def execute(self, context, scene, ray_origin, ray_vector, is_start):
data = scene.sprytile_data
grid = sprytile_utils.get_grid(context, context.object.sprytile_gridid)
tile_xy = (grid.tile_selection[0], grid.tile_selection[1])
# Get vectors for grid, without rotation
up_vector, right_vector, plane_normal = sprytile_utils.get_current_grid_vectors(
scene, with_rotation=False)
# If building on decal layer, modify plane normal to the one under mouse
if data.work_layer == 'DECAL_1' and data.lock_normal is False:
location, hit_normal, face_index, distance = self.modal.raycast_object(
context.object, ray_origin, ray_vector)
if hit_normal is not None:
face_up, face_right = self.modal.get_face_up_vector(
context, face_index, 0.4, bias_right=True)
if face_up is not None and face_right is not None:
plane_normal = hit_normal
up_vector = face_up
right_vector = face_right
# Rotate the vectors
rotation = Quaternion(plane_normal, data.mesh_rotate)
up_vector = rotation * up_vector
right_vector = rotation * right_vector
# raycast grid to get the grid position under the mouse
grid_coord, grid_right, grid_up, plane_pos = sprytile_utils.raycast_grid(
scene,
context,
up_vector,
right_vector,
plane_normal,
ray_origin,
ray_vector,
as_coord=True)
# Record starting grid position of stroke
if is_start:
self.start_coord = grid_coord
# Not starting stroke, filter out when can build
elif self.start_coord is not None:
start_offset = (grid_coord[0] - self.start_coord[0],
grid_coord[1] - self.start_coord[1])
coord_mod = (start_offset[0] % grid.tile_selection[2],
start_offset[1] % grid.tile_selection[3])
# Isn't at exact position for grid made by tile selection, with start_coord as origin
if coord_mod[0] > 0 or coord_mod[1] > 0:
# Try to snap grid_coord
tolerance_min = (floor(grid.tile_selection[2] * 0.25),
floor(grid.tile_selection[3] * 0.25))
tolerance_max = (grid.tile_selection[2] - tolerance_min[0],
grid.tile_selection[3] - tolerance_min[1])
allow_snap_x = tolerance_min[0] <= coord_mod[
0] <= tolerance_max[0]
allow_snap_y = tolerance_min[1] <= coord_mod[
1] <= tolerance_max[1]
# If neither x or y can be snapped, return
if not allow_snap_x and not allow_snap_y:
return
coord_frac = [
start_offset[0] / grid.tile_selection[2],
start_offset[1] / grid.tile_selection[3]
]
if coord_mod[0] > (grid.tile_selection[2] / 2.0):
coord_frac[0] = ceil(coord_frac[0])
else:
coord_frac[0] = floor(coord_frac[0])
if coord_mod[1] > (grid.tile_selection[3] / 2.0):
coord_frac[1] = ceil(coord_frac[1])
else:
coord_frac[1] = floor(coord_frac[1])
grid_coord = (self.start_coord[0] +
(coord_frac[0] * grid.tile_selection[2]),
self.start_coord[1] +
(coord_frac[1] * grid.tile_selection[3]))
# Get the area to build
offset_tile_id, offset_grid, coord_min, coord_max = sprytile_utils.get_grid_area(
grid.tile_selection[2], grid.tile_selection[3], data.uv_flip_x,
data.uv_flip_y)
# Check if joining multi tile faces
grid_no_spacing = sprytile_utils.grid_no_spacing(grid)
is_single_pixel = sprytile_utils.grid_is_single_pixel(grid)
do_join = is_single_pixel
if do_join is False:
do_join = grid_no_spacing and data.auto_join
# 1x1 tile selections cannot be auto joined
tile_area = grid.tile_selection[2] * grid.tile_selection[3]
if do_join and tile_area == 1:
do_join = False
# Store vertices of constructed faces for cursor flow
faces_verts = []
require_base_layer = data.work_layer != 'BASE'
# Get the work layer filter, based on layer settings
work_layer_mask = sprytile_utils.get_work_layer_data(data)
# Build mode with join multi
if do_join:
origin_coord = ((grid_coord[0] + coord_min[0]),
