
VirtualModifierData virtualModifierData;
ModifierData *md = modifiers_getVirtualModifierList(ob, &virtualModifierData);
ModifierData *pretessellatePoint;
2012-05-06 17:22:54 +00:00
Curve *cu = ob->data;
int required_mode = 0, totvert = 0;
2014-03-20 22:56:28 +11:00
const bool editmode = (!for_render && (cu->editnurb || cu->editfont));
2012-05-06 17:22:54 +00:00
DerivedMesh *dm = NULL, *ndm;
float (*vertCos)[3] = NULL;
2014-03-20 22:56:28 +11:00
int useCache = !for_render;
ModifierApplyFlag app_flag = 0;
2014-03-20 22:56:28 +11:00
if (use_render_resolution) {
app_flag |= MOD_APPLY_RENDER;
2012-05-07 06:58:03 +00:00
required_mode = eModifierMode_Render;
}
2012-05-07 06:58:03 +00:00
else
required_mode = eModifierMode_Realtime;
2014-03-20 22:56:28 +11:00
pretessellatePoint = curve_get_tessellate_point(scene, ob, use_render_resolution, editmode);
2012-05-07 06:58:03 +00:00
if (editmode)
required_mode |= eModifierMode_Editmode;
if (pretessellatePoint) {
md = pretessellatePoint->next;
}
2014-03-20 22:56:28 +11:00
if (r_dm_final && *r_dm_final) {
(*r_dm_final)->release(*r_dm_final);
}
2012-05-06 17:22:54 +00:00
for (; md; md = md->next) {
2015-03-30 21:17:07 +11:00
const ModifierTypeInfo *mti = modifierType_getInfo(md->type);
ModifierApplyFlag appf = app_flag;
2012-05-06 17:22:54 +00:00
md->scene = scene;
if (!modifier_isEnabled(scene, md, required_mode))
2012-05-07 06:58:03 +00:00
continue;
if (mti->type == eModifierTypeType_OnlyDeform ||
(mti->type == eModifierTypeType_DeformOrConstruct && !dm))
{
if (editmode)
appf |= MOD_APPLY_USECACHE;
if (dm) {
if (!vertCos) {
totvert = dm->getNumVerts(dm);
vertCos = MEM_mallocN(sizeof(*vertCos) * totvert, "dfmv");
dm->getVertCos(dm, vertCos);
}
mti->deformVerts(md, ob, dm, vertCos, totvert, appf);
}
else {
if (!vertCos) {
2012-05-06 17:22:54 +00:00
vertCos = displist_get_allverts(dispbase, &totvert);
}
mti->deformVerts(md, ob, NULL, vertCos, totvert, appf);
}
}
else {
2014-03-20 22:56:28 +11:00
if (!r_dm_final) {
2012-05-07 06:58:03 +00:00
/* makeDisplistCurveTypes could be used for beveling, where derived mesh
* is totally unnecessary, so we could stop modifiers applying
* when we found constructive modifier but derived mesh is unwanted result
*/
break;
}
if (dm) {
if (vertCos) {
DerivedMesh *tdm = CDDM_copy(dm);
dm->release(dm);
dm = tdm;
CDDM_apply_vert_coords(dm, vertCos);
}
}
else {
if (vertCos) {
displist_apply_allverts(dispbase, vertCos);
}
if (ELEM(ob->type, OB_CURVE, OB_FONT) && (cu->flag & CU_DEFORM_FILL)) {
curve_to_filledpoly(cu, nurb, dispbase);
}
dm = CDDM_from_curve_displist(ob, dispbase);
}
if (vertCos) {
/* Vertex coordinates were applied to necessary data, could free it */
MEM_freeN(vertCos);
2012-05-06 17:22:54 +00:00
vertCos = NULL;
}
if (useCache)
appf |= MOD_APPLY_USECACHE;
ndm = modwrap_applyModifier(md, ob, dm, appf);
if (ndm) {
/* Modifier returned a new derived mesh */
if (dm && dm != ndm) /* Modifier */
2012-05-06 17:22:54 +00:00
dm->release(dm);
dm = ndm;
}
}
}
if (vertCos) {
if (dm) {
DerivedMesh *tdm = CDDM_copy(dm);
dm->release(dm);
dm = tdm;
CDDM_apply_vert_coords(dm, vertCos);
CDDM_calc_normals_mapping(dm);
MEM_freeN(vertCos);
}
else {
displist_apply_allverts(dispbase, vertCos);
MEM_freeN(vertCos);
2012-05-06 17:22:54 +00:00
vertCos = NULL;
}
}
2014-03-20 22:56:28 +11:00
if (r_dm_final) {
if (dm) {
/* see: mesh_calc_modifiers */
if (dm->getNumTessFaces(dm) == 0) {
dm->recalcTessellation(dm);
}
/* Even if tessellation is not needed, some modifiers might have modified CD layers
* (like mloopcol or mloopuv), hence we have to update those. */
else if (dm->dirty & DM_DIRTY_TESS_CDLAYERS) {
