mirror of
https://github.com/Orama-Interactive/Pixelorama.git
synced 2025-02-13 09:13:07 +00:00
2d28136449
* Create a custom PixeloramaImage class, initial support for indexed mode * Convert opened projects and images to indexed mode * Use shaders for RGB to Indexed conversion and vice versa * Add `is_indexed` variable in PixeloramaImage * Basic undo/redo support for indexed mode when drawing * Make image effects respect indexed mode * Move code from image effects to ShaderImageEffect instead * Bucket tool works with indexed mode * Move and selection tools works with indexed mode * Brushes respect indexed mode * Add color_mode variable and some helper methods in Project Replace hard-coded cases of Image.FORMAT_RGBA8 with `Project.get_image_format()` just in case we want to add more formats in the future * Add a helper new_empty_image() method to Project * Set new images to indexed if the project is indexed * Change color modes from the Image menu * Fix open image to replace cel * Load/save indices in pxo files * Merging layers works with indexed mode * Layer effects respect indexed mode * Add an `other_image` parameter to `PixeloramaImage.add_data_to_dictionary()` * Scale image works with indexed mode * Resizing works with indexed mode * Fix non-shader rotation not working with indexed mode * Minor refactor of PixeloramaImage's set_pixelv_custom() * Make the text tool work with indexed mode * Remove print from PixeloramaImage * Rename "PixeloramaImage" to "ImageExtended" * Add docstrings in ImageExtended * Set color mode from the create new image dialog * Update Translations.pot * Show the color mode in the project properties dialog
714 lines
22 KiB
GDScript
714 lines
22 KiB
GDScript
extends Node
|
|
|
|
enum GradientDirection { TOP, BOTTOM, LEFT, RIGHT }
|
|
## Continuation from Image.Interpolation
|
|
enum Interpolation { SCALE3X = 5, CLEANEDGE = 6, OMNISCALE = 7 }
|
|
var blend_layers_shader := preload("res://src/Shaders/BlendLayers.gdshader")
|
|
var clean_edge_shader: Shader:
|
|
get:
|
|
if clean_edge_shader == null:
|
|
clean_edge_shader = load("res://src/Shaders/Effects/Rotation/cleanEdge.gdshader")
|
|
return clean_edge_shader
|
|
var omniscale_shader: Shader:
|
|
get:
|
|
if omniscale_shader == null:
|
|
omniscale_shader = load("res://src/Shaders/Effects/Rotation/OmniScale.gdshader")
|
|
return omniscale_shader
|
|
|
|
|
|
## Blends canvas layers into passed image starting from the origin position
|
|
func blend_layers(
|
|
image: Image,
|
|
frame: Frame,
|
|
origin := Vector2i.ZERO,
|
|
project := Global.current_project,
|
|
only_selected_cels := false,
|
|
only_selected_layers := false,
|
|
) -> void:
|
|
var frame_index := project.frames.find(frame)
|
|
var previous_ordered_layers: Array[int] = project.ordered_layers
|
|
project.order_layers(frame_index)
|
|
var textures: Array[Image] = []
|
|
# Nx4 texture, where N is the number of layers and the first row are the blend modes,
|
|
# the second are the opacities, the third are the origins and the fourth are the
|
|
# clipping mask booleans.
|
|
var metadata_image := Image.create(project.layers.size(), 4, false, Image.FORMAT_R8)
|
|
for i in project.layers.size():
|
|
var ordered_index := project.ordered_layers[i]
|
|
var layer := project.layers[ordered_index]
|
|
var include := true if layer.is_visible_in_hierarchy() else false
|
|
if only_selected_cels and include:
|
|
var test_array := [frame_index, i]
|
|
if not test_array in project.selected_cels:
|
|
include = false
|
|
if only_selected_layers and include:
|
|
var layer_is_selected := false
|
|
for selected_cel in project.selected_cels:
|
|
if i == selected_cel[1]:
|
|
layer_is_selected = true
|
|
break
|
|
