archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Packages Activity
This repository has been archived on 2023-10-09.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
a59df21ad8282c20f33a216ba9e3ba739881577d
blender-archive/source/blender/blenkernel/intern/editderivedmesh.c

1537 lines
40 KiB
C
Raw Normal View History

remove $Id from headers
2011-10-24 23:32:24 +00:00
/*
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
* ***** BEGIN GPL LICENSE BLOCK *****
*
* 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.
*
* 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,
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
*
* The Original Code is Copyright (C) 2005 Blender Foundation.
* All rights reserved.
*
* The Original Code is: all of this file.
*
* Contributor(s): none yet.
*
* ***** END GPL LICENSE BLOCK *****
*/
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
/** \file blender/blenkernel/intern/editderivedmesh.c
* \ingroup bke
move editmesh functions out editderivedmesh.c into editmesh.c
2013-04-16 05:46:17 +00:00
*
* basic design:
*
* the bmesh derivedmesh exposes the mesh as triangles. it stores pointers
* to three loops per triangle. the derivedmesh stores a cache of tessellations
* for each face. this cache will smartly update as needed (though at first
* it'll simply be more brute force). keeping track of face/edge counts may
code cleanup: spelling
2013-10-31 14:10:01 +00:00
* be a small problem.
move editmesh functions out editderivedmesh.c into editmesh.c
2013-04-16 05:46:17 +00:00
*
* this won't be the most efficient thing, considering that internal edges and
* faces of tessellations are exposed. looking up an edge by index in particular
* is likely to be a little slow.
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
*/
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
Tweak to fix for thread concurency in looptri generation. Even if pointer assignment may be atomic, it does not prevent reordering and other nifty compiler tricks, we need a memory barrier to ensure not only that transferring pointer from wip array to final one is atomic, but also that all previous writing to memory are “flushed” to (visible by) all CPUs... Thanks @sergey for finding the potential (though quite unlikely) issue.
2017-09-25 09:56:02 +02:00
#include "atomic_ops.h"
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
#include "BLI_math.h"
Cleanup: add 2d suffix to BLI files Some of these API's can have 3D versions, explicitly name them 2D.
2018-02-18 21:27:33 +11:00
#include "BLI_jitter_2d.h"
fix for problem with too low epsilon when calculating thickness, also increase limits.
2013-04-18 17:23:02 +00:00
#include "BLI_bitmap.h"
Support multiple tangents for BI render & viewport Normal Map node support for GLSL mode and the internal render (multiple tangents support). The Normal Map node is a useful node which is present in the Cycles render. It makes it possible to use normal mapping without additional material node in a node tree. This patch implements Normal Map node for GLSL mode and the internal render. Previously only the active UV layer was used to calculate tangents.
2016-04-26 18:43:02 +10:00
#include "BLI_task.h"
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
#include "BKE_cdderivedmesh.h"
#include "BKE_mesh.h"
move editmesh functions out editderivedmesh.c into editmesh.c
2013-04-16 05:46:17 +00:00
#include "BKE_editmesh.h"
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
#include "BKE_editmesh_bvh.h"
DWM: Tangent support for editmesh
2017-05-09 08:50:45 +10:00
#include "BKE_editmesh_tangent.h"
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
#include "DNA_scene_types.h"
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
#include "DNA_object_types.h"
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
#include "DNA_mesh_types.h"
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
#include "MEM_guardedalloc.h"
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
typedef struct EditDerivedBMesh {
DerivedMesh dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMEditMesh *em;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
EditMeshData emd;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
} EditDerivedBMesh;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
/* -------------------------------------------------------------------- */
/* Lazy initialize datastructures */
static void emDM_ensurePolyNormals(EditDerivedBMesh *bmdm);
static void emDM_ensureVertNormals(EditDerivedBMesh *bmdm)
{
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos && (bmdm->emd.vertexNos == NULL)) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
BMesh *bm = bmdm->em->bm;
const float (*vertexCos)[3], (*polyNos)[3];
float (*vertexNos)[3];
/* calculate vertex normals from poly normals */
emDM_ensurePolyNormals(bmdm);
BM_mesh_elem_index_ensure(bm, BM_FACE);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
polyNos = bmdm->emd.polyNos;
vertexCos = bmdm->emd.vertexCos;
Fix Editderivedmeshes vertices normals computation. Those derived meshes (used in Edit mode) were using simple sum of neighbor poly normals to get vertex normals, while everywhere else in Blender we use weighted sum of such poly normals. Patch: D311 Reviewed and enhanced by Campbell, thanks!
2014-02-12 20:48:09 +01:00
vertexNos = MEM_callocN(sizeof(*vertexNos) * bm->totvert, __func__);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
Fix Editderivedmeshes vertices normals computation. Those derived meshes (used in Edit mode) were using simple sum of neighbor poly normals to get vertex normals, while everywhere else in Blender we use weighted sum of such poly normals. Patch: D311 Reviewed and enhanced by Campbell, thanks!
2014-02-12 20:48:09 +01:00
BM_verts_calc_normal_vcos(bm, polyNos, vertexCos, vertexNos);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
bmdm->emd.vertexNos = (const float (*)[3])vertexNos;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
}
}
static void emDM_ensurePolyNormals(EditDerivedBMesh *bmdm)
{
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos && (bmdm->emd.polyNos == NULL)) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
BMesh *bm = bmdm->em->bm;
const float (*vertexCos)[3];
float (*polyNos)[3];
BMFace *efa;
BMIter fiter;
int i;
BM_mesh_elem_index_ensure(bm, BM_VERT);
polyNos = MEM_mallocN(sizeof(*polyNos) * bm->totface, __func__);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
vertexCos = bmdm->emd.vertexCos;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
BM_ITER_MESH_INDEX (efa, &fiter, bm, BM_FACES_OF_MESH, i) {
BM_elem_index_set(efa, i); /* set_inline */
BM_face_calc_normal_vcos(bm, efa, polyNos[i], vertexCos);
}
bm->elem_index_dirty &= ~BM_FACE;
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
bmdm->emd.polyNos = (const float (*)[3])polyNos;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
}
}
static void emDM_ensurePolyCenters(EditDerivedBMesh *bmdm)
{
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.polyCos == NULL) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
BMesh *bm = bmdm->em->bm;
float (*polyCos)[3];
BMFace *efa;
BMIter fiter;
int i;
polyCos = MEM_mallocN(sizeof(*polyCos) * bm->totface, __func__);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
const float (*vertexCos)[3];
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
vertexCos = bmdm->emd.vertexCos;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
Fix bad index use drawing deformed face centers
2017-06-01 15:04:22 +10:00
BM_mesh_elem_index_ensure(bm, BM_VERT);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
BM_ITER_MESH_INDEX (efa, &fiter, bm, BM_FACES_OF_MESH, i) {
BM_face_calc_center_mean_vcos(bm, efa, polyCos[i], vertexCos);
}
}
else {
BM_ITER_MESH_INDEX (efa, &fiter, bm, BM_FACES_OF_MESH, i) {
BM_face_calc_center_mean(efa, polyCos[i]);
}
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
bmdm->emd.polyCos = (const float (*)[3])polyCos;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
}
}
modifier stack: lazy initialize normals many modifiers were calculating normals, when those normals were ignored by the next modifier. now flag normals as dirty and recalculate for modifiers that set use `dependsOnNormals()` callback. Quick test on mesh with 12 modifiers (mostly build type), calculated normals 6 times, now it only runs once - so this will give some speedup too.
2013-05-30 17:36:43 +00:00
static void emDM_calcNormals(DerivedMesh *dm)
Add CD_POLYINDEX layer to reduce need for retesselations
2011-11-13 15:13:59 +00:00
{
/* Nothing to do: normals are already calculated and stored on the
style cleanup / comment formatting for bli/bke/bmesh
2012-03-03 20:19:11 +00:00
* BMVerts and BMFaces */
modifier stack: lazy initialize normals many modifiers were calculating normals, when those normals were ignored by the next modifier. now flag normals as dirty and recalculate for modifiers that set use `dependsOnNormals()` callback. Quick test on mesh with 12 modifiers (mostly build type), calculated normals 6 times, now it only runs once - so this will give some speedup too.
2013-05-30 17:36:43 +00:00
dm->dirty &= ~DM_DIRTY_NORMALS;
Add CD_POLYINDEX layer to reduce need for retesselations
2011-11-13 15:13:59 +00:00
}
Add Custom Loop Normals. This is the core code for it, tools (datatransfer and modifier) will come in next commits). RNA api is already there, though. See the code for details, but basically, we define, for each 'smooth fan' (which is a set of adjacent loops around a same vertex that are smooth, i.e. have a single same normal), a 'loop normal space' (or lnor space), using auto-computed normal and relevant edges, and store custom normal as two angular factors inside that space. This allows to have custom normals 'following' deformations of the geometry, and to only save two shorts per loop in new clnor CDLayer. Normal manipulation (editing, mixing, interpolating, etc.) shall always happen with plain 3D vectors normals, and be converted back into storage format at the end. Clnor computation has also been threaded (at least for Mesh case, not for BMesh), since the process can be rather heavy with high poly meshes. Also, bumping subversion, and fix mess in 2.70 versioning code.
2015-02-05 14:24:48 +01:00
static void emDM_calcLoopNormalsSpaceArray(
DerivedMesh *dm, const bool use_split_normals, const float split_angle, MLoopNorSpaceArray *r_lnors_spacearr);
Make use/computation of lnors consistant. Issue was, when requesting (building) lnors for a mesh that has autosmooth disabled, one would expect to simply get vnors as lnors. Until now, it wasn't the case, which was bad e.g. for normal projections of loops in recent remap code (projecting along split loop normals when you would expect projection along vertex normals...). Also, removed the 'angle' parameter from RNA's `mesh.calc_normals_split`. This should *always* use mesh settings (both autosmooth and smoothresh), otherwise once again we'd get inconsistencies in some cases. Will update fbx and obj addons too.
2015-01-19 14:11:40 +01:00
static void emDM_calcLoopNormals(DerivedMesh *dm, const bool use_split_normals, const float split_angle)
Add Custom Loop Normals. This is the core code for it, tools (datatransfer and modifier) will come in next commits). RNA api is already there, though. See the code for details, but basically, we define, for each 'smooth fan' (which is a set of adjacent loops around a same vertex that are smooth, i.e. have a single same normal), a 'loop normal space' (or lnor space), using auto-computed normal and relevant edges, and store custom normal as two angular factors inside that space. This allows to have custom normals 'following' deformations of the geometry, and to only save two shorts per loop in new clnor CDLayer. Normal manipulation (editing, mixing, interpolating, etc.) shall always happen with plain 3D vectors normals, and be converted back into storage format at the end. Clnor computation has also been threaded (at least for Mesh case, not for BMesh), since the process can be rather heavy with high poly meshes. Also, bumping subversion, and fix mess in 2.70 versioning code.
2015-02-05 14:24:48 +01:00
{
emDM_calcLoopNormalsSpaceArray(dm, use_split_normals, split_angle, NULL);
}
/* #define DEBUG_CLNORS */
static void emDM_calcLoopNormalsSpaceArray(
DerivedMesh *dm, const bool use_split_normals, const float split_angle, MLoopNorSpaceArray *r_lnors_spacearr)
Split Normals I (1/5): basis for split normals (nearly nothing user-visible here): * Add a new calcLoopNormals function to DerivedMesh struct, and implement it for CDDM and CCGDM (subsurf). EditDerivedBMesh (edit mode DM) only gets a dummy one in this commit. * Add a tessellated version of CD_LOOPNORMAL layer (CD_TESSLOOPNORMAL), with relevant code to handle it (tessellation, rna access, etc.). * Change auto_smooth options of Mesh (angle now in radian internaly, and toggle is now used to enable/disable split normals in DM creation process). Note BI render code is not touched here, hence its behavior regarding this option is now incoherent, will be addressed in a separate commit. Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D365
2014-04-13 12:18:51 +02:00
{
Split Normals I (2/5): Add basic BMesh support of split normals. * Merely a re-implementation of core split algorithm for BMesh, taking advantage of topological data available. * This code needs valid loop indices, so added BM_LOOP support to BM_mesh_elem_index_ensure() & co. Reviewers: campbellbarton Reviewed By: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D366
2014-04-13 12:25:02 +02:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
const float (*vertexCos)[3], (*vertexNos)[3], (*polyNos)[3];
float (*loopNos)[3];
Add Custom Loop Normals. This is the core code for it, tools (datatransfer and modifier) will come in next commits). RNA api is already there, though. See the code for details, but basically, we define, for each 'smooth fan' (which is a set of adjacent loops around a same vertex that are smooth, i.e. have a single same normal), a 'loop normal space' (or lnor space), using auto-computed normal and relevant edges, and store custom normal as two angular factors inside that space. This allows to have custom normals 'following' deformations of the geometry, and to only save two shorts per loop in new clnor CDLayer. Normal manipulation (editing, mixing, interpolating, etc.) shall always happen with plain 3D vectors normals, and be converted back into storage format at the end. Clnor computation has also been threaded (at least for Mesh case, not for BMesh), since the process can be rather heavy with high poly meshes. Also, bumping subversion, and fix mess in 2.70 versioning code.