DM_update_tessface_data(dm);
}
if (dm->type == DM_TYPE_CDDM) {
CDDM_calc_normals_mapping_ex(dm, (dm->dirty & DM_DIRTY_NORMALS) ? false : true);
}
}
2014-03-20 22:56:28 +11:00
(*r_dm_final) = dm;
}
}
static void displist_surf_indices(DispList *dl)
{
int a, b, p1, p2, p3, p4;
int *index;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl->totindex = 0;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
index = dl->index = MEM_mallocN(4 * sizeof(int) * (dl->parts + 1) * (dl->nr + 1), "index array nurbs");
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
for (a = 0; a < dl->parts; a++) {
2012-05-07 06:58:03 +00:00
if (BKE_displist_surfindex_get(dl, a, &b, &p1, &p2, &p3, &p4) == 0)
break;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
for (; b < dl->nr; b++, index += 4) {
index[0] = p1;
index[1] = p2;
index[2] = p4;
index[3] = p3;
2012-05-07 06:58:03 +00:00
dl->totindex++;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
p2 = p1; p1++;
p4 = p3; p3++;
}
}
}
static DerivedMesh *create_orco_dm(Scene *scene, Object *ob)
{
DerivedMesh *dm;
2012-05-06 17:22:54 +00:00
ListBase disp = {NULL, NULL};
/* OrcoDM should be created from underformed disp lists */
2012-05-07 06:58:03 +00:00
BKE_displist_make_curveTypes_forOrco(scene, ob, &disp);
dm = CDDM_from_curve_displist(ob, &disp);
2012-05-07 06:58:03 +00:00
BKE_displist_free(&disp);
return dm;
}
static void add_orco_dm(Object *ob, DerivedMesh *dm, DerivedMesh *orcodm)
{
float (*orco)[3], (*layerorco)[3];
int totvert, a;
2012-05-06 17:22:54 +00:00
Curve *cu = ob->data;
2012-05-06 17:22:54 +00:00
totvert = dm->getNumVerts(dm);
orco = MEM_callocN(sizeof(float) * 3 * totvert, "dm orco");
if (orcodm->getNumVerts(orcodm) == totvert)
orcodm->getVertCos(orcodm, orco);
else
dm->getVertCos(dm, orco);
2012-05-06 17:22:54 +00:00
for (a = 0; a < totvert; a++) {
float *co = orco[a];
2012-05-06 17:22:54 +00:00
co[0] = (co[0] - cu->loc[0]) / cu->size[0];
co[1] = (co[1] - cu->loc[1]) / cu->size[1];
co[2] = (co[2] - cu->loc[2]) / cu->size[2];
}
if ((layerorco = DM_get_vert_data_layer(dm, CD_ORCO))) {
2012-05-06 17:22:54 +00:00
memcpy(layerorco, orco, sizeof(float) * totvert);
MEM_freeN(orco);
}
else
DM_add_vert_layer(dm, CD_ORCO, CD_ASSIGN, orco);
}
2014-03-20 22:56:28 +11:00
static void curve_calc_orcodm(Scene *scene, Object *ob, DerivedMesh *dm_final,
const bool for_render, const bool use_render_resolution)
{
2012-05-07 06:58:03 +00:00
/* this function represents logic of mesh's orcodm calculation
* for displist-based objects
*/
VirtualModifierData virtualModifierData;
ModifierData *md = modifiers_getVirtualModifierList(ob, &virtualModifierData);
ModifierData *pretessellatePoint;
2012-05-06 17:22:54 +00:00
Curve *cu = ob->data;
int required_mode;
2014-03-20 22:56:28 +11:00
const bool editmode = (!for_render && (cu->editnurb || cu->editfont));
2012-05-06 17:22:54 +00:00
DerivedMesh *ndm, *orcodm = NULL;
ModifierApplyFlag app_flag = MOD_APPLY_ORCO;
2014-03-20 22:56:28 +11:00
if (use_render_resolution) {
app_flag |= MOD_APPLY_RENDER;
2012-05-07 06:58:03 +00:00
required_mode = eModifierMode_Render;
}
2012-05-07 06:58:03 +00:00
else
required_mode = eModifierMode_Realtime;
2014-03-20 22:56:28 +11:00
pretessellatePoint = curve_get_tessellate_point(scene, ob, use_render_resolution, editmode);
2012-05-07 06:58:03 +00:00
if (editmode)
required_mode |= eModifierMode_Editmode;
if (pretessellatePoint) {
md = pretessellatePoint->next;
}
/* If modifiers are disabled, we wouldn't be here because
* this function is only called if there're enabled constructive
* modifiers applied on the curve.