(grid_coord[1] + coord_min[1]))
size_x = (coord_max[0] - coord_min[0]) + 1
size_y = (coord_max[1] - coord_min[1]) + 1
tile_origin = (grid.tile_selection[0], grid.tile_selection[1])
tile_coord = (tile_origin[0] + grid.tile_selection[2],
tile_origin[1] + grid.tile_selection[3])
face_index = self.modal.construct_face(
context,
origin_coord, [size_x, size_y],
tile_coord,
tile_origin,
grid_up,
grid_right,
up_vector,
right_vector,
plane_normal,
require_base_layer=require_base_layer,
work_layer_mask=work_layer_mask)
if face_index is not None:
face_verts = self.modal.face_to_world_verts(
context, face_index)
faces_verts.extend(face_verts)
# Build mode without auto join, try operation on each build coordinate
else:
virtual_cursor = scene.cursor_location + \
(grid_coord[0] * grid_right) + \
(grid_coord[1] * grid_up)
self.modal.add_virtual_cursor(virtual_cursor)
# Loop through grid coordinates to build
for i in range(len(offset_grid)):
grid_offset = offset_grid[i]
tile_offset = offset_tile_id[i]
grid_pos = [
grid_coord[0] + grid_offset[0],
grid_coord[1] + grid_offset[1]
]
tile_pos = [
tile_xy[0] + tile_offset[0], tile_xy[1] + tile_offset[1]
]
face_index = self.modal.construct_face(
context,
grid_pos, [1, 1],
tile_pos,
tile_xy,
grid_up,
grid_right,
up_vector,
right_vector,
plane_normal,
require_base_layer=require_base_layer,
work_layer_mask=work_layer_mask)
if face_index is not None:
face_verts = self.modal.face_to_world_verts(
context, face_index)
faces_verts.extend(face_verts)
if plane_pos is not None:
self.modal.add_virtual_cursor(plane_pos)
if data.cursor_flow and data.work_layer == "BASE" and len(
faces_verts) > 0:
# Find which vertex the cursor should flow to
new_cursor_pos = self.modal.flow_cursor_verts(
context, faces_verts, plane_pos)
if new_cursor_pos is not None:
# Not base layer, move position back by offset
if data.work_layer != 'BASE':
new_cursor_pos -= plane_normal * data.mesh_decal_offset
# Calculate the world position of old start_coord
old_start_pos = scene.cursor_location + (
self.start_coord[0] * grid_right) + (self.start_coord[1] *
grid_up)
# find the offset of the old start position from the new cursor position
new_start_offset = old_start_pos - new_cursor_pos
# get how much the grid x/y vectors need to scale by to normalize
scale_right = 1.0 / grid_right.magnitude
scale_up = 1.0 / grid_up.magnitude
# scale the offset by grid x/y, so can use normalized dot product to
# find the grid coordinates the start position is from new cursor pos
new_start_coord = Vector(
((new_start_offset * scale_right).dot(
grid_right.normalized()),
(new_start_offset * scale_up).dot(grid_up.normalized())))
# Record the new offset starting coord,
# for the nice painting snap
self.start_coord = new_start_coord
scene.cursor_location = new_cursor_pos
def execute(self, context, scene, ray_origin, ray_vector, is_start):
data = scene.sprytile_data
grid = sprytile_utils.get_grid(context, context.object.sprytile_gridid)
tile_xy = (grid.tile_selection[0], grid.tile_selection[1])
# Get vectors for grid, without rotation
up_vector, right_vector, plane_normal = sprytile_utils.get_current_grid_vectors(
scene,
with_rotation=False
)
# Rotate the vectors
rotation = Quaternion(plane_normal, data.mesh_rotate)
up_vector = rotation * up_vector
right_vector = rotation * right_vector
# Used to move raycast slightly along ray vector
shift_vec = ray_vector.normalized() * 0.001
# raycast grid to get the grid position under the mouse
grid_coord, grid_right, grid_up, plane_pos = sprytile_utils.raycast_grid(
scene, context,
up_vector, right_vector, plane_normal,
ray_origin, ray_vector,