2015-02-05 14:24:48 +01:00
short (*clnors_data)[2];
int cd_loop_clnors_offset;
Split Normals I (2/5): Add basic BMesh support of split normals. * Merely a re-implementation of core split algorithm for BMesh, taking advantage of topological data available. * This code needs valid loop indices, so added BM_LOOP support to BM_mesh_elem_index_ensure() & co. Reviewers: campbellbarton Reviewed By: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D366
2014-04-13 12:25:02 +02:00
/* calculate loop normals from poly and vertex normals */
emDM_ensureVertNormals(bmdm);
Fix stupid typo that broke split normal shading in edit mode, when there was some deform modifiers.
2014-04-20 23:12:18 +02:00
emDM_ensurePolyNormals(bmdm);
Split Normals I (2/5): Add basic BMesh support of split normals. * Merely a re-implementation of core split algorithm for BMesh, taking advantage of topological data available. * This code needs valid loop indices, so added BM_LOOP support to BM_mesh_elem_index_ensure() & co. Reviewers: campbellbarton Reviewed By: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D366
2014-04-13 12:25:02 +02:00
dm->dirty &= ~DM_DIRTY_NORMALS;
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
vertexCos = bmdm->emd.vertexCos;
vertexNos = bmdm->emd.vertexNos;
polyNos = bmdm->emd.polyNos;
Split Normals I (2/5): Add basic BMesh support of split normals. * Merely a re-implementation of core split algorithm for BMesh, taking advantage of topological data available. * This code needs valid loop indices, so added BM_LOOP support to BM_mesh_elem_index_ensure() & co. Reviewers: campbellbarton Reviewed By: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D366
2014-04-13 12:25:02 +02:00
loopNos = dm->getLoopDataArray(dm, CD_NORMAL);
if (!loopNos) {
DM_add_loop_layer(dm, CD_NORMAL, CD_CALLOC, NULL);
loopNos = dm->getLoopDataArray(dm, CD_NORMAL);
}
Add Custom Loop Normals. This is the core code for it, tools (datatransfer and modifier) will come in next commits). RNA api is already there, though. See the code for details, but basically, we define, for each 'smooth fan' (which is a set of adjacent loops around a same vertex that are smooth, i.e. have a single same normal), a 'loop normal space' (or lnor space), using auto-computed normal and relevant edges, and store custom normal as two angular factors inside that space. This allows to have custom normals 'following' deformations of the geometry, and to only save two shorts per loop in new clnor CDLayer. Normal manipulation (editing, mixing, interpolating, etc.) shall always happen with plain 3D vectors normals, and be converted back into storage format at the end. Clnor computation has also been threaded (at least for Mesh case, not for BMesh), since the process can be rather heavy with high poly meshes. Also, bumping subversion, and fix mess in 2.70 versioning code.
2015-02-05 14:24:48 +01:00
/* We can have both, give priority to dm's data, and fallback to bm's ones. */
clnors_data = dm->getLoopDataArray(dm, CD_CUSTOMLOOPNORMAL);
cd_loop_clnors_offset = clnors_data ? -1 : CustomData_get_offset(&bm->ldata, CD_CUSTOMLOOPNORMAL);
BM_loops_calc_normal_vcos(bm, vertexCos, vertexNos, polyNos, use_split_normals, split_angle, loopNos,
r_lnors_spacearr, clnors_data, cd_loop_clnors_offset);
#ifdef DEBUG_CLNORS
if (r_lnors_spacearr) {
int i;
for (i = 0; i < numLoops; i++) {
if (r_lnors_spacearr->lspacearr[i]->ref_alpha != 0.0f) {
LinkNode *loops = r_lnors_spacearr->lspacearr[i]->loops;
printf("Loop %d uses lnor space %p:\n", i, r_lnors_spacearr->lspacearr[i]);
print_v3("\tfinal lnor:", loopNos[i]);
print_v3("\tauto lnor:", r_lnors_spacearr->lspacearr[i]->vec_lnor);
print_v3("\tref_vec:", r_lnors_spacearr->lspacearr[i]->vec_ref);
printf("\talpha: %f\n\tbeta: %f\n\tloops: %p\n", r_lnors_spacearr->lspacearr[i]->ref_alpha,
r_lnors_spacearr->lspacearr[i]->ref_beta, r_lnors_spacearr->lspacearr[i]->loops);
printf("\t\t(shared with loops");
while (loops) {
printf(" %d", GET_INT_FROM_POINTER(loops->link));
loops = loops->next;
}
printf(")\n");
}
else {
printf("Loop %d has no lnor space\n", i);
}
}
}
#endif
Split Normals I (1/5): basis for split normals (nearly nothing user-visible here): * Add a new calcLoopNormals function to DerivedMesh struct, and implement it for CDDM and CCGDM (subsurf). EditDerivedBMesh (edit mode DM) only gets a dummy one in this commit. * Add a tessellated version of CD_LOOPNORMAL layer (CD_TESSLOOPNORMAL), with relevant code to handle it (tessellation, rna access, etc.). * Change auto_smooth options of Mesh (angle now in radian internaly, and toggle is now used to enable/disable split normals in DM creation process). Note BI render code is not touched here, hence its behavior regarding this option is now incoherent, will be addressed in a separate commit. Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D365
2014-04-13 12:18:51 +02:00
}
Support multiple tangents for BI render & viewport Normal Map node support for GLSL mode and the internal render (multiple tangents support). The Normal Map node is a useful node which is present in the Cycles render. It makes it possible to use normal mapping without additional material node in a node tree. This patch implements Normal Map node for GLSL mode and the internal render. Previously only the active UV layer was used to calculate tangents.
2016-04-26 18:43:02 +10:00
static void emDM_calc_loop_tangents(
DerivedMesh *dm, bool calc_active_tangent,
DWM: Tangent support for editmesh
2017-05-09 08:50:45 +10:00
const char (*tangent_names)[MAX_NAME], int tangent_names_len)
Fix editmesh GLSL tangent drawing The generic tangent calculation relied on CDDM arrays which aren't available in edit-mode. Add a tangent calculation callback, which has its own implementation for editmesh data.
2015-08-24 15:14:34 +10:00
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMEditMesh *em = bmdm->em;
Fix T51561: Normal maps fail w/ quad + eevee Generalize derived-mesh tangent calculation so it can be used by Batch cache creation too.
2017-05-24 22:33:21 +10:00
if (CustomData_number_of_layers(&em->bm->ldata, CD_MLOOPUV) == 0) {
return;
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
const float (*poly_normals)[3] = bmdm->emd.polyNos;
DWM: Tangent support for editmesh
2017-05-09 08:50:45 +10:00
const float (*loop_normals)[3] = CustomData_get_layer(&dm->loopData, CD_NORMAL);
const float (*vert_orco)[3] = dm->getVertDataArray(dm, CD_ORCO); /* can be NULL */
BKE_editmesh_loop_tangent_calc(
em, calc_active_tangent,
tangent_names, tangent_names_len,
poly_normals, loop_normals,
vert_orco,
&dm->loopData, dm->numLoopData,
&dm->tangent_mask);
Fix editmesh GLSL tangent drawing The generic tangent calculation relied on CDDM arrays which aren't available in edit-mode. Add a tangent calculation callback, which has its own implementation for editmesh data.
2015-08-24 15:14:34 +10:00
}
Cleanup: style, use const
2016-01-20 15:05:01 +11:00
style cleanup - spelling - turns out we had tessellation spelt wrong all over. - use \directive for doxy (not @directive) - remove BLI_sparsemap.h - was from bmesh merge IIRC but entire file commented and not used.
2012-03-02 16:05:54 +00:00
static void emDM_recalcTessellation(DerivedMesh *UNUSED(dm))
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
{
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
/* do nothing */
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
Fix (unreported) looptri array not being recalculated in ccgDM and emDM In ccgDM and emDM, looptri array recalculation was being handled directly by `*DM_getLoopTriArray` (`getLoopTriArray` callback), while `*DM_recalcLoopTri` (`recalcLoopTri` callback) was doing nothing. This results in the array not being recalculated when other functions that depend on the array data called the recalc function. This moves all the recalculation code to `*DM_recalcLoopTri` and makes `*DM_getLoopTriArray` call that. This commit also makes a minor change to the `getNumLoopTri` function, so that it returns the correct number without having to recalculate the looptri array. Reviewed By: mont29 Differential Revision: https://developer.blender.org/D2375
2016-11-23 17:19:03 -02:00
static void emDM_recalcLoopTri(DerivedMesh *dm)
Add tessellation data to DerivedMesh (LoopTri) This stores loop indices into the loop array giving easier acess to data such as vertex-colors and UV's, removing the need to store an MFace duplicate of custom-data. This doesn't yet move all internal code from MFace to LoopTri just yet. Only applies to: - opengl drawing - sculpting (pbvh) - vertex/weight paint Thanks to @psy-fi for review, fixes and improvements to drawing!
2015-07-17 03:36:03 +10:00
{
Fix (unreported) looptri array not being recalculated in ccgDM and emDM In ccgDM and emDM, looptri array recalculation was being handled directly by `*DM_getLoopTriArray` (`getLoopTriArray` callback), while `*DM_recalcLoopTri` (`recalcLoopTri` callback) was doing nothing. This results in the array not being recalculated when other functions that depend on the array data called the recalc function. This moves all the recalculation code to `*DM_recalcLoopTri` and makes `*DM_getLoopTriArray` call that. This commit also makes a minor change to the `getNumLoopTri` function, so that it returns the correct number without having to recalculate the looptri array. Reviewed By: mont29 Differential Revision: https://developer.blender.org/D2375
2016-11-23 17:19:03 -02:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMLoop *(*looptris)[3] = bmdm->em->looptris;
MLoopTri *mlooptri;
const int tottri = bmdm->em->tottri;
int i;
DM_ensure_looptri_data(dm);
Fix T52816: regression can't open file in 2.79 (crash). Tentative fix, since I cannot reproduce thenissue for some reason here on linux. Core of the problem is pretty clear though, thanks to Germano Cavalcante (@mano-wii): another thread could try to use looptris data after worker one had allocated it, but before it had actually computed looptris. So now, we use a temp 'wip' pointer to store looptris being computed (since this is protected by a mutex, other threads will have to wait on it, no possibility for them to double-compute the looptris here). This should probably be backported to 2.79a if done.
2017-09-19 13:57:46 +02:00
mlooptri = dm->looptris.array_wip;
Fix (unreported) looptri array not being recalculated in ccgDM and emDM In ccgDM and emDM, looptri array recalculation was being handled directly by `*DM_getLoopTriArray` (`getLoopTriArray` callback), while `*DM_recalcLoopTri` (`recalcLoopTri` callback) was doing nothing. This results in the array not being recalculated when other functions that depend on the array data called the recalc function. This moves all the recalculation code to `*DM_recalcLoopTri` and makes `*DM_getLoopTriArray` call that. This commit also makes a minor change to the `getNumLoopTri` function, so that it returns the correct number without having to recalculate the looptri array. Reviewed By: mont29 Differential Revision: https://developer.blender.org/D2375
2016-11-23 17:19:03 -02:00
Fix T52852: Assert in looptri calculation after recent changes. Wrong condition in asserts...