if not layer_is_selected:
|
|
include = false
|
|
var cel := frame.cels[ordered_index]
|
|
if DisplayServer.get_name() == "headless":
|
|
blend_layers_headless(image, project, layer, cel, origin)
|
|
else:
|
|
if layer is GroupLayer and layer.blend_mode != BaseLayer.BlendModes.PASS_THROUGH:
|
|
var cel_image := (layer as GroupLayer).blend_children(frame)
|
|
textures.append(cel_image)
|
|
else:
|
|
var cel_image := layer.display_effects(cel)
|
|
textures.append(cel_image)
|
|
if (
|
|
layer.is_blended_by_ancestor()
|
|
and not only_selected_cels
|
|
and not only_selected_layers
|
|
):
|
|
include = false
|
|
set_layer_metadata_image(layer, cel, metadata_image, ordered_index, include)
|
|
if DisplayServer.get_name() != "headless":
|
|
var texture_array := Texture2DArray.new()
|
|
texture_array.create_from_images(textures)
|
|
var params := {
|
|
"layers": texture_array,
|
|
"metadata": ImageTexture.create_from_image(metadata_image),
|
|
}
|
|
var blended := Image.create(project.size.x, project.size.y, false, image.get_format())
|
|
var gen := ShaderImageEffect.new()
|
|
gen.generate_image(blended, blend_layers_shader, params, project.size)
|
|
image.blend_rect(blended, Rect2i(Vector2i.ZERO, project.size), origin)
|
|
# Re-order the layers again to ensure correct canvas drawing
|
|
project.ordered_layers = previous_ordered_layers
|
|
|
|
|
|
func set_layer_metadata_image(
|
|
layer: BaseLayer, cel: BaseCel, image: Image, index: int, include := true
|
|
) -> void:
|
|
# Store the blend mode
|
|
image.set_pixel(index, 0, Color(layer.blend_mode / 255.0, 0.0, 0.0, 0.0))
|
|
# Store the opacity
|
|
if layer.is_visible_in_hierarchy() and include:
|
|
var opacity := cel.get_final_opacity(layer)
|
|
image.set_pixel(index, 1, Color(opacity, 0.0, 0.0, 0.0))
|
|
else:
|
|
image.set_pixel(index, 1, Color())
|
|
# Store the clipping mask boolean
|
|
if layer.clipping_mask:
|
|
image.set_pixel(index, 3, Color.RED)
|
|
else:
|
|
image.set_pixel(index, 3, Color.BLACK)
|
|
if not include:
|
|
# Store a small red value as a way to indicate that this layer should be skipped
|
|
# Used for layers such as child layers of a group, so that the group layer itself can
|
|
# successfully be used as a clipping mask with the layer below it.
|
|
image.set_pixel(index, 3, Color(0.2, 0.0, 0.0, 0.0))
|
|
|
|
|
|
func blend_layers_headless(
|
|
image: Image, project: Project, layer: BaseLayer, cel: BaseCel, origin: Vector2i
|
|
) -> void:
|
|
var opacity := cel.get_final_opacity(layer)
|
|
var cel_image := Image.new()
|
|
cel_image.copy_from(cel.get_image())
|
|
if opacity < 1.0: # If we have cel or layer transparency
|
|
for xx in cel_image.get_size().x:
|
|
for yy in cel_image.get_size().y:
|
|
var pixel_color := cel_image.get_pixel(xx, yy)
|
|
pixel_color.a *= opacity
|
|
cel_image.set_pixel(xx, yy, pixel_color)
|
|
image.blend_rect(cel_image, Rect2i(Vector2i.ZERO, project.size), origin)
|
|
|
|
|
|
## Algorithm based on http://members.chello.at/easyfilter/bresenham.html
|
|
func get_ellipse_points(pos: Vector2i, size: Vector2i) -> Array[Vector2i]:
|
|
var array: Array[Vector2i] = []
|
|
var x0 := pos.x
|
|
var x1 := pos.x + (size.x - 1)
|
|
var y0 := pos.y
|
|
var y1 := pos.y + (size.y - 1)
|
|
var a := absi(x1 - x0)
|
|
var b := absi(y1 - x0)
|
|
var b1 := b & 1
|
|
var dx := 4 * (1 - a) * b * b
|
|
var dy := 4 * (b1 + 1) * a * a
|
|
var err := dx + dy + b1 * a * a
|
|
var e2 := 0
|
|
|
|
if x0 > x1:
|
|
x0 = x1
|
|
x1 += a
|
|
|
|
if y0 > y1:
|
|
y0 = y1
|
|
|
|
y0 += (b + 1) / 2
|
|
y1 = y0 - b1
|
|
a *= 8 * a
|
|
b1 = 8 * b * b
|
|
|
|
while x0 <= x1:
|
|
var v1 := Vector2i(x1, y0)
|
|
var v2 := Vector2i(x0, y0)
|
|
var v3 := Vector2i(x0, y1)
|
|