as_coord=True
)
# Record starting position of stroke
if is_start:
self.start_coord = grid_coord
# Not starting, filter out when can build
elif self.start_coord is not None:
tolerance_min = (floor(grid.tile_selection[2] * 0.25),
floor(grid.tile_selection[3] * 0.25))
coord_offset = (
(grid_coord[0] - self.start_coord[0]) % grid.tile_selection[2],
(grid_coord[1] - self.start_coord[1]) % grid.tile_selection[3]
)
if coord_offset[0] > 0 or coord_offset[1] > 0:
return
# Get the area to build
offset_tile_id, offset_grid, coord_min, coord_max = sprytile_utils.get_grid_area(
grid.tile_selection[2],
grid.tile_selection[3],
data.uv_flip_x, data.uv_flip_y
)
# Check if joining multi tile faces
grid_no_spacing = sprytile_utils.grid_no_spacing(grid)
is_single_pixel = sprytile_utils.grid_is_single_pixel(grid)
do_join = is_single_pixel
if do_join is False:
do_join = grid_no_spacing and data.auto_join
# Raycast under mouse
hit_loc, hit_normal, hit_face_idx, hit_dist = self.modal.raycast_object(
context.object, ray_origin, ray_vector)
# Hit a face
if hit_face_idx is not None:
# Determine if face can be painted
plane_hit = intersect_line_plane(ray_origin, ray_origin + ray_vector,
scene.cursor_location, plane_normal)
plane_dist = (plane_hit - ray_origin).magnitude
difference = abs(hit_dist - plane_dist)
# If valid for painting instead of creating…
if difference < 0.01 or hit_dist < plane_dist:
if do_join is False:
return
check_dot = abs(plane_normal.dot(hit_normal))
check_dot -= 1
check_coplanar = distance_point_to_plane(hit_loc, scene.cursor_location, plane_normal)
check_coplanar = abs(check_coplanar) < 0.05
check_dot = abs(check_dot) < 0.05
# Paint if coplanar and dot angle checks out
if check_coplanar and check_dot:
face, verts, uvs, target_grid, data, target_img, tile_xy = tool_paint.ToolPaint.process_preview(
self.modal, context,
scene, hit_face_idx)
sprytile_uv.apply_uvs(context, face, uvs, target_grid,
self.modal.bmesh, data, target_img,
tile_xy, origin_xy=tile_xy)
# Hit a face, that could have been painted, don't do anything else
return
# Build mode with auto join
if do_join:
origin_coord = scene.cursor_location + \
(grid_coord[0] + coord_min[0]) * grid_right + \
(grid_coord[1] + coord_min[1]) * grid_up
self.modal.add_virtual_cursor(origin_coord)
size_x = (coord_max[0] - coord_min[0]) + 1
size_y = (coord_max[1] - coord_min[1]) + 1
size_x *= grid.grid[0]
size_y *= grid.grid[1]
grid_right *= size_x / grid.grid[0]
grid_up *= size_y / grid.grid[1]
verts = self.modal.get_build_vertices(origin_coord,
grid_right, grid_up,
up_vector, right_vector)
face_index = self.modal.create_face(context, verts)
if face_index is None:
return
vtx_center = Vector((0, 0, 0))
for vtx in verts:
vtx_center += vtx
vtx_center /= len(verts)
origin_xy = (grid.tile_selection[0],
grid.tile_selection[1])
target_img = sprytile_utils.get_grid_texture(context.object, grid)
uvs = sprytile_uv.get_uv_pos_size(data, target_img.size, grid,
origin_xy, size_x, size_y,
up_vector, right_vector,
verts, vtx_center)
sprytile_uv.apply_uvs(context, self.modal.bmesh.faces[face_index],
uvs, grid, self.modal.bmesh,
data, target_img, origin_xy)
if data.auto_merge:
threshold = (1 / data.world_pixels) * min(2, grid.grid[0], grid.grid[1])
face = self.modal.bmesh.faces[face_index]
self.modal.merge_doubles(context, face, vtx_center, -plane_normal, threshold)
# Build mode without auto join
else:
virtual_cursor = scene.cursor_location + \
(grid_coord[0] * grid_right) + \
(grid_coord[1] * grid_up)
self.modal.add_virtual_cursor(virtual_cursor)
# Loop through grid coordinates to build
for i in range(len(offset_grid)):