2017-09-20 20:45:51 +02:00
BLI_assert(tottri == 0 || mlooptri != NULL);
Fix (unreported) looptri array not being recalculated in ccgDM and emDM In ccgDM and emDM, looptri array recalculation was being handled directly by `*DM_getLoopTriArray` (`getLoopTriArray` callback), while `*DM_recalcLoopTri` (`recalcLoopTri` callback) was doing nothing. This results in the array not being recalculated when other functions that depend on the array data called the recalc function. This moves all the recalculation code to `*DM_recalcLoopTri` and makes `*DM_getLoopTriArray` call that. This commit also makes a minor change to the `getNumLoopTri` function, so that it returns the correct number without having to recalculate the looptri array. Reviewed By: mont29 Differential Revision: https://developer.blender.org/D2375
2016-11-23 17:19:03 -02:00
BLI_assert(poly_to_tri_count(dm->numPolyData, dm->numLoopData) == dm->looptris.num);
BLI_assert(tottri == dm->looptris.num);
BM_mesh_elem_index_ensure(bmdm->em->bm, BM_FACE | BM_LOOP);
for (i = 0; i < tottri; i++) {
BMLoop **ltri = looptris[i];
MLoopTri *lt = &mlooptri[i];
ARRAY_SET_ITEMS(
Cleanup: code style & cmake
2017-03-12 02:40:04 +11:00
lt->tri,
BM_elem_index_get(ltri[0]),
BM_elem_index_get(ltri[1]),
BM_elem_index_get(ltri[2]));
Fix (unreported) looptri array not being recalculated in ccgDM and emDM In ccgDM and emDM, looptri array recalculation was being handled directly by `*DM_getLoopTriArray` (`getLoopTriArray` callback), while `*DM_recalcLoopTri` (`recalcLoopTri` callback) was doing nothing. This results in the array not being recalculated when other functions that depend on the array data called the recalc function. This moves all the recalculation code to `*DM_recalcLoopTri` and makes `*DM_getLoopTriArray` call that. This commit also makes a minor change to the `getNumLoopTri` function, so that it returns the correct number without having to recalculate the looptri array. Reviewed By: mont29 Differential Revision: https://developer.blender.org/D2375
2016-11-23 17:19:03 -02:00
lt->poly = BM_elem_index_get(ltri[0]->f);
}
Fix T52816: regression can't open file in 2.79 (crash). Tentative fix, since I cannot reproduce thenissue for some reason here on linux. Core of the problem is pretty clear though, thanks to Germano Cavalcante (@mano-wii): another thread could try to use looptris data after worker one had allocated it, but before it had actually computed looptris. So now, we use a temp 'wip' pointer to store looptris being computed (since this is protected by a mutex, other threads will have to wait on it, no possibility for them to double-compute the looptris here). This should probably be backported to 2.79a if done.
2017-09-19 13:57:46 +02:00
BLI_assert(dm->looptris.array == NULL);
Add an 'atomic cas' wrapper for pointers. Avoids having to repeat obfuscating castings everywhere...
2017-09-25 10:40:50 +02:00
atomic_cas_ptr((void **)&dm->looptris.array, dm->looptris.array, dm->looptris.array_wip);
Tweak to fix for thread concurency in looptri generation. Even if pointer assignment may be atomic, it does not prevent reordering and other nifty compiler tricks, we need a memory barrier to ensure not only that transferring pointer from wip array to final one is atomic, but also that all previous writing to memory are “flushed” to (visible by) all CPUs... Thanks @sergey for finding the potential (though quite unlikely) issue.
2017-09-25 09:56:02 +02:00
dm->looptris.array_wip = NULL;
Add tessellation data to DerivedMesh (LoopTri) This stores loop indices into the loop array giving easier acess to data such as vertex-colors and UV's, removing the need to store an MFace duplicate of custom-data. This doesn't yet move all internal code from MFace to LoopTri just yet. Only applies to: - opengl drawing - sculpting (pbvh) - vertex/weight paint Thanks to @psy-fi for review, fixes and improvements to drawing!
2015-07-17 03:36:03 +10:00
}
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
static void emDM_foreachMappedVert(
DerivedMesh *dm,
void (*func)(void *userData, int index, const float co[3], const float no_f[3], const short no_s[3]),
void *userData,
DMForeachFlag flag)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMVert *eve;
BMIter iter;
int i;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
const float (*vertexCos)[3] = bmdm->emd.vertexCos;
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
const float (*vertexNos)[3];
if (flag & DM_FOREACH_USE_NORMAL) {
emDM_ensureVertNormals(bmdm);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
vertexNos = bmdm->emd.vertexNos;
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
}
else {
vertexNos = NULL;
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
const float *no = (flag & DM_FOREACH_USE_NORMAL) ? vertexNos[i] : NULL;
func(userData, i, vertexCos[i], no, NULL);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
}
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
}
else {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
const float *no = (flag & DM_FOREACH_USE_NORMAL) ? eve->no : NULL;
func(userData, i, eve->co, no, NULL);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
}
Cleanup: ws
2015-07-18 05:54:31 +10:00
static void emDM_foreachMappedEdge(
DerivedMesh *dm,
void (*func)(void *userData, int index, const float v0co[3], const float v1co[3]),
void *userData)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMEdge *eed;
BMIter iter;
int i;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_mesh_elem_index_ensure(bm, BM_VERT);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eed, &iter, bm, BM_EDGES_OF_MESH, i) {
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
func(userData, i,
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
bmdm->emd.vertexCos[BM_elem_index_get(eed->v1)],
bmdm->emd.vertexCos[BM_elem_index_get(eed->v2)]);
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
}
else {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eed, &iter, bm, BM_EDGES_OF_MESH, i) {
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
func(userData, i, eed->v1->co, eed->v2->co);
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
Fix T39781: Autosmooth feature: erroneously showing hidden vertices' normals too when in "vertex per face" display So, turned out after all we need a foreachmapped helper for loops as well... :/ CDDM and EBDM were reasonably simple, but I fought hours with CCGDM (subsurf) to find only a bad working solution (see D478). So I fallback to a code similar to CDDM one. Probably not nice for performances, but loops and subsurf are not matching well... Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D478
2014-04-23 15:18:11 +02:00
static void emDM_foreachMappedLoop(
DerivedMesh *dm,
void (*func)(void *userData, int vertex_index, int face_index, const float co[3], const float no[3]),
void *userData,
DMForeachFlag flag)
{
/* We can't use dm->getLoopDataLayout(dm) here, we want to always access dm->loopData, EditDerivedBMesh would
* return loop data from bmesh itself. */
const float (*lnors)[3] = (flag & DM_FOREACH_USE_NORMAL) ? DM_get_loop_data_layer(dm, CD_NORMAL) : NULL;
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
BMFace *efa;
BMIter iter;
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
const float (*vertexCos)[3] = bmdm->emd.vertexCos;
Fix T39781: Autosmooth feature: erroneously showing hidden vertices' normals too when in "vertex per face" display So, turned out after all we need a foreachmapped helper for loops as well... :/ CDDM and EBDM were reasonably simple, but I fought hours with CCGDM (subsurf) to find only a bad working solution (see D478). So I fallback to a code similar to CDDM one. Probably not nice for performances, but loops and subsurf are not matching well... Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D478
2014-04-23 15:18:11 +02:00
int f_idx;
BM_mesh_elem_index_ensure(bm, BM_VERT);
BM_ITER_MESH_INDEX (efa, &iter, bm, BM_FACES_OF_MESH, f_idx) {
BMLoop *l_iter, *l_first;
l_iter = l_first = BM_FACE_FIRST_LOOP(efa);
do {
const BMVert *eve = l_iter->v;
const int v_idx = BM_elem_index_get(eve);
const float *no = lnors ? *lnors++ : NULL;
func(userData, v_idx, f_idx, vertexCos ? vertexCos[v_idx] : eve->co, no);
} while ((l_iter = l_iter->next) != l_first);
}
}
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
static void emDM_foreachMappedFaceCenter(
DerivedMesh *dm,
void (*func)(void *userData, int index, const float co[3], const float no[3]),
void *userData,
DMForeachFlag flag)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BMesh *bm = bmdm->em->bm;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
const float (*polyNos)[3];
const float (*polyCos)[3];
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMFace *efa;
BMIter iter;
int i;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
emDM_ensurePolyCenters(bmdm);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
polyCos = bmdm->emd.polyCos; /* always set */
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
if (flag & DM_FOREACH_USE_NORMAL) {
emDM_ensurePolyNormals(bmdm);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
polyNos = bmdm->emd.polyNos; /* maybe NULL */
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
}
else {
polyNos = NULL;
}
if (polyNos) {
BM_ITER_MESH_INDEX (efa, &iter, bm, BM_FACES_OF_MESH, i) {
const float *no = polyNos[i];
func(userData, i, polyCos[i], no);
}
}
else {
BM_ITER_MESH_INDEX (efa, &iter, bm, BM_FACES_OF_MESH, i) {
const float *no = (flag & DM_FOREACH_USE_NORMAL) ? efa->no : NULL;
func(userData, i, polyCos[i], no);
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_getMinMax(DerivedMesh *dm, float r_min[3], float r_max[3])
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMVert *eve;
BMIter iter;
int i;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
if (bm->totvert) {
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
minmax_v3v3_v3(r_min, r_max, bmdm->emd.vertexCos[i]);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
}
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
}
else {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH (eve, &iter, bm, BM_VERTS_OF_MESH) {
minmax_v3v3_v3(r_min, r_max, eve->co);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
}
else {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
zero_v3(r_min);
zero_v3(r_max);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static int emDM_getNumVerts(DerivedMesh *dm)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return bmdm->em->bm->totvert;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static int emDM_getNumEdges(DerivedMesh *dm)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return bmdm->em->bm->totedge;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static int emDM_getNumTessFaces(DerivedMesh *dm)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return bmdm->em->tottri;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static int emDM_getNumLoops(DerivedMesh *dm)
More DM func renames, fixing some build breaks, renaming more stuff, also seems like it might be fixing the recent CDDM_copy corruption/leak bug
2011-11-30 18:03:56 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return bmdm->em->bm->totloop;
More DM func renames, fixing some build breaks, renaming more stuff, also seems like it might be fixing the recent CDDM_copy corruption/leak bug
2011-11-30 18:03:56 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static int emDM_getNumPolys(DerivedMesh *dm)
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return bmdm->em->bm->totface;
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
}
Code cleanup: use bool for static methods
2014-02-05 22:36:15 +11:00
static void bmvert_to_mvert(BMesh *bm, BMVert *ev, MVert *r_vert)
(NOTE: DO NOT TEST) Start of planned DerivedMesh refactoring. The mface interfaces in DerivedMesh have been renamed to reflect their new status as tesselated face interfaces (rather then the primary ones, which are now stored in mpolys). short review: mpolys store "primary" face data, while mfaces store the tesselated form of the mesh (generally as triangles). mpolys are defined by mloops, and each mpoly defines a range of loops it "owns" in the main mloop array. I've also added basic read-only face iterators, which are implemented for CDDM, ccgsubsurf, and the bmeditmesh derivedmesh. Since faces are now variable-length things, trying to implement the same interface as mfaces would not have worked well (especially since faces are stored as an mpoly + a range of mloops). I figure first we can evaluate these simple read-only face iterators, then decide if a) we like using iterators in DerivedMesh, b) how much of it should use them, and c) if we want write-capable iterators. I plan to write official docs on this design after I get it more stable; I'm committing now because there's a rather lot of changes, and I might do a merge soon.
2009-06-10 10:06:25 +00:00
{
Code cleanup: use 'const' for arrays (blenkernel)
2014-04-27 00:20:13 +10:00
const float *f;
speedup for editmesh getVert and getEdge, were doing 2x customdata layer lookups for crease and bevel weight, also make use of the existing meshdata array for lookups.
2013-01-09 18:20:11 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
copy_v3_v3(r_vert->co, ev->co);
(NOTE: DO NOT TEST) Start of planned DerivedMesh refactoring. The mface interfaces in DerivedMesh have been renamed to reflect their new status as tesselated face interfaces (rather then the primary ones, which are now stored in mpolys). short review: mpolys store "primary" face data, while mfaces store the tesselated form of the mesh (generally as triangles). mpolys are defined by mloops, and each mpoly defines a range of loops it "owns" in the main mloop array. I've also added basic read-only face iterators, which are implemented for CDDM, ccgsubsurf, and the bmeditmesh derivedmesh. Since faces are now variable-length things, trying to implement the same interface as mfaces would not have worked well (especially since faces are stored as an mpoly + a range of mloops). I figure first we can evaluate these simple read-only face iterators, then decide if a) we like using iterators in DerivedMesh, b) how much of it should use them, and c) if we want write-capable iterators. I plan to write official docs on this design after I get it more stable; I'm committing now because there's a rather lot of changes, and I might do a merge soon.
2009-06-10 10:06:25 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
normal_float_to_short_v3(r_vert->no, ev->no);
(NOTE: DO NOT TEST) Start of planned DerivedMesh refactoring. The mface interfaces in DerivedMesh have been renamed to reflect their new status as tesselated face interfaces (rather then the primary ones, which are now stored in mpolys). short review: mpolys store "primary" face data, while mfaces store the tesselated form of the mesh (generally as triangles). mpolys are defined by mloops, and each mpoly defines a range of loops it "owns" in the main mloop array. I've also added basic read-only face iterators, which are implemented for CDDM, ccgsubsurf, and the bmeditmesh derivedmesh. Since faces are now variable-length things, trying to implement the same interface as mfaces would not have worked well (especially since faces are stored as an mpoly + a range of mloops). I figure first we can evaluate these simple read-only face iterators, then decide if a) we like using iterators in DerivedMesh, b) how much of it should use them, and c) if we want write-capable iterators. I plan to write official docs on this design after I get it more stable; I'm committing now because there's a rather lot of changes, and I might do a merge soon.
2009-06-10 10:06:25 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert->flag = BM_vert_flag_to_mflag(ev);
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
speedup for editmesh getVert and getEdge, were doing 2x customdata layer lookups for crease and bevel weight, also make use of the existing meshdata array for lookups.