*
* This means we can create ORCO DM in advance and assume it's
* never NULL.
*/
orcodm = create_orco_dm(scene, ob);
2012-05-06 17:22:54 +00:00
for (; md; md = md->next) {
2015-03-30 21:17:07 +11:00
const ModifierTypeInfo *mti = modifierType_getInfo(md->type);
2012-05-06 17:22:54 +00:00
md->scene = scene;
if (!modifier_isEnabled(scene, md, required_mode))
2012-05-07 06:58:03 +00:00
continue;
if (mti->type != eModifierTypeType_Constructive)
continue;
ndm = modwrap_applyModifier(md, ob, orcodm, app_flag);
if (ndm) {
/* if the modifier returned a new dm, release the old one */
if (orcodm && orcodm != ndm) {
orcodm->release(orcodm);
}
orcodm = ndm;
}
}
/* add an orco layer if needed */
2014-03-20 22:56:28 +11:00
add_orco_dm(ob, dm_final, orcodm);
orcodm->release(orcodm);
}
2012-05-07 06:58:03 +00:00
void BKE_displist_make_surf(Scene *scene, Object *ob, ListBase *dispbase,
2014-03-20 22:56:28 +11:00
DerivedMesh **r_dm_final,
const bool for_render, const bool for_orco, const bool use_render_resolution)
{
ListBase nubase = {NULL, NULL};
Nurb *nu;
Curve *cu = ob->data;
DispList *dl;
float *data;
int len;
2014-03-20 22:56:28 +11:00
if (!for_render && cu->editnurb) {
BKE_nurbList_duplicate(&nubase, BKE_curve_editNurbs_get(cu));
}
else {
BKE_nurbList_duplicate(&nubase, &cu->nurb);
}
2014-03-20 22:56:28 +11:00
if (!for_orco)
curve_calc_modifiers_pre(scene, ob, &nubase, for_render, use_render_resolution);
for (nu = nubase.first; nu; nu = nu->next) {
if ((for_render || nu->hide == 0) && BKE_nurb_check_valid_uv(nu)) {
2012-05-06 17:22:54 +00:00
int resolu = nu->resolu, resolv = nu->resolv;
2014-03-20 22:56:28 +11:00
if (use_render_resolution) {
2012-05-07 06:58:03 +00:00
if (cu->resolu_ren)
resolu = cu->resolu_ren;
if (cu->resolv_ren)
resolv = cu->resolv_ren;
}
2012-05-06 17:22:54 +00:00
if (nu->pntsv == 1) {
len = SEGMENTSU(nu) * resolu;
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListsurf");
dl->verts = MEM_mallocN(len * sizeof(float[3]), "dlverts");
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
dl->parts = 1;
dl->nr = len;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
/* dl->rt will be used as flag for render face and */
/* CU_2D conflicts with R_NOPUNOFLIP */
2012-05-06 17:22:54 +00:00
dl->rt = nu->flag & ~CU_2D;
2012-05-06 17:22:54 +00:00
data = dl->verts;
if (nu->flagu & CU_NURB_CYCLIC) dl->type = DL_POLY;
else dl->type = DL_SEGM;
2012-05-06 17:22:54 +00:00
BKE_nurb_makeCurve(nu, data, NULL, NULL, NULL, resolu, 3 * sizeof(float));
}
else {
2012-05-06 17:22:54 +00:00
len = (nu->pntsu * resolu) * (nu->pntsv * resolv);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListsurf");
dl->verts = MEM_mallocN(len * sizeof(float[3]), "dlverts");
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
/* dl->rt will be used as flag for render face and */
/* CU_2D conflicts with R_NOPUNOFLIP */
2012-05-06 17:22:54 +00:00
dl->rt = nu->flag & ~CU_2D;
2012-05-06 17:22:54 +00:00
data = dl->verts;
dl->type = DL_SURF;
2012-05-06 17:22:54 +00:00
dl->parts = (nu->pntsu * resolu); /* in reverse, because makeNurbfaces works that way */
dl->nr = (nu->pntsv * resolv);
if (nu->flagv & CU_NURB_CYCLIC) dl->flag |= DL_CYCL_U; /* reverse too! */
if (nu->flagu & CU_NURB_CYCLIC) dl->flag |= DL_CYCL_V;
BKE_nurb_makeFaces(nu, data, 0, resolu, resolv);