grid_offset = offset_grid[i]
tile_offset = offset_tile_id[i]
grid_pos = [grid_coord[0] + grid_offset[0], grid_coord[1] + grid_offset[1]]
tile_pos = [tile_xy[0] + tile_offset[0], tile_xy[1] + tile_offset[1]]
self.modal.construct_face(context, grid_pos,
tile_pos, tile_xy,
grid_up, grid_right,
up_vector, right_vector, plane_normal,
shift_vec=shift_vec)
def build_preview(self, context, scene, ray_origin, ray_vector):
obj = context.object
data = scene.sprytile_data
grid_id = obj.sprytile_gridid
target_grid = sprytile_utils.get_grid(context, grid_id)
# Reset can build flag
self.can_build = False
target_img = sprytile_utils.get_grid_texture(obj, target_grid)
if target_img is None:
self.modal.clear_preview_data()
return
# If building on base layer, get from current virtual grid
up_vector, right_vector, plane_normal = sprytile_utils.get_current_grid_vectors(
scene, False)
# Building on decal layer, get from face under mouse
if data.work_layer == 'DECAL_1' and data.lock_normal is False:
location, hit_normal, face_index, distance = self.modal.raycast_object(
context.object, ray_origin, ray_vector)
# For decals, if not hitting the object don't draw preview
if hit_normal is None:
self.modal.clear_preview_data()
return
# Do a coplanar check between hit location and cursor
grid_origin = scene.cursor_location.copy()
grid_origin += hit_normal * data.mesh_decal_offset
check_coplanar = distance_point_to_plane(location, grid_origin,
hit_normal)
check_coplanar = abs(check_coplanar) < 0.05
if check_coplanar is False:
self.modal.clear_preview_data()
return
face_up, face_right = self.modal.get_face_up_vector(
context, face_index, 0.4, bias_right=True)
if face_up is not None and face_right is not None:
plane_normal = hit_normal
up_vector = face_up
right_vector = face_right
else:
self.modal.clear_preview_data()
return
rotation = Quaternion(plane_normal, data.mesh_rotate)
up_vector = rotation * up_vector
right_vector = rotation * right_vector
# Raycast to the virtual grid
face_position, x_vector, y_vector, plane_cursor = sprytile_utils.raycast_grid(
scene, context, up_vector, right_vector, plane_normal, ray_origin,
ray_vector)
if face_position is None:
self.modal.clear_preview_data()
return
# Passed can build checks, set flag to true
self.can_build = True
offset_tile_id, offset_grid, coord_min, coord_max = sprytile_utils.get_grid_area(
target_grid.tile_selection[2], target_grid.tile_selection[3],
data.uv_flip_x, data.uv_flip_y)
grid_no_spacing = sprytile_utils.grid_no_spacing(target_grid)
# No spacing in grid, automatically join the preview together
if grid_no_spacing:
origin_coord = face_position + coord_min[0] * x_vector + coord_min[
1] * y_vector
size_x = (coord_max[0] - coord_min[0]) + 1
size_y = (coord_max[1] - coord_min[1]) + 1
size_x *= target_grid.grid[0]
size_y *= target_grid.grid[1]
x_vector *= size_x / target_grid.grid[0]
y_vector *= size_y / target_grid.grid[1]
preview_verts = self.modal.get_build_vertices(
origin_coord, x_vector, y_vector, up_vector, right_vector)
vtx_center = Vector((0, 0, 0))
for vtx in preview_verts:
vtx_center += vtx
vtx_center /= len(preview_verts)
origin_xy = (target_grid.tile_selection[0],
target_grid.tile_selection[1])
preview_uvs = sprytile_uv.get_uv_pos_size(
data, target_img.size, target_grid, origin_xy, size_x, size_y,
up_vector, right_vector, preview_verts, vtx_center)
self.modal.set_preview_data(preview_verts, preview_uvs)
return
# Spaced grids need to be tiled
preview_verts = []
preview_uvs = []
for i in range(len(offset_tile_id)):
grid_offset = offset_grid[i]
tile_offset = offset_tile_id[i]