2013-01-09 18:20:11 +00:00
if ((f = CustomData_bmesh_get(&bm->vdata, ev->head.data, CD_BWEIGHT))) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert->bweight = (unsigned char)((*f) * 255.0f);
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
}
(NOTE: DO NOT TEST) Start of planned DerivedMesh refactoring. The mface interfaces in DerivedMesh have been renamed to reflect their new status as tesselated face interfaces (rather then the primary ones, which are now stored in mpolys). short review: mpolys store "primary" face data, while mfaces store the tesselated form of the mesh (generally as triangles). mpolys are defined by mloops, and each mpoly defines a range of loops it "owns" in the main mloop array. I've also added basic read-only face iterators, which are implemented for CDDM, ccgsubsurf, and the bmeditmesh derivedmesh. Since faces are now variable-length things, trying to implement the same interface as mfaces would not have worked well (especially since faces are stored as an mpoly + a range of mloops). I figure first we can evaluate these simple read-only face iterators, then decide if a) we like using iterators in DerivedMesh, b) how much of it should use them, and c) if we want write-capable iterators. I plan to write official docs on this design after I get it more stable; I'm committing now because there's a rather lot of changes, and I might do a merge soon.
2009-06-10 10:06:25 +00:00
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_getVert(DerivedMesh *dm, int index, MVert *r_vert)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMVert *ev;
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
if (UNLIKELY(index < 0 || index >= bm->totvert)) {
fix for error in editmesh derived callbacks, emDM_getVert(), emDM_getEdge(), emDM_getTessFace(). Currently the functions aren't used since they would fail have failed immediately.
2013-04-13 20:40:34 +00:00
BLI_assert(!"error in emDM_getVert");
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
return;
}
move bmesh array lookup data and utility functions from editmesh into bmesh, since enough bmesh operations can also take advantage of direct index lookups on verts/edges/faces. developers note: - EDBM_index_arrays_init/ensure/free -> BM_mesh_elem_table_ensure/init/free - EDBM_vert/edge/face_at_index -> BM_vert/edge/face_at_index - EDBM_uv_element_map_create/free -> BM_uv_element_map_create/free - ED_uv_element_get -> BM_uv_element_get
2013-10-28 02:05:33 +00:00
BLI_assert((bm->elem_table_dirty & BM_VERT) == 0);
ev = bm->vtable[index]; /* should be BM_vert_at_index() */
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
// ev = BM_vert_at_index(bm, index); /* warning, does list loop, _not_ ideal */
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
bmvert_to_mvert(bm, ev, r_vert);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos)
copy_v3_v3(r_vert->co, bmdm->emd.vertexCos[index]);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
static void emDM_getVertCo(DerivedMesh *dm, int index, float r_co[3])
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
if (UNLIKELY(index < 0 || index >= bm->totvert)) {
BLI_assert(!"error in emDM_getVertCo");
return;
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
copy_v3_v3(r_co, bmdm->emd.vertexCos[index]);
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
}
else {
move bmesh array lookup data and utility functions from editmesh into bmesh, since enough bmesh operations can also take advantage of direct index lookups on verts/edges/faces. developers note: - EDBM_index_arrays_init/ensure/free -> BM_mesh_elem_table_ensure/init/free - EDBM_vert/edge/face_at_index -> BM_vert/edge/face_at_index - EDBM_uv_element_map_create/free -> BM_uv_element_map_create/free - ED_uv_element_get -> BM_uv_element_get
2013-10-28 02:05:33 +00:00
BMVert *ev;
BLI_assert((bm->elem_table_dirty & BM_VERT) == 0);
ev = bm->vtable[index]; /* should be BM_vert_at_index() */
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
// ev = BM_vert_at_index(bm, index); /* warning, does list loop, _not_ ideal */
copy_v3_v3(r_co, ev->co);
}
}
static void emDM_getVertNo(DerivedMesh *dm, int index, float r_no[3])
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
if (UNLIKELY(index < 0 || index >= bm->totvert)) {
BLI_assert(!"error in emDM_getVertNo");
return;
}
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
emDM_ensureVertNormals(bmdm);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
copy_v3_v3(r_no, bmdm->emd.vertexNos[index]);
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
}
else {
move bmesh array lookup data and utility functions from editmesh into bmesh, since enough bmesh operations can also take advantage of direct index lookups on verts/edges/faces. developers note: - EDBM_index_arrays_init/ensure/free -> BM_mesh_elem_table_ensure/init/free - EDBM_vert/edge/face_at_index -> BM_vert/edge/face_at_index - EDBM_uv_element_map_create/free -> BM_uv_element_map_create/free - ED_uv_element_get -> BM_uv_element_get
2013-10-28 02:05:33 +00:00
BMVert *ev;
BLI_assert((bm->elem_table_dirty & BM_VERT) == 0);
ev = bm->vtable[index]; /* should be BM_vert_at_index() */
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
// ev = BM_vert_at_index(bm, index); /* warning, does list loop, _not_ ideal */
copy_v3_v3(r_no, ev->no);
}
}
static void emDM_getPolyNo(DerivedMesh *dm, int index, float r_no[3])
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
if (UNLIKELY(index < 0 || index >= bm->totface)) {
BLI_assert(!"error in emDM_getPolyNo");
return;
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
emDM_ensurePolyNormals(bmdm);
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
copy_v3_v3(r_no, bmdm->emd.polyNos[index]);
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
}
else {
move bmesh array lookup data and utility functions from editmesh into bmesh, since enough bmesh operations can also take advantage of direct index lookups on verts/edges/faces. developers note: - EDBM_index_arrays_init/ensure/free -> BM_mesh_elem_table_ensure/init/free - EDBM_vert/edge/face_at_index -> BM_vert/edge/face_at_index - EDBM_uv_element_map_create/free -> BM_uv_element_map_create/free - ED_uv_element_get -> BM_uv_element_get
2013-10-28 02:05:33 +00:00
BMFace *efa;
BLI_assert((bm->elem_table_dirty & BM_FACE) == 0);
efa = bm->ftable[index]; /* should be BM_vert_at_index() */
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
// efa = BM_face_at_index(bm, index); /* warning, does list loop, _not_ ideal */
copy_v3_v3(r_no, efa->no);
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_getEdge(DerivedMesh *dm, int index, MEdge *r_edge)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMEdge *e;
Code cleanup: use 'const' for arrays (blenkernel)
2014-04-27 00:20:13 +10:00
const float *f;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
if (UNLIKELY(index < 0 || index >= bm->totedge)) {
fix for error in editmesh derived callbacks, emDM_getVert(), emDM_getEdge(), emDM_getTessFace(). Currently the functions aren't used since they would fail have failed immediately.
2013-04-13 20:40:34 +00:00
BLI_assert(!"error in emDM_getEdge");
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
return;
}
move bmesh array lookup data and utility functions from editmesh into bmesh, since enough bmesh operations can also take advantage of direct index lookups on verts/edges/faces. developers note: - EDBM_index_arrays_init/ensure/free -> BM_mesh_elem_table_ensure/init/free - EDBM_vert/edge/face_at_index -> BM_vert/edge/face_at_index - EDBM_uv_element_map_create/free -> BM_uv_element_map_create/free - ED_uv_element_get -> BM_uv_element_get
2013-10-28 02:05:33 +00:00
BLI_assert((bm->elem_table_dirty & BM_EDGE) == 0);
e = bm->etable[index]; /* should be BM_edge_at_index() */
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
// e = BM_edge_at_index(bm, index); /* warning, does list loop, _not_ ideal */
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->flag = BM_edge_flag_to_mflag(e);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->v1 = BM_elem_index_get(e->v1);
r_edge->v2 = BM_elem_index_get(e->v2);
speedup for editmesh getVert and getEdge, were doing 2x customdata layer lookups for crease and bevel weight, also make use of the existing meshdata array for lookups.
2013-01-09 18:20:11 +00:00
if ((f = CustomData_bmesh_get(&bm->edata, e->head.data, CD_BWEIGHT))) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->bweight = (unsigned char)((*f) * 255.0f);
speedup for editmesh getVert and getEdge, were doing 2x customdata layer lookups for crease and bevel weight, also make use of the existing meshdata array for lookups.
2013-01-09 18:20:11 +00:00
}
if ((f = CustomData_bmesh_get(&bm->edata, e->head.data, CD_CREASE))) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->crease = (unsigned char)((*f) * 255.0f);
speedup for editmesh getVert and getEdge, were doing 2x customdata layer lookups for crease and bevel weight, also make use of the existing meshdata array for lookups.
2013-01-09 18:20:11 +00:00
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_getTessFace(DerivedMesh *dm, int index, MFace *r_face)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMFace *ef;
code cleanup: editmesh draw functions, make face drawing more consistent.
2013-04-13 17:57:11 +00:00
BMLoop **ltri;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
fix for error in editmesh derived callbacks, emDM_getVert(), emDM_getEdge(), emDM_getTessFace(). Currently the functions aren't used since they would fail have failed immediately.
2013-04-13 20:40:34 +00:00
if (UNLIKELY(index < 0 || index >= bmdm->em->tottri)) {
BLI_assert(!"error in emDM_getTessFace");
Array modifier is now implemented using bmesh (though it's not completely feature-complete yet). I ported over the remove doubles code from the old bmesh branch for this, and split it into two bmops, "Weld Verts" and "Remove Doubles". Weld verts welds specific verts together, while remove doubles finds doubles and welds them. I also reverted the hotkey change I made earlier.
2009-06-23 05:35:49 +00:00
return;
}
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
ltri = bmdm->em->looptris[index];
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: editmesh draw functions, make face drawing more consistent.
2013-04-13 17:57:11 +00:00
ef = ltri[0]->f;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_face->mat_nr = (unsigned char) ef->mat_nr;
r_face->flag = BM_face_flag_to_mflag(ef);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_face->v1 = BM_elem_index_get(ltri[0]->v);
r_face->v2 = BM_elem_index_get(ltri[1]->v);
r_face->v3 = BM_elem_index_get(ltri[2]->v);
r_face->v4 = 0;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
test_index_face(r_face, NULL, 0, 3);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_copyVertArray(DerivedMesh *dm, MVert *r_vert)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
Fix part of #30982: dupliface object with modifiers did not show duplis correct while in edit mode.
2012-04-17 13:07:13 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BMVert *eve;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMIter iter;
don't store bevel weights or edge crease customdata layers in editmode unless they are needed. configurable in 'Geometry Data' panel, will be added when running crease edges transform for example.
2013-01-10 04:43:31 +00:00
const int cd_vert_bweight_offset = CustomData_get_offset(&bm->vdata, CD_BWEIGHT);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
int i;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
copy_v3_v3(r_vert->co, bmdm->emd.vertexCos[i]);
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
normal_float_to_short_v3(r_vert->no, eve->no);
r_vert->flag = BM_vert_flag_to_mflag(eve);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert->bweight = (cd_vert_bweight_offset != -1) ? BM_ELEM_CD_GET_FLOAT_AS_UCHAR(eve, cd_vert_bweight_offset) : 0;
don't store bevel weights or edge crease customdata layers in editmode unless they are needed. configurable in 'Geometry Data' panel, will be added when running crease edges transform for example.
2013-01-10 04:43:31 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert++;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
}
}
else {
BM_ITER_MESH (eve, &iter, bm, BM_VERTS_OF_MESH) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
copy_v3_v3(r_vert->co, eve->co);
normal_float_to_short_v3(r_vert->no, eve->no);
r_vert->flag = BM_vert_flag_to_mflag(eve);
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert->bweight = (cd_vert_bweight_offset != -1) ? BM_ELEM_CD_GET_FLOAT_AS_UCHAR(eve, cd_vert_bweight_offset) : 0;
don't store bevel weights or edge crease customdata layers in editmode unless they are needed. configurable in 'Geometry Data' panel, will be added when running crease edges transform for example.
2013-01-10 04:43:31 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_vert++;
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_copyEdgeArray(DerivedMesh *dm, MEdge *r_edge)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = ((EditDerivedBMesh *)dm)->em->bm;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BMEdge *eed;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMIter iter;
don't store bevel weights or edge crease customdata layers in editmode unless they are needed. configurable in 'Geometry Data' panel, will be added when running crease edges transform for example.
2013-01-10 04:43:31 +00:00
const int cd_edge_bweight_offset = CustomData_get_offset(&bm->edata, CD_BWEIGHT);
const int cd_edge_crease_offset = CustomData_get_offset(&bm->edata, CD_CREASE);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code refactor, function renaming for bmesh. These changes are to make the bmesh api more consistent and easier to learn, grouping similar functions which is convenient for autocomplete. This uses similar convention to RNA. * use face/loop/edge/vert as a prefix for functions. * use 'elem' as a prefix too for functions that can take any type with a BMHeader. * changed from camel case to underscore separated (like RNA).