2012-05-07 06:58:03 +00:00
/* gl array drawing: using indices */
displist_surf_indices(dl);
2002-10-12 11:37:38 +00:00
}
}
}
2014-03-20 22:56:28 +11:00
if (!for_orco) {
BKE_nurbList_duplicate(&ob->curve_cache->deformed_nurbs, &nubase);
2014-03-20 22:56:28 +11:00
curve_calc_modifiers_post(scene, ob, &nubase, dispbase, r_dm_final,
for_render, use_render_resolution);
2012-05-07 06:58:03 +00:00
}
BKE_nurbList_free(&nubase);
}
static void rotateBevelPiece(Curve *cu, BevPoint *bevp, BevPoint *nbevp, DispList *dlb, float bev_blend, float widfac, float fac, float **r_data)
{
float *fp, *data = *r_data;
int b;
fp = dlb->verts;
2012-05-06 17:22:54 +00:00
for (b = 0; b < dlb->nr; b++, fp += 3, data += 3) {
if (cu->flag & CU_3D) {
float vec[3], quat[4];
2012-05-06 17:22:54 +00:00
vec[0] = fp[1] + widfac;
vec[1] = fp[2];
2012-05-06 17:22:54 +00:00
vec[2] = 0.0;
if (nbevp == NULL) {
copy_v3_v3(data, bevp->vec);
copy_qt_qt(quat, bevp->quat);
}
else {
interp_v3_v3v3(data, bevp->vec, nbevp->vec, bev_blend);
interp_qt_qtqt(quat, bevp->quat, nbevp->quat, bev_blend);
}
mul_qt_v3(quat, vec);
data[0] += fac * vec[0];
data[1] += fac * vec[1];
data[2] += fac * vec[2];
}
else {
float sina, cosa;
if (nbevp == NULL) {
copy_v3_v3(data, bevp->vec);
sina = bevp->sina;
cosa = bevp->cosa;
}
else {
interp_v3_v3v3(data, bevp->vec, nbevp->vec, bev_blend);
/* perhaps we need to interpolate angles instead. but the thing is
* cosa and sina are not actually sine and cosine
*/
sina = nbevp->sina * bev_blend + bevp->sina * (1.0f - bev_blend);
cosa = nbevp->cosa * bev_blend + bevp->cosa * (1.0f - bev_blend);
}
data[0] += fac * (widfac + fp[1]) * sina;
data[1] += fac * (widfac + fp[1]) * cosa;
data[2] += fac * fp[2];
}
}
*r_data = data;
}
static void fillBevelCap(Nurb *nu, DispList *dlb, float *prev_fp, ListBase *dispbase)
{
DispList *dl;
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListbev2");
dl->verts = MEM_mallocN(sizeof(float[3]) * dlb->nr, "dlverts");
memcpy(dl->verts, prev_fp, 3 * sizeof(float) * dlb->nr);
2012-05-06 17:22:54 +00:00
dl->type = DL_POLY;
2012-05-06 17:22:54 +00:00
dl->parts = 1;
dl->nr = dlb->nr;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
/* dl->rt will be used as flag for render face and */
/* CU_2D conflicts with R_NOPUNOFLIP */
2012-05-06 17:22:54 +00:00
dl->rt = nu->flag & ~CU_2D;
BLI_addtail(dispbase, dl);
}
static void calc_bevfac_segment_mapping(BevList *bl, float bevfac, float spline_length, int *r_bev, float *r_blend)
{
float normlen, normsum = 0.0f;
float *seglen = bl->seglen;
int *segbevcount = bl->segbevcount;
int bevcount = 0, nr = bl->nr;
float bev_fl = bevfac * (bl->nr - 1);
*r_bev = (int)bev_fl;
while (bevcount < nr - 1) {
normlen = *seglen / spline_length;
2014-08-20 08:47:32 +10:00
if (normsum + normlen > bevfac) {
bev_fl = bevcount + (bevfac - normsum) / normlen * *segbevcount;
*r_bev = (int) bev_fl;
*r_blend = bev_fl - *r_bev;
break;
}
normsum += normlen;
bevcount += *segbevcount;
segbevcount++;
seglen++;
}
}
static void calc_bevfac_spline_mapping(BevList *bl, float bevfac,
float spline_length,
int *r_bev, float *r_blend)
{
const float len_target = bevfac * spline_length;
BevPoint *bevp = bl->bevpoints;
float len_next = 0.0f, len = 0.0f;
int i = 0, nr = bl->nr;
while (nr--) {
bevp++;
len_next = len + bevp->offset;
