[Bf-blender-cvs] SVN commit: /data/svn/bf-blender [59868] trunk/blender/source/blender/bmesh /operators/bmo_fill_grid.c: grid-fill support for customdata, so filled areas have UV's, vertex colors etc interpolated from the boundary.
Campbell Barton
ideasman42 at gmail.com
Fri Sep 6 08:48:17 CEST 2013
Revision: 59868
http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=59868
Author: campbellbarton
Date: 2013-09-06 06:48:17 +0000 (Fri, 06 Sep 2013)
Log Message:
-----------
grid-fill support for customdata, so filled areas have UV's, vertex colors etc interpolated from the boundary.
also support cases where only some of the boundary edges have faces to interpolate from.
Modified Paths:
--------------
trunk/blender/source/blender/bmesh/operators/bmo_fill_grid.c
Modified: trunk/blender/source/blender/bmesh/operators/bmo_fill_grid.c
===================================================================
--- trunk/blender/source/blender/bmesh/operators/bmo_fill_grid.c 2013-09-06 06:27:22 UTC (rev 59867)
+++ trunk/blender/source/blender/bmesh/operators/bmo_fill_grid.c 2013-09-06 06:48:17 UTC (rev 59868)
@@ -103,7 +103,95 @@
#endif
+/* -------------------------------------------------------------------- */
+/* Handle Loop Pairs */
+
+/** \name Loop Pairs
+ * \{ */
+
/**
+ * Assign a loop pair from 2 verts (which _must_ share an edge)
+ */
+static void bm_loop_pair_from_verts(BMVert *v_a, BMVert *v_b,
+ BMLoop *l_pair[2])
+{
+ BMEdge *e = BM_edge_exists(v_a, v_b);
+ if (e->l) {
+ if (e->l->v == v_a) {
+ l_pair[0] = e->l;
+ l_pair[1] = e->l->next;
+ }
+ else {
+ l_pair[0] = e->l->next;
+ l_pair[1] = e->l;
+ }
+ }
+ else {
+ l_pair[0] = NULL;
+ l_pair[1] = NULL;
+ }
+}
+
+/**
+ * Copy loop pair from one side to the other if either is missing,
+ * this simplifies interpolation code so we only need to check if x/y are missing,
+ * rather then checking each loop.
+ */
+static void bm_loop_pair_test_copy(BMLoop *l_pair_a[2], BMLoop *l_pair_b[2])
+{
+ /* if the first one is set, we know the second is too */
+ if (l_pair_a[0] && l_pair_b[0] == NULL) {
+ l_pair_b[0] = l_pair_a[1];
+ l_pair_b[1] = l_pair_a[0];
+ }
+ else if (l_pair_b[0] && l_pair_a[0] == NULL) {
+ l_pair_a[0] = l_pair_b[1];
+ l_pair_a[1] = l_pair_b[0];
+ }
+}
+
+/**
+ * Interpolate from boundary loops.
+ *
+ * \note These weights will be calculated multiple times per vertex.
+ * Since this runs on each loop, could reduce calls to #barycentric_weights_v2_quad.
+ */
+static void bm_loop_interp_from_grid_boundary_4(BMesh *bm, BMLoop *l, BMLoop *l_bound[4], const float uv[2])
+{
+ float cos[4][2] = {
+ {uv[0], 0},
+ {0, uv[1]},
+ {uv[0], 1},
+ {1, uv[1]}};
+
+ void *l_cdata[4] = {
+ l_bound[0]->head.data,
+ l_bound[1]->head.data,
+ l_bound[2]->head.data,
+ l_bound[3]->head.data};
+
+ float w[4];
+
+ barycentric_weights_v2_quad(UNPACK4(cos), uv, w);
+ CustomData_bmesh_interp(&bm->ldata, l_cdata, w, NULL, 4, l->head.data);
+}
+
+static void bm_loop_interp_from_grid_boundary_2(BMesh *bm, BMLoop *l, BMLoop *l_bound[2], const float t)
+{
+
+ void *l_cdata[2] = {
+ l_bound[0]->head.data,
+ l_bound[1]->head.data};
+
+ const float w[2] = {1.0f - t, t};
+
+ CustomData_bmesh_interp(&bm->ldata, l_cdata, w, NULL, 2, l->head.data);
+}
+
+/** \} */
+
+
+/**
* This may be useful outside the bmesh operator.
*
* \param v_grid 2d array of verts, all boundary verts must be set, we fill in the middle.