2012-02-12 10:51:45 +00:00
BM_mesh_elem_index_ensure(bm, BM_VERT);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BM_ITER_MESH (eed, &iter, bm, BM_EDGES_OF_MESH) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->v1 = BM_elem_index_get(eed->v1);
r_edge->v2 = BM_elem_index_get(eed->v2);
wip commit; DO NOT USE. almost done with phase 1 of this restructuring, basically just some things that needed to be done before trunk (and some things that needed to be started, but can be finished much later).
2010-07-14 22:06:10 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->flag = BM_edge_flag_to_mflag(eed);
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge->crease = (cd_edge_crease_offset != -1) ? BM_ELEM_CD_GET_FLOAT_AS_UCHAR(eed, cd_edge_crease_offset) : 0;
r_edge->bweight = (cd_edge_bweight_offset != -1) ? BM_ELEM_CD_GET_FLOAT_AS_UCHAR(eed, cd_edge_bweight_offset) : 0;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_edge++;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_copyTessFaceArray(DerivedMesh *dm, MFace *r_face)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
struct BMLoop *(*looptris)[3] = bmdm->em->looptris;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMFace *ef;
int i;
code refactor, function renaming for bmesh. These changes are to make the bmesh api more consistent and easier to learn, grouping similar functions which is convenient for autocomplete. This uses similar convention to RNA. * use face/loop/edge/vert as a prefix for functions. * use 'elem' as a prefix too for functions that can take any type with a BMHeader. * changed from camel case to underscore separated (like RNA).
2012-02-12 10:51:45 +00:00
BM_mesh_elem_index_ensure(bm, BM_VERT);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
for (i = 0; i < bmdm->em->tottri; i++, r_face++) {
code cleanup: editmesh draw functions, make face drawing more consistent.
2013-04-13 17:57:11 +00:00
BMLoop **ltri = looptris[i];
ef = ltri[0]->f;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_face->mat_nr = (unsigned char) ef->mat_nr;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_face->flag = BM_face_flag_to_mflag(ef);
r_face->edcode = 0;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_face->v1 = BM_elem_index_get(ltri[0]->v);
r_face->v2 = BM_elem_index_get(ltri[1]->v);
r_face->v3 = BM_elem_index_get(ltri[2]->v);
r_face->v4 = 0;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
test_index_face(r_face, NULL, 0, 3);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_copyLoopArray(DerivedMesh *dm, MLoop *r_loop)
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
{
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
optimization: - halve the number of allocs in layerInterp_mdeformvert list creation. - use direct loop access in emDM_copyLoopArray
2013-07-22 18:01:27 +00:00
BMIter iter;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BMFace *efa;
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
code refactor, function renaming for bmesh. These changes are to make the bmesh api more consistent and easier to learn, grouping similar functions which is convenient for autocomplete. This uses similar convention to RNA. * use face/loop/edge/vert as a prefix for functions. * use 'elem' as a prefix too for functions that can take any type with a BMHeader. * changed from camel case to underscore separated (like RNA).
2012-02-12 10:51:45 +00:00
BM_mesh_elem_index_ensure(bm, BM_VERT | BM_EDGE);
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BM_ITER_MESH (efa, &iter, bm, BM_FACES_OF_MESH) {
optimization: - halve the number of allocs in layerInterp_mdeformvert list creation. - use direct loop access in emDM_copyLoopArray
2013-07-22 18:01:27 +00:00
BMLoop *l_iter, *l_first;
l_iter = l_first = BM_FACE_FIRST_LOOP(efa);
do {
correct own error in recent edits to editderivedmesh
2013-07-23 15:05:33 +00:00
r_loop->v = BM_elem_index_get(l_iter->v);
r_loop->e = BM_elem_index_get(l_iter->e);
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_loop++;
optimization: - halve the number of allocs in layerInterp_mdeformvert list creation. - use direct loop access in emDM_copyLoopArray
2013-07-22 18:01:27 +00:00
} while ((l_iter = l_iter->next) != l_first);
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
}
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_copyPolyArray(DerivedMesh *dm, MPoly *r_poly)
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
{
edit-derived-mesh had a table for vert/edge/face lookups - that were never used. removed these and replaced them with calls to BM_vert_at_index, BM_edge_at_index - which are not ideal but same as trunk does.
2012-02-19 01:51:36 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
fixing compile errors with VisualStudio 2008. * macro ABS has no effect with uintptr_t anyway and was throwing warning (promoted to error) -> commented out and marked with TODO * removed two unused variables trhowing warning also promoted to error when compiling.
2011-07-03 17:07:07 +00:00
BMIter iter;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BMFace *efa;
Fix some typos & bugs from some of my previous patches and commits
2011-09-26 05:10:37 +00:00
int i;
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
Fix some typos & bugs from some of my previous patches and commits
2011-09-26 05:10:37 +00:00
i = 0;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
BM_ITER_MESH (efa, &iter, bm, BM_FACES_OF_MESH) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_poly->flag = BM_face_flag_to_mflag(efa);
r_poly->loopstart = i;
r_poly->totloop = efa->len;
r_poly->mat_nr = efa->mat_nr;
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
r_poly++;
code cleanup: - use bmesh iterator macros in more places - rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
i += efa->len;
=bmesh= Removed the DerivedMesh face iterators (they sucked). This should make subsurf faster. Also sped up multires a bit (not sure if it's strictly correct, need to look at it later).
2011-06-14 03:16:08 +00:00
}
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static void *emDM_getTessFaceDataArray(DerivedMesh *dm, int type)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
BMesh *bm = bmdm->em->bm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
void *datalayer;
made the crash handler disabled by a command line flag, insted of being #ifdef'd out on release builds. also did little bit of work on subsurf. and face select can now deselect again, and fixed some other bugs with it (and subsurf).
2009-08-15 17:31:28 +00:00
datalayer = DM_get_tessface_data_layer(dm, type);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
if (datalayer)
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
return datalayer;
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
/* layers are store per face for editmesh, we convert to a temporary
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
* data layer array in the derivedmesh when these are requested */
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
if (type == CD_MTFACE || type == CD_MCOL) {
cleanup: redundant casts & const cast correctness
2015-01-01 23:26:03 +11:00
const char *bmdata;
char *data;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
TexFace removal part 3 - MTexPoly structure & layer type. - The 'Mesh.uv_textures' layers. - DerivedMesh TexFace drawing. - Scripts & UI.
2017-05-25 15:11:00 +10:00
bool has_type_source = CustomData_has_layer(&bm->ldata, (type == CD_MTFACE) ? CD_MLOOPUV : CD_MLOOPCOL);
Fix error accessing tessface color in edit-mode Was checking for loop-color in poly-layer.
2017-05-25 02:44:58 +10:00
if (has_type_source) {
code cleanup - commant/remove unused vars and reduce scope.
2011-09-07 06:49:20 +00:00
/* offset = bm->pdata.layers[index].offset; */ /* UNUSED */
avoid customdata lookups for selection test/enable disable. also add uvedit_face_select_set, uvedit_edge_select_set, uvedit_uv_select_set - since quite a few areas where setting based on a boolean.
2013-05-10 06:46:32 +00:00
BMLoop *(*looptris)[3] = bmdm->em->looptris;
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
const int size = CustomData_sizeof(type);
int i, j;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
made the crash handler disabled by a command line flag, insted of being #ifdef'd out on release builds. also did little bit of work on subsurf. and face select can now deselect again, and fixed some other bugs with it (and subsurf).
2009-08-15 17:31:28 +00:00
DM_add_tessface_layer(dm, type, CD_CALLOC, NULL);
Fix error accessing tessface color in edit-mode Was checking for loop-color in poly-layer.
2017-05-25 02:44:58 +10:00
const int index = CustomData_get_layer_index(&dm->faceData, type);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
dm->faceData.layers[index].flag |= CD_FLAG_TEMPORARY;
made the crash handler disabled by a command line flag, insted of being #ifdef'd out on release builds. also did little bit of work on subsurf. and face select can now deselect again, and fixed some other bugs with it (and subsurf).
2009-08-15 17:31:28 +00:00
data = datalayer = DM_get_tessface_data_layer(dm, type);
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
if (type == CD_MTFACE) {
avoid customdata lookups for selection test/enable disable. also add uvedit_face_select_set, uvedit_edge_select_set, uvedit_uv_select_set - since quite a few areas where setting based on a boolean.
2013-05-10 06:46:32 +00:00
const int cd_loop_uv_offset = CustomData_get_offset(&bm->ldata, CD_MLOOPUV);
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
for (i = 0; i < bmdm->em->tottri; i++, data += size) {
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
for (j = 0; j < 3; j++) {
avoid customdata lookups for selection test/enable disable. also add uvedit_face_select_set, uvedit_edge_select_set, uvedit_uv_select_set - since quite a few areas where setting based on a boolean.
2013-05-10 06:46:32 +00:00
// bmdata = CustomData_bmesh_get(&bm->ldata, looptris[i][j]->head.data, CD_MLOOPUV);
bmdata = BM_ELEM_CD_GET_VOID_P(looptris[i][j], cd_loop_uv_offset);
cleanup: redundant casts & const cast correctness
2015-01-01 23:26:03 +11:00
copy_v2_v2(((MTFace *)data)->uv[j], ((const MLoopUV *)bmdata)->uv);
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
}
}
}
else {
avoid customdata lookups for selection test/enable disable. also add uvedit_face_select_set, uvedit_edge_select_set, uvedit_uv_select_set - since quite a few areas where setting based on a boolean.
2013-05-10 06:46:32 +00:00
const int cd_loop_color_offset = CustomData_get_offset(&bm->ldata, CD_MLOOPCOL);
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
for (i = 0; i < bmdm->em->tottri; i++, data += size) {
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
for (j = 0; j < 3; j++) {
avoid customdata lookups for selection test/enable disable. also add uvedit_face_select_set, uvedit_edge_select_set, uvedit_uv_select_set - since quite a few areas where setting based on a boolean.
2013-05-10 06:46:32 +00:00
// bmdata = CustomData_bmesh_get(&bm->ldata, looptris[i][j]->head.data, CD_MLOOPCOL);
bmdata = BM_ELEM_CD_GET_VOID_P(looptris[i][j], cd_loop_color_offset);
cleanup: redundant casts & const cast correctness
2015-01-01 23:26:03 +11:00
MESH_MLOOPCOL_TO_MCOL(((const MLoopCol *)bmdata), (((MCol *)data) + j));
bmesh api todo: emDM_getTessFaceDataArray now returns UV's and Colors from editmode loops.
2012-04-16 04:17:33 +00:00
}
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
}
}
Split Normals I (2/5): Add basic BMesh support of split normals. * Merely a re-implementation of core split algorithm for BMesh, taking advantage of topological data available. * This code needs valid loop indices, so added BM_LOOP support to BM_mesh_elem_index_ensure() & co. Reviewers: campbellbarton Reviewed By: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D366
2014-04-13 12:25:02 +02:00
/* Special handling for CD_TESSLOOPNORMAL, we generate it on demand as well. */
if (type == CD_TESSLOOPNORMAL) {
const float (*lnors)[3] = dm->getLoopDataArray(dm, CD_NORMAL);
if (lnors) {
BMLoop *(*looptris)[3] = bmdm->em->looptris;
short (*tlnors)[4][3], (*tlnor)[4][3];
int index, i, j;
DM_add_tessface_layer(dm, type, CD_CALLOC, NULL);
index = CustomData_get_layer_index(&dm->faceData, type);
dm->faceData.layers[index].flag |= CD_FLAG_TEMPORARY;
tlnor = tlnors = DM_get_tessface_data_layer(dm, type);
BM_mesh_elem_index_ensure(bm, BM_LOOP);
for (i = 0; i < bmdm->em->tottri; i++, tlnor++, looptris++) {
for (j = 0; j < 3; j++) {
normal_float_to_short_v3((*tlnor)[j], lnors[BM_elem_index_get((*looptris)[j])]);
}
}
}
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
return datalayer;
}
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
static void emDM_getVertCos(DerivedMesh *dm, float (*r_cos)[3])
=bmesh= fixed mesh deform crasher
2011-05-09 22:16:31 +00:00
{
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BMesh *bm = bmdm->em->bm;
=bmesh= fixed mesh deform crasher
2011-05-09 22:16:31 +00:00
BMVert *eve;
BMIter iter;
int i;
fixed 4 uses of un-inirialized vars + some compiler warnings.