if (len_next > len_target) {
break;
}
len = len_next;
i++;
}
*r_bev = i;
*r_blend = (len_target - len) / bevp->offset;
}
static void calc_bevfac_mapping_default(BevList *bl,
int *r_start, float *r_firstblend,
int *r_steps, float *r_lastblend)
{
*r_start = 0;
*r_steps = bl->nr;
*r_firstblend = 1.0f;
*r_lastblend = 1.0f;
}
static void calc_bevfac_mapping(Curve *cu, BevList *bl, Nurb *nu,
int *r_start, float *r_firstblend, int *r_steps, float *r_lastblend)
{
float tmpf, total_length = 0.0f;
int end = 0, i;
if ((BKE_nurb_check_valid_u(nu) == false) ||
/* not essential, but skips unnecessary calculation */
(min_ff(cu->bevfac1, cu->bevfac2) == 0.0f &&
max_ff(cu->bevfac1, cu->bevfac2) == 1.0f))
{
calc_bevfac_mapping_default(bl, r_start, r_firstblend, r_steps, r_lastblend);
return;
}
if (ELEM(cu->bevfac1_mapping, CU_BEVFAC_MAP_SEGMENT, CU_BEVFAC_MAP_SPLINE) ||
ELEM(cu->bevfac2_mapping, CU_BEVFAC_MAP_SEGMENT, CU_BEVFAC_MAP_SPLINE))
{
for (i = 0; i < SEGMENTSU(nu); i++) {
total_length += bl->seglen[i];
}
}
switch (cu->bevfac1_mapping) {
case CU_BEVFAC_MAP_RESOLU:
{
const float start_fl = cu->bevfac1 * (bl->nr - 1);
*r_start = (int)start_fl;
*r_firstblend = 1.0f - (start_fl - (*r_start));
break;
}
case CU_BEVFAC_MAP_SEGMENT:
{
calc_bevfac_segment_mapping(bl, cu->bevfac1, total_length, r_start, r_firstblend);
*r_firstblend = 1.0f - *r_firstblend;
break;
}
case CU_BEVFAC_MAP_SPLINE:
{
calc_bevfac_spline_mapping(bl, cu->bevfac1, total_length, r_start, r_firstblend);
*r_firstblend = 1.0f - *r_firstblend;
break;
}
}
switch (cu->bevfac2_mapping) {
case CU_BEVFAC_MAP_RESOLU:
{
const float end_fl = cu->bevfac2 * (bl->nr - 1);
end = (int)end_fl;
*r_steps = 2 + end - *r_start;
*r_lastblend = end_fl - end;
break;
}
case CU_BEVFAC_MAP_SEGMENT:
{
calc_bevfac_segment_mapping(bl, cu->bevfac2, total_length, &end, r_lastblend);
*r_steps = end - *r_start + 2;
break;
}
case CU_BEVFAC_MAP_SPLINE:
{
calc_bevfac_spline_mapping(bl, cu->bevfac2, total_length, &end, r_lastblend);
*r_steps = end - *r_start + 2;
break;
}
}
if (end < *r_start || (end == *r_start && *r_lastblend < 1.0f - *r_firstblend )) {
SWAP(int, *r_start, end);
tmpf = *r_lastblend;
*r_lastblend = 1.0f - *r_firstblend;
*r_firstblend = 1.0f - tmpf;
*r_steps = end - *r_start + 2;
}
if (*r_start + *r_steps > bl->nr) {
*r_steps = bl->nr - *r_start;
*r_lastblend = 1.0f;
}
}
static void do_makeDispListCurveTypes(Scene *scene, Object *ob, ListBase *dispbase,
2014-03-20 22:56:28 +11:00
DerivedMesh **r_dm_final,
const bool for_render, const bool for_orco, const bool use_render_resolution)
{
Curve *cu = ob->data;
/* we do allow duplis... this is only displist on curve level */
if (!ELEM(ob->type, OB_SURF, OB_CURVE, OB_FONT)) return;
2012-05-06 17:22:54 +00:00
if (ob->type == OB_SURF) {
2014-03-20 22:56:28 +11:00
BKE_displist_make_surf(scene, ob, dispbase, r_dm_final, for_render, for_orco, use_render_resolution);
2002-10-12 11:37:38 +00:00
}
else if (ELEM(ob->type, OB_CURVE, OB_FONT)) {
ListBase dlbev;
ListBase nubase = {NULL, NULL};
BKE_curve_bevelList_free(&ob->curve_cache->bev);
2015-10-07 15:02:06 +11:00
/* We only re-evaluate path if evaluation is not happening for orco.
* If the calculation happens for orco, we should never free data which
* was needed before and only not needed for orco calculation.