@@ -113,8 +201,12 @@
const bool use_flip)
{
const bool use_vert_interp = CustomData_has_interp(&bm->vdata);
+ const bool use_loop_interp = CustomData_has_interp(&bm->ldata);
int x, y;
+ /* for use_loop_interp */
+ BMLoop *((*larr_x_a)[2]), *((*larr_x_b)[2]), *((*larr_y_a)[2]), *((*larr_y_b)[2]);
+
#define XY(_x, _y) ((_x) + ((_y) * (xtot)))
#ifdef BARYCENTRIC_INTERP
@@ -141,6 +233,31 @@
true);
#endif
+
+ /* Store loops */
+ if (use_loop_interp) {
+ /* x2 because each edge connects 2 loops */
+ larr_x_a = MEM_mallocN(sizeof(*larr_x_a) * (xtot - 1), __func__);
+ larr_x_b = MEM_mallocN(sizeof(*larr_x_b) * (xtot - 1), __func__);
+
+ larr_y_a = MEM_mallocN(sizeof(*larr_y_a) * (ytot - 1), __func__);
+ larr_y_b = MEM_mallocN(sizeof(*larr_y_b) * (ytot - 1), __func__);
+
+ /* fill in the loops */
+ for (x = 0; x < xtot - 1; x++) {
+ bm_loop_pair_from_verts(v_grid[XY(x, 0)], v_grid[XY(x + 1, 0)], larr_x_a[x]);
+ bm_loop_pair_from_verts(v_grid[XY(x, ytot - 1)], v_grid[XY(x + 1, ytot - 1)], larr_x_b[x]);
+ bm_loop_pair_test_copy(larr_x_a[x], larr_x_b[x]);
+ }
+
+ for (y = 0; y < ytot - 1; y++) {
+ bm_loop_pair_from_verts(v_grid[XY(0, y)], v_grid[XY(0, y + 1)], larr_y_a[y]);
+ bm_loop_pair_from_verts(v_grid[XY(xtot - 1, y)], v_grid[XY(xtot - 1, y + 1)], larr_y_b[y]);
+ bm_loop_pair_test_copy(larr_y_a[y], larr_y_b[y]);
+ }
+ }
+
+
/* Build Verts */
for (y = 1; y < ytot - 1; y++) {
#ifdef BARYCENTRIC_INTERP
@@ -230,6 +347,88 @@
NULL,
BM_CREATE_NOP);
}
+
+
+ if (use_loop_interp && (larr_x_a[x][0] || larr_y_a[y][0])) {
+ /* bottom/left/top/right */
+ BMLoop *l_quad[4];
+ BMLoop *l_bound[4];
+ BMLoop *l_tmp;
+ float uv[2]; /* xy in the grid */
+ int x_side, y_side, i;
+ char interp_from;
+
+
+ if (larr_x_a[x][0] && larr_y_a[y][0]) {
+ interp_from = 'B'; /* B == both */
+ l_tmp = larr_x_a[x][0];
+ }
+ else if (larr_x_a[x][0]) {
+ interp_from = 'X';
+ l_tmp = larr_x_a[x][0];
+ }
+ else {
+ interp_from = 'Y';
+ l_tmp = larr_y_a[y][0];
+ }
+
+ BM_elem_attrs_copy(bm, bm, l_tmp->f, f);
+
+
+ BM_face_as_array_loop_quad(f, l_quad);
+
+ l_tmp = BM_FACE_FIRST_LOOP(f);
+
+ if (use_flip) {
+ l_quad[0] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[1] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[3] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[2] = l_tmp;
+ }
+ else {
+ l_quad[2] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[3] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[1] = l_tmp; l_tmp = l_tmp->next;
+ l_quad[0] = l_tmp;
+ }
+
+ i = 0;
+
+ for (x_side = 0; x_side < 2; x_side++) {
+ for (y_side = 0; y_side < 2; y_side++) {
+
+ uv[0] = (float)(x + x_side) / (xtot - 1);
+ uv[1] = (float)(y + y_side) / (ytot - 1);
+
+ if (interp_from == 'B') {
+ l_bound[0] = larr_x_a[x][x_side]; /* B */
+ l_bound[1] = larr_y_a[y][y_side]; /* L */
+ l_bound[2] = larr_x_b[x][x_side]; /* T */
+ l_bound[3] = larr_y_b[y][y_side]; /* R */
+
+ bm_loop_interp_from_grid_boundary_4(bm, l_quad[i++], l_bound, uv);
+ }
+ else if (interp_from == 'X') {
+ l_bound[0] = larr_x_a[x][x_side]; /* B */
+ l_bound[1] = larr_x_b[x][x_side]; /* T */
+
+ bm_loop_interp_from_grid_boundary_2(bm, l_quad[i++], l_bound, uv[1]);
+ }
+ else if (interp_from == 'Y') {
+ l_bound[0] = larr_y_a[y][y_side]; /* L */
+ l_bound[1] = larr_y_b[y][y_side]; /* R */
+
+ bm_loop_interp_from_grid_boundary_2(bm, l_quad[i++], l_bound, uv[0]);
+ }
+ else {
+ BLI_assert(0);
+ }
+ }
+ }
+ }
+ /* end interp */
+
+
BMO_elem_flag_enable(bm, f, FACE_OUT);
f->mat_nr = mat_nr;
if (use_smooth) {
@@ -237,6 +436,14 @@
}
}
}
+
+ if (use_loop_interp) {
+ MEM_freeN(larr_x_a);
+ MEM_freeN(larr_y_a);
+ MEM_freeN(larr_x_b);
+ MEM_freeN(larr_y_b);
+ }
+
#undef XY
}
More information about the Bf-blender-cvs
mailing list