2011-05-10 23:48:09 +00:00
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
copy_v3_v3(r_cos[i], bmdm->emd.vertexCos[i]);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
}
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
}
else {
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
copy_v3_v3(r_cos[i], eve->co);
=bmesh= fixed mesh deform crasher
2011-05-09 22:16:31 +00:00
}
}
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
static void emDM_release(DerivedMesh *dm)
(NOTE: DO NOT TEST) Start of planned DerivedMesh refactoring. The mface interfaces in DerivedMesh have been renamed to reflect their new status as tesselated face interfaces (rather then the primary ones, which are now stored in mpolys). short review: mpolys store "primary" face data, while mfaces store the tesselated form of the mesh (generally as triangles). mpolys are defined by mloops, and each mpoly defines a range of loops it "owns" in the main mloop array. I've also added basic read-only face iterators, which are implemented for CDDM, ccgsubsurf, and the bmeditmesh derivedmesh. Since faces are now variable-length things, trying to implement the same interface as mfaces would not have worked well (especially since faces are stored as an mpoly + a range of mloops). I figure first we can evaluate these simple read-only face iterators, then decide if a) we like using iterators in DerivedMesh, b) how much of it should use them, and c) if we want write-capable iterators. I plan to write official docs on this design after I get it more stable; I'm committing now because there's a rather lot of changes, and I might do a merge soon.
2009-06-10 10:06:25 +00:00
{
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
if (DM_release(dm)) {
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.vertexCos) {
MEM_freeN((void *)bmdm->emd.vertexCos);
if (bmdm->emd.vertexNos) {
MEM_freeN((void *)bmdm->emd.vertexNos);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.polyNos) {
MEM_freeN((void *)bmdm->emd.polyNos);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
}
}
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
if (bmdm->emd.polyCos) {
MEM_freeN((void *)bmdm->emd.polyCos);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
MEM_freeN(bmdm);
}
}
set many vars & functions as static, also commented invalid pin theme alpha assignment so bmesh branch builds with -Werror
2011-05-11 02:14:43 +00:00
static CustomData *bmDm_getVertDataLayout(DerivedMesh *dm)
rearranged some code a bit
2009-08-18 20:05:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return &bmdm->em->bm->vdata;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
}
set many vars & functions as static, also commented invalid pin theme alpha assignment so bmesh branch builds with -Werror
2011-05-11 02:14:43 +00:00
static CustomData *bmDm_getEdgeDataLayout(DerivedMesh *dm)
rearranged some code a bit
2009-08-18 20:05:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return &bmdm->em->bm->edata;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
}
set many vars & functions as static, also commented invalid pin theme alpha assignment so bmesh branch builds with -Werror
2011-05-11 02:14:43 +00:00
static CustomData *bmDm_getTessFaceDataLayout(DerivedMesh *dm)
rearranged some code a bit
2009-08-18 20:05:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
=bmesh= glsl drawing in editmode works again, also fixed some minor customdata bugs
2011-05-10 17:01:26 +00:00
return &bmdm->dm.faceData;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
}
set many vars & functions as static, also commented invalid pin theme alpha assignment so bmesh branch builds with -Werror
2011-05-11 02:14:43 +00:00
static CustomData *bmDm_getLoopDataLayout(DerivedMesh *dm)
rearranged some code a bit
2009-08-18 20:05:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return &bmdm->em->bm->ldata;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
}
More DM func renames, this one includes renames of the function pointer members of the DerivedMesh struct
2011-11-29 13:01:51 +00:00
static CustomData *bmDm_getPolyDataLayout(DerivedMesh *dm)
rearranged some code a bit
2009-08-18 20:05:08 +00:00
{
style cleanup: blenkernel
2012-05-12 19:18:02 +00:00
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
return &bmdm->em->bm->pdata;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
}
EditMesh: Avoid creating deform-vert layer every redraw Getting a new edit-derived-bmesh was always creating a deform-vert array, even when it wasn't needed. Since this was called on redraw, in many cases it was doing it unnecessarily. Now pass in a custom-data mask and only fill in deform-verts when needed. Gives noticeable drawing speedup (~10-30% here).
2016-06-22 21:20:09 +10:00
/**
* \note This may be called per-draw,
* avoid allocating large arrays where possible and keep this a thin wrapper for #BMesh.
*/
Cleanup: ws
2015-07-18 05:54:31 +10:00
DerivedMesh *getEditDerivedBMesh(
EditMesh: Avoid creating deform-vert layer every redraw Getting a new edit-derived-bmesh was always creating a deform-vert array, even when it wasn't needed. Since this was called on redraw, in many cases it was doing it unnecessarily. Now pass in a custom-data mask and only fill in deform-verts when needed. Gives noticeable drawing speedup (~10-30% here).
2016-06-22 21:20:09 +10:00
BMEditMesh *em, struct Object *UNUSED(ob),
CustomDataMask data_mask,
Cleanup: ws
2015-07-18 05:54:31 +10:00
float (*vertexCos)[3])
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
{
fix memory leak in array modifier not freeing editmesh, also use better names for allocs.
2011-09-12 02:44:17 +00:00
EditDerivedBMesh *bmdm = MEM_callocN(sizeof(*bmdm), __func__);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMesh *bm = em->bm;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: rename BKE_tessmesh -> BKE_editmesh, rename EditDerivedBMesh.tc -> em. ('tc' is odd name which isn't used elsewhere).
2013-04-13 20:31:52 +00:00
bmdm->em = em;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
DM_init((DerivedMesh *)bmdm, DM_TYPE_EDITBMESH, bm->totvert,
bm->totedge, em->tottri, bm->totloop, bm->totface);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
don't store bevel weights or edge crease customdata layers in editmode unless they are needed. configurable in 'Geometry Data' panel, will be added when running crease edges transform for example.
2013-01-10 04:43:31 +00:00
/* could also get from the objects mesh directly */
bmdm->dm.cd_flag = BM_mesh_cd_flag_from_bmesh(bm);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
bmdm->dm.getVertCos = emDM_getVertCos;
bmdm->dm.getMinMax = emDM_getMinMax;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
rearranged some code a bit
2009-08-18 20:05:08 +00:00
bmdm->dm.getVertDataLayout = bmDm_getVertDataLayout;
bmdm->dm.getEdgeDataLayout = bmDm_getEdgeDataLayout;
bmdm->dm.getTessFaceDataLayout = bmDm_getTessFaceDataLayout;
bmdm->dm.getLoopDataLayout = bmDm_getLoopDataLayout;
More DM func renames, this one includes renames of the function pointer members of the DerivedMesh struct
2011-11-29 13:01:51 +00:00
bmdm->dm.getPolyDataLayout = bmDm_getPolyDataLayout;
rearranged some code a bit
2009-08-18 20:05:08 +00:00
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
bmdm->dm.getNumVerts = emDM_getNumVerts;
bmdm->dm.getNumEdges = emDM_getNumEdges;
bmdm->dm.getNumTessFaces = emDM_getNumTessFaces;
bmdm->dm.getNumLoops = emDM_getNumLoops;
bmdm->dm.getNumPolys = emDM_getNumPolys;
bmdm->dm.getVert = emDM_getVert;
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
bmdm->dm.getVertCo = emDM_getVertCo;
bmdm->dm.getVertNo = emDM_getVertNo;
bmdm->dm.getPolyNo = emDM_getPolyNo;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
bmdm->dm.getEdge = emDM_getEdge;
bmdm->dm.getTessFace = emDM_getTessFace;
bmdm->dm.copyVertArray = emDM_copyVertArray;
bmdm->dm.copyEdgeArray = emDM_copyEdgeArray;
bmdm->dm.copyTessFaceArray = emDM_copyTessFaceArray;
bmdm->dm.copyLoopArray = emDM_copyLoopArray;
bmdm->dm.copyPolyArray = emDM_copyPolyArray;
bmdm->dm.getTessFaceDataArray = emDM_getTessFaceDataArray;
bmdm->dm.calcNormals = emDM_calcNormals;
Split Normals I (1/5): basis for split normals (nearly nothing user-visible here): * Add a new calcLoopNormals function to DerivedMesh struct, and implement it for CDDM and CCGDM (subsurf). EditDerivedBMesh (edit mode DM) only gets a dummy one in this commit. * Add a tessellated version of CD_LOOPNORMAL layer (CD_TESSLOOPNORMAL), with relevant code to handle it (tessellation, rna access, etc.). * Change auto_smooth options of Mesh (angle now in radian internaly, and toggle is now used to enable/disable split normals in DM creation process). Note BI render code is not touched here, hence its behavior regarding this option is now incoherent, will be addressed in a separate commit. Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D365
2014-04-13 12:18:51 +02:00
bmdm->dm.calcLoopNormals = emDM_calcLoopNormals;
Add Custom Loop Normals. This is the core code for it, tools (datatransfer and modifier) will come in next commits). RNA api is already there, though. See the code for details, but basically, we define, for each 'smooth fan' (which is a set of adjacent loops around a same vertex that are smooth, i.e. have a single same normal), a 'loop normal space' (or lnor space), using auto-computed normal and relevant edges, and store custom normal as two angular factors inside that space. This allows to have custom normals 'following' deformations of the geometry, and to only save two shorts per loop in new clnor CDLayer. Normal manipulation (editing, mixing, interpolating, etc.) shall always happen with plain 3D vectors normals, and be converted back into storage format at the end. Clnor computation has also been threaded (at least for Mesh case, not for BMesh), since the process can be rather heavy with high poly meshes. Also, bumping subversion, and fix mess in 2.70 versioning code.
2015-02-05 14:24:48 +01:00
bmdm->dm.calcLoopNormalsSpaceArray = emDM_calcLoopNormalsSpaceArray;
Support multiple tangents for BI render & viewport Normal Map node support for GLSL mode and the internal render (multiple tangents support). The Normal Map node is a useful node which is present in the Cycles render. It makes it possible to use normal mapping without additional material node in a node tree. This patch implements Normal Map node for GLSL mode and the internal render. Previously only the active UV layer was used to calculate tangents.
2016-04-26 18:43:02 +10:00
bmdm->dm.calcLoopTangents = emDM_calc_loop_tangents;
style cleanup - spelling - turns out we had tessellation spelt wrong all over. - use \directive for doxy (not @directive) - remove BLI_sparsemap.h - was from bmesh merge IIRC but entire file commented and not used.
2012-03-02 16:05:54 +00:00
bmdm->dm.recalcTessellation = emDM_recalcTessellation;
Add tessellation data to DerivedMesh (LoopTri) This stores loop indices into the loop array giving easier acess to data such as vertex-colors and UV's, removing the need to store an MFace duplicate of custom-data. This doesn't yet move all internal code from MFace to LoopTri just yet. Only applies to: - opengl drawing - sculpting (pbvh) - vertex/weight paint Thanks to @psy-fi for review, fixes and improvements to drawing!
2015-07-17 03:36:03 +10:00
bmdm->dm.recalcLoopTri = emDM_recalcLoopTri;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
bmdm->dm.foreachMappedVert = emDM_foreachMappedVert;
Fix T39781: Autosmooth feature: erroneously showing hidden vertices' normals too when in "vertex per face" display So, turned out after all we need a foreachmapped helper for loops as well... :/ CDDM and EBDM were reasonably simple, but I fought hours with CCGDM (subsurf) to find only a bad working solution (see D478). So I fallback to a code similar to CDDM one. Probably not nice for performances, but loops and subsurf are not matching well... Reviewers: campbellbarton CC: brecht Differential Revision: https://developer.blender.org/D478
2014-04-23 15:18:11 +02:00
bmdm->dm.foreachMappedLoop = emDM_foreachMappedLoop;
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
bmdm->dm.foreachMappedEdge = emDM_foreachMappedEdge;
bmdm->dm.foreachMappedFaceCenter = emDM_foreachMappedFaceCenter;
bmdm->dm.release = emDM_release;
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
bmdm->emd.vertexCos = (const float (*)[3])vertexCos;
fix [#36055] Edge/Face Info display on wrong position with key frames display editmesh stats with deform modifiers.
2013-07-09 12:59:31 +00:00
bmdm->dm.deformedOnly = (vertexCos != NULL);
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
EditMesh: Avoid creating deform-vert layer every redraw Getting a new edit-derived-bmesh was always creating a deform-vert array, even when it wasn't needed. Since this was called on redraw, in many cases it was doing it unnecessarily. Now pass in a custom-data mask and only fill in deform-verts when needed. Gives noticeable drawing speedup (~10-30% here).
2016-06-22 21:20:09 +10:00
const int cd_dvert_offset = (data_mask & CD_MASK_MDEFORMVERT) ?
CustomData_get_offset(&bm->vdata, CD_MDEFORMVERT) : -1;
bmesh optimization: use offsets for deform-vert lookups to avoid customdata layer lookups within loops.
2013-05-27 12:42:48 +00:00
if (cd_dvert_offset != -1) {
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
BMIter iter;
BMVert *eve;
int i;
DM_add_vert_layer(&bmdm->dm, CD_MDEFORMVERT, CD_CALLOC, NULL);
changes to derived mesh editmode functions to be in closer sync with trunk.
2011-12-02 03:18:34 +00:00
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
DM_set_vert_data(&bmdm->dm, i, CD_MDEFORMVERT,
bmesh optimization: use offsets for deform-vert lookups to avoid customdata layer lookups within loops.