*/
2014-03-20 22:56:28 +11:00
if (!for_orco) {
if (ob->curve_cache->path) free_path(ob->curve_cache->path);
ob->curve_cache->path = NULL;
}
if (ob->type == OB_FONT) {
BKE_vfont_to_curve_nubase(G.main, ob, FO_EDIT, &nubase);
}
else {
BKE_nurbList_duplicate(&nubase, BKE_curve_nurbs_get(cu));
}
2014-03-20 22:56:28 +11:00
if (!for_orco)
curve_calc_modifiers_pre(scene, ob, &nubase, for_render, use_render_resolution);
BKE_curve_bevelList_make(ob, &nubase, for_render != false);
/* If curve has no bevel will return nothing */
2014-03-20 22:56:28 +11:00
BKE_curve_bevel_make(scene, ob, &dlbev, for_render, use_render_resolution);
2002-10-12 11:37:38 +00:00
/* no bevel or extrude, and no width correction? */
2012-05-06 17:22:54 +00:00
if (!dlbev.first && cu->width == 1.0f) {
2014-03-20 22:56:28 +11:00
curve_to_displist(cu, &nubase, dispbase, for_render, use_render_resolution);
}
else {
2012-05-06 17:22:54 +00:00
float widfac = cu->width - 1.0f;
BevList *bl = ob->curve_cache->bev.first;
Nurb *nu = nubase.first;
2012-05-06 17:22:54 +00:00
for (; bl && nu; bl = bl->next, nu = nu->next) {
DispList *dl;
float *data;
int a;
if (bl->nr) { /* blank bevel lists can happen */
/* exception handling; curve without bevel or extrude, with width correction */
if (BLI_listbase_is_empty(&dlbev)) {
BevPoint *bevp;
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListbev");
dl->verts = MEM_mallocN(sizeof(float[3]) * bl->nr, "dlverts");
BLI_addtail(dispbase, dl);
2012-05-06 17:22:54 +00:00
if (bl->poly != -1) dl->type = DL_POLY;
else dl->type = DL_SEGM;
2012-05-06 17:22:54 +00:00
if (dl->type == DL_SEGM) dl->flag = (DL_FRONT_CURVE | DL_BACK_CURVE);
2012-05-06 17:22:54 +00:00
dl->parts = 1;
dl->nr = bl->nr;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
/* dl->rt will be used as flag for render face and */
/* CU_2D conflicts with R_NOPUNOFLIP */
2012-05-06 17:22:54 +00:00
dl->rt = nu->flag & ~CU_2D;
2012-05-06 17:22:54 +00:00
a = dl->nr;
bevp = bl->bevpoints;
2012-05-06 17:22:54 +00:00
data = dl->verts;
while (a--) {
2012-05-06 17:22:54 +00:00
data[0] = bevp->vec[0] + widfac * bevp->sina;
data[1] = bevp->vec[1] + widfac * bevp->cosa;
data[2] = bevp->vec[2];
bevp++;
2012-05-06 17:22:54 +00:00
data += 3;
}
}
else {
DispList *dlb;
ListBase bottom_capbase = {NULL, NULL};
ListBase top_capbase = {NULL, NULL};
float bottom_no[3] = {0.0f};
float top_no[3] = {0.0f};
float firstblend = 0.0f, lastblend = 0.0f;
int i, start, steps = 0;
if (nu->flagu & CU_NURB_CYCLIC) {
calc_bevfac_mapping_default(bl,
&start, &firstblend, &steps, &lastblend);
}
else {
if (fabsf(cu->bevfac2 - cu->bevfac1) < FLT_EPSILON) {
continue;
}
calc_bevfac_mapping(cu, bl, nu, &start, &firstblend, &steps, &lastblend);
}
2012-05-06 17:22:54 +00:00
for (dlb = dlbev.first; dlb; dlb = dlb->next) {
BevPoint *bevp_first, *bevp_last;
BevPoint *bevp;
/* for each part of the bevel use a separate displblock */
2012-05-06 17:22:54 +00:00
dl = MEM_callocN(sizeof(DispList), "makeDispListbev1");
dl->verts = data = MEM_mallocN(sizeof(float[3]) * dlb->nr * steps, "dlverts");
BLI_addtail(dispbase, dl);
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl->type = DL_SURF;
2012-05-07 06:58:03 +00:00
2012-05-06 17:22:54 +00:00
dl->flag = dlb->flag & (DL_FRONT_CURVE | DL_BACK_CURVE);
if (dlb->type == DL_POLY) {
dl->flag |= DL_CYCL_U;
}