x_offset = x_vector * grid_offset[0]
y_offset = y_vector * grid_offset[1]
coord_position = face_position + x_offset + y_offset
coord_verts = self.modal.get_build_vertices(
coord_position, x_vector, y_vector, up_vector, right_vector)
# Get the center of the preview verts
vtx_center = Vector((0, 0, 0))
for vtx in coord_verts:
vtx_center += vtx
vtx_center /= len(coord_verts)
# Calculate the tile with offset
tile_xy = (target_grid.tile_selection[0] + tile_offset[0],
target_grid.tile_selection[1] + tile_offset[1])
coord_uvs = sprytile_uv.get_uv_positions(data, target_img.size,
target_grid, up_vector,
right_vector, tile_xy,
coord_verts, vtx_center)
preview_verts.extend(coord_verts)
preview_uvs.extend(coord_uvs)
self.modal.set_preview_data(preview_verts, preview_uvs)
def build_preview(self, context, scene, ray_origin, ray_vector):
obj = context.object
data = scene.sprytile_data
grid_id = obj.sprytile_gridid
target_grid = sprytile_utils.get_grid(context, grid_id)
target_img = sprytile_utils.get_grid_texture(obj, target_grid)
if target_img is None:
return
up_vector, right_vector, plane_normal = sprytile_utils.get_current_grid_vectors(scene, False)
rotation = Quaternion(plane_normal, data.mesh_rotate)
up_vector = rotation * up_vector
right_vector = rotation * right_vector
# Raycast to the virtual grid
face_position, x_vector, y_vector, plane_cursor = sprytile_utils.raycast_grid(
scene, context,
up_vector, right_vector, plane_normal,
ray_origin, ray_vector
)
if face_position is None:
return
offset_tile_id, offset_grid, coord_min, coord_max = sprytile_utils.get_grid_area(
target_grid.tile_selection[2],
target_grid.tile_selection[3],
data.uv_flip_x,
data.uv_flip_y)
grid_no_spacing = sprytile_utils.grid_no_spacing(target_grid)
# No spacing in grid, automatically join the preview together
if grid_no_spacing:
origin_coord = face_position + coord_min[0] * x_vector + coord_min[1] * y_vector
size_x = (coord_max[0] - coord_min[0]) + 1
size_y = (coord_max[1] - coord_min[1]) + 1
size_x *= target_grid.grid[0]
size_y *= target_grid.grid[1]
x_vector *= size_x / target_grid.grid[0]
y_vector *= size_y / target_grid.grid[1]
preview_verts = self.modal.get_build_vertices(origin_coord,
x_vector, y_vector,
up_vector, right_vector)
vtx_center = Vector((0, 0, 0))
for vtx in preview_verts:
vtx_center += vtx
vtx_center /= len(preview_verts)
origin_xy = (target_grid.tile_selection[0],
target_grid.tile_selection[1])
preview_uvs = sprytile_uv.get_uv_pos_size(data, target_img.size, target_grid,
origin_xy, size_x, size_y,
up_vector, right_vector,
preview_verts, vtx_center)
self.modal.set_preview_data(preview_verts, preview_uvs)
return
# Spaced grids need to be tiled
preview_verts = []
preview_uvs = []
for i in range(len(offset_tile_id)):
grid_offset = offset_grid[i]
tile_offset = offset_tile_id[i]
x_offset = x_vector * grid_offset[0]
y_offset = y_vector * grid_offset[1]
coord_position = face_position + x_offset + y_offset
coord_verts = self.modal.get_build_vertices(coord_position, x_vector, y_vector,
up_vector, right_vector)
# Get the center of the preview verts
vtx_center = Vector((0, 0, 0))
for vtx in coord_verts:
vtx_center += vtx
vtx_center /= len(coord_verts)
# Calculate the tile with offset
tile_xy = (target_grid.tile_selection[0] + tile_offset[0],
target_grid.tile_selection[1] + tile_offset[1])
coord_uvs = sprytile_uv.get_uv_positions(data, target_img.size, target_grid,
up_vector, right_vector, tile_xy,
coord_verts, vtx_center)
preview_verts.extend(coord_verts)
preview_uvs.extend(coord_uvs)
self.modal.set_preview_data(preview_verts, preview_uvs)