2013-05-27 12:42:48 +00:00
BM_ELEM_CD_GET_VOID_P(eve, cd_dvert_offset));
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
}
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
EditMesh: Avoid creating deform-vert layer every redraw Getting a new edit-derived-bmesh was always creating a deform-vert array, even when it wasn't needed. Since this was called on redraw, in many cases it was doing it unnecessarily. Now pass in a custom-data mask and only fill in deform-verts when needed. Gives noticeable drawing speedup (~10-30% here).
2016-06-22 21:20:09 +10:00
const int cd_skin_offset = (data_mask & CD_MASK_MVERT_SKIN) ?
CustomData_get_offset(&bm->vdata, CD_MVERT_SKIN) : -1;
bmesh optimization: use offsets for deform-vert lookups to avoid customdata layer lookups within loops.
2013-05-27 12:42:48 +00:00
if (cd_skin_offset != -1) {
Draw skin vertex roots in edit mode. Roots are drawn with a view-oriented dashed red circle around the vertex location. The circle's radius is the average if the skin vertex's X and Y radii. Skin modifier documentation: http://wiki.blender.org/index.php/User:Nicholasbishop/SkinModifier
2012-05-22 15:29:27 +00:00
BMIter iter;
BMVert *eve;
int i;
DM_add_vert_layer(&bmdm->dm, CD_MVERT_SKIN, CD_CALLOC, NULL);
code cleanup: edit-derived-mesh, define 'BMesh *bm' as local variable, avoid 'bmdm->em->bm' within functions.
2013-04-13 20:58:49 +00:00
BM_ITER_MESH_INDEX (eve, &iter, bm, BM_VERTS_OF_MESH, i) {
Draw skin vertex roots in edit mode. Roots are drawn with a view-oriented dashed red circle around the vertex location. The circle's radius is the average if the skin vertex's X and Y radii. Skin modifier documentation: http://wiki.blender.org/index.php/User:Nicholasbishop/SkinModifier
2012-05-22 15:29:27 +00:00
DM_set_vert_data(&bmdm->dm, i, CD_MVERT_SKIN,
bmesh optimization: use offsets for deform-vert lookups to avoid customdata layer lookups within loops.
2013-05-27 12:42:48 +00:00
BM_ELEM_CD_GET_VOID_P(eve, cd_skin_offset));
Draw skin vertex roots in edit mode. Roots are drawn with a view-oriented dashed red circle around the vertex location. The circle's radius is the average if the skin vertex's X and Y radii. Skin modifier documentation: http://wiki.blender.org/index.php/User:Nicholasbishop/SkinModifier
2012-05-22 15:29:27 +00:00
}
}
code cleanup: - replace iterators with macros - move vertexCos checks outside the for loops (use 2 for loops). - style cleanup
2012-05-03 21:19:31 +00:00
return (DerivedMesh *)bmdm;
NOTE: do not test. work-in-progress commit with editmesh ripped out and replaced with bmesh. this is not usable by any means. for those who read through this, note the design is still fairly messy in places, and fyi BMTessMesh is the replacement for EditMesh, I need to rename it to BMEditMesh.
2009-05-16 16:18:08 +00:00
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* -------------------------------------------------------------------- */
/* StatVis Functions */
static void axis_from_enum_v3(float v[3], const char axis)
{
zero_v3(v);
if (axis < 3) v[axis] = 1.0f;
else v[axis - 3] = -1.0f;
}
static void statvis_calc_overhang(
BMEditMesh *em,
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
const float (*polyNos)[3],
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* values for calculating */
const float min, const float max, const char axis,
/* result */
unsigned char (*r_face_colors)[4])
{
BMIter iter;
BMesh *bm = em->bm;
BMFace *f;
float dir[3];
int index;
const float minmax_irange = 1.0f / (max - min);
Fix T38795: Mesh Analysis overhang faces show gray
2014-02-24 19:53:44 +11:00
bool is_max;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* fallback */
Fix T39351: Blender crashes when enabling Mesh Analysis. This code was using a //const// char array for fallback colors and then writing to it (hidden to the compiler by explicit casting). Bad!
2014-03-23 11:35:52 +01:00
unsigned char col_fallback[4] = {64, 64, 64, 255}; /* gray */
unsigned char col_fallback_max[4] = {0, 0, 0, 255}; /* max color */
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
BLI_assert(min <= max);
axis_from_enum_v3(dir, axis);
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
if (LIKELY(em->ob)) {
mul_transposed_mat3_m4_v3(em->ob->obmat, dir);
normalize_v3(dir);
}
Fix T38795: Mesh Analysis overhang faces show gray
2014-02-24 19:53:44 +11:00
/* fallback max */
{
float fcol[3];
weight_to_rgb(fcol, 1.0f);
Fix T39351: Blender crashes when enabling Mesh Analysis. This code was using a //const// char array for fallback colors and then writing to it (hidden to the compiler by explicit casting). Bad!
2014-03-23 11:35:52 +01:00
rgb_float_to_uchar(col_fallback_max, fcol);
Fix T38795: Mesh Analysis overhang faces show gray
2014-02-24 19:53:44 +11:00
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* now convert into global space */
BM_ITER_MESH_INDEX (f, &iter, bm, BM_FACES_OF_MESH, index) {
float fac = angle_normalized_v3v3(polyNos ? polyNos[index] : f->no, dir) / (float)M_PI;
/* remap */
Fix T38795: Mesh Analysis overhang faces show gray
2014-02-24 19:53:44 +11:00
if ((is_max = (fac <= max)) && (fac >= min)) {
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
float fcol[3];
fac = (fac - min) * minmax_irange;
fac = 1.0f - fac;
CLAMP(fac, 0.0f, 1.0f);
weight_to_rgb(fcol, fac);
rgb_float_to_uchar(r_face_colors[index], fcol);
}
else {
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
const unsigned char *fallback = is_max ? col_fallback_max : col_fallback;
copy_v4_v4_uchar(r_face_colors[index], fallback);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
}
}
/* so we can use jitter values for face interpolation */
static void uv_from_jitter_v2(float uv[2])
{
uv[0] += 0.5f;
uv[1] += 0.5f;
if (uv[0] + uv[1] > 1.0f) {
uv[0] = 1.0f - uv[0];
uv[1] = 1.0f - uv[1];
}
CLAMP(uv[0], 0.0f, 1.0f);
CLAMP(uv[1], 0.0f, 1.0f);
}
static void statvis_calc_thickness(
BMEditMesh *em,
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
const float (*vertexCos)[3],
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* values for calculating */
const float min, const float max, const int samples,
/* result */
unsigned char (*r_face_colors)[4])
{
tweak for epsilon of intersection check, was too small for some cases.
2013-04-28 22:35:08 +00:00
const float eps_offset = 0.00002f; /* values <= 0.00001 give errors */
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
float *face_dists = (float *)r_face_colors; /* cheating */
const bool use_jit = samples < 32;
float jit_ofs[32][2];
BMesh *bm = em->bm;
const int tottri = em->tottri;
const float minmax_irange = 1.0f / (max - min);
int i;
struct BMLoop *(*looptris)[3] = em->looptris;
/* fallback */
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
const unsigned char col_fallback[4] = {64, 64, 64, 255};
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
struct BMBVHTree *bmtree;
BLI_assert(min <= max);
Math Lib: rename fill_*, to copy_* matching convention for fixed length api, eg: copy_v3_fl
2015-05-05 17:08:29 +10:00
copy_vn_fl(face_dists, em->bm->totface, max);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
if (use_jit) {
int j;
BLI_assert(samples < 32);
Code cleanup: jitter, use 2d float array
2014-03-18 10:50:24 +11:00
BLI_jitter_init(jit_ofs, samples);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
for (j = 0; j < samples; j++) {
uv_from_jitter_v2(jit_ofs[j]);
}
}
BM_mesh_elem_index_ensure(bm, BM_FACE);
if (vertexCos) {
BM_mesh_elem_index_ensure(bm, BM_VERT);
}
BMesh Refactor: BKE_bmbvh_new can now be created without an EditMesh. This adds BM_bmesh_calc_tessellation() so we can get triangles from a bmesh without having to have an editmesh available.
2013-11-18 18:20:21 +11:00
bmtree = BKE_bmbvh_new_from_editmesh(em, 0, vertexCos, false);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
for (i = 0; i < tottri; i++) {
BMFace *f_hit;
BMLoop **ltri = looptris[i];
const int index = BM_elem_index_get(ltri[0]->f);
const float *cos[3];
float ray_co[3];
float ray_no[3];
if (vertexCos) {
cos[0] = vertexCos[BM_elem_index_get(ltri[0]->v)];
cos[1] = vertexCos[BM_elem_index_get(ltri[1]->v)];
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
cos[2] = vertexCos[BM_elem_index_get(ltri[2]->v)];
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
else {
cos[0] = ltri[0]->v->co;
cos[1] = ltri[1]->v->co;
cos[2] = ltri[2]->v->co;
}
normal_tri_v3(ray_no, cos[2], cos[1], cos[0]);
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
#define FACE_RAY_TEST_ANGLE \
Some knife fixes. Avoids duplicating verts; better handling of cut-through ortho. Now cut lines detect vertices that they pass (almost) exactly over and snap to them, to avoid making verts vert close to other ones. Added radius arg to BKE_bmbvh_ray_cast so that can detect an obscuring face when the ray might otherwise go exactly between two triangles. Needed an isect_line_tri_epsilon function for similar reason. Fixes last part of bug #35002. Other knife bugs still present but getting this commit in now before continuing bug fixing.
2013-08-30 16:34:44 +00:00
f_hit = BKE_bmbvh_ray_cast(bmtree, ray_co, ray_no, 0.0f, \
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
&dist, NULL, NULL); \
skip checks in statvis_calc_thickness(). also remove paranoid NULL checks in smallhash.c
2013-04-20 16:49:02 +00:00
if (f_hit && dist < face_dists[index]) { \
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
float angle_fac = fabsf(dot_v3v3(ltri[0]->f->no, f_hit->no)); \
angle_fac = 1.0f - angle_fac; \
angle_fac = angle_fac * angle_fac * angle_fac; \
angle_fac = 1.0f - angle_fac; \
dist /= angle_fac; \
if (dist < face_dists[index]) { \
face_dists[index] = dist; \
} \
} (void)0
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
if (use_jit) {
int j;
for (j = 0; j < samples; j++) {
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
float dist = face_dists[index];
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
interp_v3_v3v3v3_uv(ray_co, cos[0], cos[1], cos[2], jit_ofs[j]);
madd_v3_v3fl(ray_co, ray_no, eps_offset);
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
FACE_RAY_TEST_ANGLE;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
}
else {
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
float dist = face_dists[index];
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
mid_v3_v3v3v3(ray_co, cos[0], cos[1], cos[2]);
madd_v3_v3fl(ray_co, ray_no, eps_offset);
fix [#35025] Mesh Debug samples strange behavior
2013-04-20 16:29:37 +00:00
FACE_RAY_TEST_ANGLE;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
}
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
BKE_bmbvh_free(bmtree);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* convert floats into color! */
for (i = 0; i < bm->totface; i++) {
float fac = face_dists[i];
/* important not '<=' */
if (fac < max) {
float fcol[3];
fac = (fac - min) * minmax_irange;
fac = 1.0f - fac;
CLAMP(fac, 0.0f, 1.0f);
weight_to_rgb(fcol, fac);
rgb_float_to_uchar(r_face_colors[i], fcol);
}
else {
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
copy_v4_v4_uchar(r_face_colors[i], col_fallback);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
}
}
static void statvis_calc_intersect(
BMEditMesh *em,
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
const float (*vertexCos)[3],
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* result */
unsigned char (*r_face_colors)[4])
{
BMesh *bm = em->bm;
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
int i;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
/* fallback */
// const char col_fallback[4] = {64, 64, 64, 255};
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
float fcol[3];
unsigned char col[3];
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
struct BMBVHTree *bmtree;
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
BVHTreeOverlap *overlap;
unsigned int overlap_len;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
memset(r_face_colors, 64, sizeof(int) * em->bm->totface);
BM_mesh_elem_index_ensure(bm, BM_FACE);
if (vertexCos) {
BM_mesh_elem_index_ensure(bm, BM_VERT);
}
BMesh Refactor: BKE_bmbvh_new can now be created without an EditMesh. This adds BM_bmesh_calc_tessellation() so we can get triangles from a bmesh without having to have an editmesh available.