if ((bl->poly >= 0) && (steps > 2)) {
dl->flag |= DL_CYCL_V;
}
2012-05-06 17:22:54 +00:00
dl->parts = steps;
dl->nr = dlb->nr;
dl->col = nu->mat_nr;
dl->charidx = nu->charidx;
/* dl->rt will be used as flag for render face and */
/* CU_2D conflicts with R_NOPUNOFLIP */
2012-05-06 17:22:54 +00:00
dl->rt = nu->flag & ~CU_2D;
dl->bevel_split = BLI_BITMAP_NEW(steps, "bevel_split");
2012-05-07 06:58:03 +00:00
/* for each point of poly make a bevel piece */
bevp_first = bl->bevpoints;
bevp_last = &bl->bevpoints[bl->nr - 1];
bevp = &bl->bevpoints[start];
2012-05-06 17:22:54 +00:00
for (i = start, a = 0; a < steps; i++, bevp++, a++) {
float fac = 1.0;
float *cur_data = data;
2012-05-06 17:22:54 +00:00
if (cu->taperobj == NULL) {
fac = bevp->radius;
}
else {
float len, taper_fac;
if (cu->flag & CU_MAP_TAPER) {
len = (steps - 3) + firstblend + lastblend;
if (a == 0)
taper_fac = 0.0f;
else if (a == steps - 1)
taper_fac = 1.0f;
else
taper_fac = ((float) a - (1.0f - firstblend)) / len;
}
else {
len = bl->nr - 1;
taper_fac = (float) i / len;
if (a == 0)
taper_fac += (1.0f - firstblend) / len;
else if (a == steps - 1)
taper_fac -= (1.0f - lastblend) / len;
}
fac = displist_calc_taper(scene, cu->taperobj, taper_fac);
}
if (bevp->split_tag) {
BLI_BITMAP_ENABLE(dl->bevel_split, a);
}
2012-05-07 06:58:03 +00:00
/* rotate bevel piece and write in data */
if ((a == 0) && (bevp != bevp_last)) {
rotateBevelPiece(cu, bevp, bevp + 1, dlb, 1.0f - firstblend, widfac, fac, &data);
}
else if ((a == steps - 1) && (bevp != bevp_first) ) {
rotateBevelPiece(cu, bevp, bevp - 1, dlb, 1.0f - lastblend, widfac, fac, &data);
}
else {
rotateBevelPiece(cu, bevp, NULL, dlb, 0.0f, widfac, fac, &data);
}
if (cu->bevobj && (cu->flag & CU_FILL_CAPS) && !(nu->flagu & CU_NURB_CYCLIC)) {
if (a == 1) {
2012-05-06 17:22:54 +00:00
fillBevelCap(nu, dlb, cur_data - 3 * dlb->nr, &bottom_capbase);
negate_v3_v3(bottom_no, bevp->dir);
}
if (a == steps - 1) {
fillBevelCap(nu, dlb, cur_data, &top_capbase);
copy_v3_v3(top_no, bevp->dir);
}
}
2002-10-12 11:37:38 +00:00
}
/* gl array drawing: using indices */
displist_surf_indices(dl);
2002-10-12 11:37:38 +00:00
}
if (bottom_capbase.first) {
BKE_displist_fill(&bottom_capbase, dispbase, bottom_no, false);
BKE_displist_fill(&top_capbase, dispbase, top_no, false);
2012-05-07 06:58:03 +00:00
BKE_displist_free(&bottom_capbase);
BKE_displist_free(&top_capbase);
2002-10-12 11:37:38 +00:00
}
}
}
}
2012-05-07 06:58:03 +00:00
BKE_displist_free(&dlbev);
2002-10-12 11:37:38 +00:00
}
if (!(cu->flag & CU_DEFORM_FILL)) {
curve_to_filledpoly(cu, &nubase, dispbase);
}
2014-03-20 22:56:28 +11:00
if (!for_orco) {
if ((cu->flag & CU_PATH) ||
DAG_get_eval_flags_for_object(scene, ob) & DAG_EVAL_NEED_CURVE_PATH)
{
calc_curvepath(ob, &nubase);
}
}
2002-10-12 11:37:38 +00:00
if (!for_orco) {
BKE_nurbList_duplicate(&ob->curve_cache->deformed_nurbs, &nubase);
2014-03-20 22:56:28 +11:00
curve_calc_modifiers_post(scene, ob, &nubase, dispbase, r_dm_final, for_render, use_render_resolution);
}
if (cu->flag & CU_DEFORM_FILL && !ob->derivedFinal) {
curve_to_filledpoly(cu, &nubase, dispbase);
}
BKE_nurbList_free(&nubase);
2002-10-12 11:37:38 +00:00
}
}
void BKE_displist_make_curveTypes(Scene *scene, Object *ob, const bool for_orco)
{
ListBase *dispbase;
/* The same check for duplis as in do_makeDispListCurveTypes.
2012-03-09 18:28:30 +00:00
* Happens when curve used for constraint/bevel was converted to mesh.
* check there is still needed for render displist and orco displists. */
if (!ELEM(ob->type, OB_SURF, OB_CURVE, OB_FONT))
2012-05-07 06:58:03 +00:00
return;
BKE_object_free_derived_caches(ob);
if (!ob->curve_cache) {
ob->curve_cache = MEM_callocN(sizeof(CurveCache), "CurveCache for curve types");
}
dispbase = &(ob->curve_cache->disp);
2014-03-20 22:56:28 +11:00
do_makeDispListCurveTypes(scene, ob, dispbase, &ob->derivedFinal, 0, for_orco, 0);
boundbox_displist_object(ob);
}
2012-05-07 06:58:03 +00:00
void BKE_displist_make_curveTypes_forRender(Scene *scene, Object *ob, ListBase *dispbase,
2014-05-28 15:56:54 +06:00
DerivedMesh **r_dm_final, const bool for_orco,
const bool use_render_resolution)
{
if (ob->curve_cache == NULL) {
ob->curve_cache = MEM_callocN(sizeof(CurveCache), "CurveCache for Curve");
}
2014-03-20 22:56:28 +11:00
do_makeDispListCurveTypes(scene, ob, dispbase, r_dm_final, true, for_orco, use_render_resolution);
}
2012-05-07 06:58:03 +00:00
void BKE_displist_make_curveTypes_forOrco(struct Scene *scene, struct Object *ob, struct ListBase *dispbase)
{
if (ob->curve_cache == NULL) {
ob->curve_cache = MEM_callocN(sizeof(CurveCache), "CurveCache for Curve");
}
do_makeDispListCurveTypes(scene, ob, dispbase, NULL, 1, 1, 1);
}
/* add Orco layer to the displist object which has got derived mesh and return orco */
2014-03-20 22:56:28 +11:00
float *BKE_displist_make_orco(Scene *scene, Object *ob, DerivedMesh *dm_final,
const bool for_render,
const bool use_render_resolution)
{
float *orco;
2014-03-20 22:56:28 +11:00
if (dm_final == NULL)
dm_final = ob->derivedFinal;
2014-03-20 22:56:28 +11:00
if (!dm_final->getVertDataArray(dm_final, CD_ORCO)) {
curve_calc_orcodm(scene, ob, dm_final, for_render, use_render_resolution);
}
2014-03-20 22:56:28 +11:00
orco = dm_final->getVertDataArray(dm_final, CD_ORCO);
if (orco) {
2012-05-06 17:22:54 +00:00
orco = MEM_dupallocN(orco);
}
return orco;
2002-10-12 11:37:38 +00:00
}
void BKE_displist_minmax(ListBase *dispbase, float min[3], float max[3])
2002-10-12 11:37:38 +00:00
{
DispList *dl;
const float *vert;
2012-05-06 17:22:54 +00:00
int a, tot = 0;
int doit = 0;
2012-05-07 06:58:03 +00:00
for (dl = dispbase->first; dl; dl = dl->next) {
tot = (dl->type == DL_INDEX3) ? dl->nr : dl->nr * dl->parts;
vert = dl->verts;
for (a = 0; a < tot; a++, vert += 3) {
minmax_v3v3_v3(min, max, vert);
}
doit |= (tot != 0);
}
if (!doit) {
/* there's no geometry in displist, use zero-sized boundbox */
zero_v3(min);
zero_v3(max);
}
}
/* this is confusing, there's also min_max_object, appplying the obmat... */
static void boundbox_displist_object(Object *ob)
{
if (ELEM(ob->type, OB_CURVE, OB_SURF, OB_FONT)) {
/* Curve's BB is already calculated as a part of modifier stack,
* here we only calculate object BB based on final display list.
*/
2012-05-07 06:58:03 +00:00
/* object's BB is calculated from final displist */
if (ob->bb == NULL)
ob->bb = MEM_callocN(sizeof(BoundBox), "boundbox");
2012-05-07 06:58:03 +00:00
if (ob->derivedFinal) {
DM_set_object_boundbox(ob, ob->derivedFinal);
}
else {
float min[3], max[3];
INIT_MINMAX(min, max);
BKE_displist_minmax(&ob->curve_cache->disp, min, max);
BKE_boundbox_init_from_minmax(ob->bb, min, max);
ob->bb->flag &= ~BOUNDBOX_DIRTY;
}
2002-10-12 11:37:38 +00:00
}
}