2013-11-18 18:20:21 +11:00
bmtree = BKE_bmbvh_new_from_editmesh(em, 0, vertexCos, false);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
overlap = BKE_bmbvh_overlap(bmtree, bmtree, &overlap_len);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
/* same for all faces */
weight_to_rgb(fcol, 1.0f);
rgb_float_to_uchar(col, fcol);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
if (overlap) {
for (i = 0; i < overlap_len; i++) {
BMFace *f_hit_pair[2] = {
em->looptris[overlap[i].indexA][0]->f,
em->looptris[overlap[i].indexB][0]->f,
};
int j;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
for (j = 0; j < 2; j++) {
BMFace *f_hit = f_hit_pair[j];
int index;
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
index = BM_elem_index_get(f_hit);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
copy_v3_v3_uchar(r_face_colors[index], col);
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
Use BVH-overlap for mesh self-intersection display Add BKE_bmbvh_overlap, remove BKE_bmbvh_find_face_segment
2015-08-20 19:09:20 +10:00
MEM_freeN(overlap);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
BKE_bmbvh_free(bmtree);
}
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
static void statvis_calc_distort(
BMEditMesh *em,
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
const float (*vertexCos)[3], const float (*polyNos)[3],
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
/* values for calculating */
const float min, const float max,
/* result */
unsigned char (*r_face_colors)[4])
{
BMIter iter;
BMesh *bm = em->bm;
BMFace *f;
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
const float *f_no;
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
int index;
const float minmax_irange = 1.0f / (max - min);
/* fallback */
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
const unsigned char col_fallback[4] = {64, 64, 64, 255};
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
/* now convert into global space */
BM_ITER_MESH_INDEX (f, &iter, bm, BM_FACES_OF_MESH, index) {
float fac;
if (f->len == 3) {
fac = -1.0f;
}
else {
BMLoop *l_iter, *l_first;
if (vertexCos) {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
f_no = polyNos[index];
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
}
else {
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
f_no = f->no;
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
}
fac = 0.0f;
l_iter = l_first = BM_FACE_FIRST_LOOP(f);
do {
float no_corner[3];
if (vertexCos) {
normal_tri_v3(no_corner,
vertexCos[BM_elem_index_get(l_iter->prev->v)],
vertexCos[BM_elem_index_get(l_iter->v)],
vertexCos[BM_elem_index_get(l_iter->next->v)]);
}
else {
BMesh: move safe loop-normal-calc to own function This makes sense when we want to avoid float precision error for near co-linear edges. OTOH, this is an arbitrary decision, so keep functions separate.
2017-07-12 23:06:59 +10:00
BM_loop_calc_face_normal_safe(l_iter, no_corner);
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
}
fix range for face distortion and display for concave ngons.
2013-04-18 09:12:06 +00:00
/* simple way to detect (what is most likely) concave */
if (dot_v3v3(f_no, no_corner) < 0.0f) {
negate_v3(no_corner);
}
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
fac = max_ff(fac, angle_normalized_v3v3(f_no, no_corner));
} while ((l_iter = l_iter->next) != l_first);
fix range for face distortion and display for concave ngons.
2013-04-18 09:12:06 +00:00
fac *= 2.0f;
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
}
/* remap */
if (fac >= min) {
float fcol[3];
fac = (fac - min) * minmax_irange;
CLAMP(fac, 0.0f, 1.0f);
weight_to_rgb(fcol, fac);
rgb_float_to_uchar(r_face_colors[index], fcol);
}
else {
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
copy_v4_v4_uchar(r_face_colors[index], col_fallback);
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
}
}
}
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
static void statvis_calc_sharp(
BMEditMesh *em,
const float (*vertexCos)[3],
/* values for calculating */
const float min, const float max,
/* result */
unsigned char (*r_vert_colors)[4])
{
float *vert_angles = (float *)r_vert_colors; /* cheating */
BMIter iter;
BMesh *bm = em->bm;
BMEdge *e;
//float f_no[3];
const float minmax_irange = 1.0f / (max - min);
int i;
/* fallback */
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
const unsigned char col_fallback[4] = {64, 64, 64, 255};
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
(void)vertexCos; /* TODO */
Math Lib: rename fill_*, to copy_* matching convention for fixed length api, eg: copy_v3_fl
2015-05-05 17:08:29 +10:00
copy_vn_fl(vert_angles, em->bm->totvert, -M_PI);
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
/* first assign float values to verts */
BM_ITER_MESH (e, &iter, bm, BM_EDGES_OF_MESH) {
float angle = BM_edge_calc_face_angle_signed(e);
float *col1 = &vert_angles[BM_elem_index_get(e->v1)];
float *col2 = &vert_angles[BM_elem_index_get(e->v2)];
*col1 = max_ff(*col1, angle);
*col2 = max_ff(*col2, angle);
}
/* convert floats into color! */
for (i = 0; i < bm->totvert; i++) {
float fac = vert_angles[i];
/* important not '<=' */
if (fac > min) {
float fcol[3];
fac = (fac - min) * minmax_irange;
CLAMP(fac, 0.0f, 1.0f);
weight_to_rgb(fcol, fac);
rgb_float_to_uchar(r_vert_colors[i], fcol);
}
else {
Fix warnings and remove casts by adding copy_vx_vx_uchar() functions.
2015-10-11 01:44:47 +02:00
copy_v4_v4_uchar(r_vert_colors[i], col_fallback);
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
}
}
}
Cleanup: ws
2015-07-18 05:54:31 +10:00
void BKE_editmesh_statvis_calc(
BMEditMesh *em, DerivedMesh *dm,
const MeshStatVis *statvis)
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
{
EditDerivedBMesh *bmdm = (EditDerivedBMesh *)dm;
BLI_assert(dm == NULL || dm->type == DM_TYPE_EDITBMESH);
switch (statvis->type) {
case SCE_STATVIS_OVERHANG:
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
{
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
BKE_editmesh_color_ensure(em, BM_FACE);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
statvis_calc_overhang(
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
em, bmdm ? bmdm->emd.polyNos : NULL,
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
statvis->overhang_min / (float)M_PI,
statvis->overhang_max / (float)M_PI,
statvis->overhang_axis,
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
em->derivedFaceColor);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
break;
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
case SCE_STATVIS_THICKNESS:
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
{
const float scale = 1.0f / mat4_to_scale(em->ob->obmat);
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
BKE_editmesh_color_ensure(em, BM_FACE);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
statvis_calc_thickness(
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
em, bmdm ? bmdm->emd.vertexCos : NULL,
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
statvis->thickness_min * scale,
statvis->thickness_max * scale,
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
statvis->thickness_samples,
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
em->derivedFaceColor);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
break;
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
case SCE_STATVIS_INTERSECT:
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
{
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
BKE_editmesh_color_ensure(em, BM_FACE);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
statvis_calc_intersect(
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
em, bmdm ? bmdm->emd.vertexCos : NULL,
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
em->derivedFaceColor);
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
break;
Support object scale for mesh display options. (currently only uniform scale for thickness test)
2013-04-17 23:30:19 +00:00
}
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
case SCE_STATVIS_DISTORT:
{
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
BKE_editmesh_color_ensure(em, BM_FACE);
optimization: lazy initialize EditDerivedBMesh members vertexNos, polyNos. also add polyCos array which cache's face centers, gives overall ~20% speedup to drawing on a high-poly mesh in face-editmode.
2013-07-22 16:49:37 +00:00
if (bmdm)
emDM_ensurePolyNormals(bmdm);
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
statvis_calc_distort(
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
em, bmdm ? bmdm->emd.vertexCos : NULL, bmdm ? bmdm->emd.polyNos : NULL,
fix range for face distortion and display for concave ngons.
2013-04-18 09:12:06 +00:00
statvis->distort_min,
statvis->distort_max,
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
em->derivedFaceColor);
break;
}
case SCE_STATVIS_SHARP:
{
BKE_editmesh_color_ensure(em, BM_VERT);
statvis_calc_sharp(
Modifiers: Simple Deform & Build, DerivedMesh → Mesh This commit introduces `EditMeshData`. The fields in this struct are extracted from `EditDerivedBMesh` into their own struct `EditMeshData`, which can then also be used by the `Mesh` struct. This allows passing deformed vertices efficiently to the draw routines. The modifier code constructs a new Mesh instead of writing to ob->data; even when ob->data is a CoW copy, it can still be used by different objects and thus shouldn't be modified by a modifier.
2018-04-19 11:03:58 +02:00
em, bmdm ? bmdm->emd.vertexCos : NULL,
add display mode for sharp edges.
2013-04-18 17:09:56 +00:00
statvis->sharp_min,
statvis->sharp_max,
/* in this case they are vertex colors */
em->derivedVertColor);
add mesh distort display mode (highlights distorted faces)
2013-04-18 04:24:18 +00:00
break;
}
display options to help with 3d printing. editmesh debug info, - overhang (with axis angle options) - wall thickness (with min/max distance) - self-intersections. access below 'Mesh Display' panel.
2013-04-17 09:27:23 +00:00
}
}
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
/* -------------------------------------------------------------------- */
/* Editmesh Vert Coords */
struct CageUserData {
int totvert;
float (*cos_cage)[3];
remove the pointer from BLI_bitmap's typedef, hides that an arg passed is really an array which may be modified by other functions.
2013-07-22 23:20:48 +00:00
BLI_bitmap *visit_bitmap;
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
};
Cleanup: ws
2015-07-18 05:54:31 +10:00
static void cage_mapped_verts_callback(
void *userData, int index, const float co[3],
const float UNUSED(no_f[3]), const short UNUSED(no_s[3]))
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
{
struct CageUserData *data = userData;
BLI_bitmap: rename macros - BLI_BITMAP_SET -> BLI_BITMAP_ENABLE - BLI_BITMAP_CLEAR -> BLI_BITMAP_DISABLE - BLI_BITMAP_GET -> BLI_BITMAP_TEST - BLI_BITMAP_MODIFY -> BLI_BITMAP_SET
2014-06-06 16:05:15 +10:00
if ((index >= 0 && index < data->totvert) && (!BLI_BITMAP_TEST(data->visit_bitmap, index))) {
BLI_BITMAP_ENABLE(data->visit_bitmap, index);
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
copy_v3_v3(data->cos_cage[index], co);
}
}
Depsgraph: remove EvaluationContext, pass Depsgraph instead. The depsgraph was always created within a fixed evaluation context. Passing both risks the depsgraph and evaluation context not matching, and it complicates the Python API where we'd have to expose both which is not so easy to understand. This also removes the global evaluation context in main, which assumed there to be a single active scene and view layer. Differential Revision: https://developer.blender.org/D3152
2018-04-06 12:07:27 +02:00
float (*BKE_editmesh_vertexCos_get(struct Depsgraph *depsgraph, BMEditMesh *em, Scene *scene, int *r_numVerts))[3]
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
{
DerivedMesh *cage, *final;
remove the pointer from BLI_bitmap's typedef, hides that an arg passed is really an array which may be modified by other functions.
2013-07-22 23:20:48 +00:00
BLI_bitmap *visit_bitmap;
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
struct CageUserData data;
float (*cos_cage)[3];
Depsgraph: remove EvaluationContext, pass Depsgraph instead. The depsgraph was always created within a fixed evaluation context. Passing both risks the depsgraph and evaluation context not matching, and it complicates the Python API where we'd have to expose both which is not so easy to understand. This also removes the global evaluation context in main, which assumed there to be a single active scene and view layer. Differential Revision: https://developer.blender.org/D3152
2018-04-06 12:07:27 +02:00
cage = editbmesh_get_derived_cage_and_final(depsgraph, scene, em->ob, em, CD_MASK_BAREMESH, &final);
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
cos_cage = MEM_callocN(sizeof(*cos_cage) * em->bm->totvert, "bmbvh cos_cage");
/* when initializing cage verts, we only want the first cage coordinate for each vertex,
* so that e.g. mirror or array use original vertex coordinates and not mirrored or duplicate */
visit_bitmap = BLI_BITMAP_NEW(em->bm->totvert, __func__);
data.totvert = em->bm->totvert;
data.cos_cage = cos_cage;
data.visit_bitmap = visit_bitmap;
optimization: only calculate the normals for passing into derivedMesh foreachMappedVert/foreachMappedFaceCenter when needed, this means in editmode with wire draw, face and vertex normals don't have to be calculated at all. in most cases the normals are not used so add a flag that makes calculating them only for functions that need them. also fix face normal calculation for CDDM, was using quad calculation for ngons too.
2013-07-22 22:59:47 +00:00
cage->foreachMappedVert(cage, cage_mapped_verts_callback, &data, DM_FOREACH_NOP);
use modified vertex coords for calculating display thickness and intersections. internal improvement to editmesh_bvh.c - optionally pass cage-coords as an arg, rather then calculating the coords in BKE_bmbvh_new(), since all callers already have coords calculated. - de-duplicate coords creation function from knife and bmbvhm, move into own generic function: BKE_editmesh_vertexCos_get()
2013-04-18 01:20:04 +00:00
MEM_freeN(visit_bitmap);
if (r_numVerts) {
*r_numVerts = em->bm->totvert;
}
return cos_cage;
}
Copy Permalink