var v4 := Vector2i(x1, y1)
|
|
array.append(v1)
|
|
array.append(v2)
|
|
array.append(v3)
|
|
array.append(v4)
|
|
|
|
e2 = 2 * err
|
|
if e2 <= dy:
|
|
y0 += 1
|
|
y1 -= 1
|
|
dy += a
|
|
err += dy
|
|
|
|
if e2 >= dx || 2 * err > dy:
|
|
x0 += 1
|
|
x1 -= 1
|
|
dx += b1
|
|
err += dx
|
|
|
|
while y0 - y1 < b:
|
|
var v1 := Vector2i(x0 - 1, y0)
|
|
var v2 := Vector2i(x1 + 1, y0)
|
|
var v3 := Vector2i(x0 - 1, y1)
|
|
var v4 := Vector2i(x1 + 1, y1)
|
|
array.append(v1)
|
|
array.append(v2)
|
|
array.append(v3)
|
|
array.append(v4)
|
|
y0 += 1
|
|
y1 -= 1
|
|
|
|
return array
|
|
|
|
|
|
func get_ellipse_points_filled(pos: Vector2i, size: Vector2i, thickness := 1) -> Array[Vector2i]:
|
|
var offsetted_size := size + Vector2i.ONE * (thickness - 1)
|
|
var border := get_ellipse_points(pos, offsetted_size)
|
|
var filling: Array[Vector2i] = []
|
|
|
|
for x in range(1, ceili(offsetted_size.x / 2.0)):
|
|
var fill := false
|
|
var prev_is_true := false
|
|
for y in range(0, ceili(offsetted_size.y / 2.0)):
|
|
var top_l_p := Vector2i(x, y)
|
|
var bit := border.has(pos + top_l_p)
|
|
|
|
if bit and not fill:
|
|
prev_is_true = true
|
|
continue
|
|
|
|
if not bit and (fill or prev_is_true):
|
|
filling.append(pos + top_l_p)
|
|
filling.append(pos + Vector2i(x, offsetted_size.y - y - 1))
|
|
filling.append(pos + Vector2i(offsetted_size.x - x - 1, y))
|
|
filling.append(pos + Vector2i(offsetted_size.x - x - 1, offsetted_size.y - y - 1))
|
|
|
|
if prev_is_true:
|
|
fill = true
|
|
prev_is_true = false
|
|
elif bit and fill:
|
|
break
|
|
|
|
return border + filling
|
|
|
|
|
|
func scale_3x(sprite: Image, tol := 0.196078) -> Image:
|
|
var scaled := Image.create(
|
|
sprite.get_width() * 3, sprite.get_height() * 3, sprite.has_mipmaps(), sprite.get_format()
|
|
)
|
|
var width_minus_one := sprite.get_width() - 1
|
|
var height_minus_one := sprite.get_height() - 1
|
|
for x in range(0, sprite.get_width()):
|
|
for y in range(0, sprite.get_height()):
|
|
var xs := 3 * x
|
|
var ys := 3 * y
|
|
|
|
var a := sprite.get_pixel(maxi(x - 1, 0), maxi(y - 1, 0))
|
|
var b := sprite.get_pixel(mini(x, width_minus_one), maxi(y - 1, 0))
|
|
var c := sprite.get_pixel(mini(x + 1, width_minus_one), maxi(y - 1, 0))
|
|
var d := sprite.get_pixel(maxi(x - 1, 0), mini(y, height_minus_one))
|
|
var e := sprite.get_pixel(mini(x, width_minus_one), mini(y, height_minus_one))
|
|
var f := sprite.get_pixel(mini(x + 1, width_minus_one), mini(y, height_minus_one))
|
|
var g := sprite.get_pixel(maxi(x - 1, 0), mini(y + 1, height_minus_one))
|
|
var h := sprite.get_pixel(mini(x, width_minus_one), mini(y + 1, height_minus_one))
|
|
var i := sprite.get_pixel(mini(x + 1, width_minus_one), mini(y + 1, height_minus_one))
|
|
|
|
var db: bool = similar_colors(d, b, tol)
|
|
var dh: bool = similar_colors(d, h, tol)
|
|
var bf: bool = similar_colors(f, b, tol)
|
|
var ec: bool = similar_colors(e, c, tol)
|
|
var ea: bool = similar_colors(e, a, tol)
|
|
var fh: bool = similar_colors(f, h, tol)
|
|
var eg: bool = similar_colors(e, g, tol)
|
|
var ei: bool = similar_colors(e, i, tol)
|
|
|
|
scaled.set_pixel(maxi(xs - 1, 0), maxi(ys - 1, 0), d if (db and !dh and !bf) else e)
|
|
scaled.set_pixel(
|
|
xs,
|
|
maxi(ys - 1, 0),
|
|
b if (db and !dh and !bf and !ec) or (bf and !db and !fh and !ea) else e
|
|
)
|
|
scaled.set_pixel(xs + 1, maxi(ys - 1, 0), f if (bf and !db and !fh) else e)
|
|
scaled.set_pixel(
|
|
maxi(xs - 1, 0),
|
|
ys,
|
|
d if (dh and !fh and !db and !ea) or (db and !dh and !bf and !eg) else e
|
|
)
|
|
scaled.set_pixel(xs, ys, e)
|
|
scaled.set_pixel(
|
|
xs + 1, ys, f if (bf and !db and !fh and !ei) or (fh and !bf and !dh and !ec) else e
|
|
)
|
|
scaled.set_pixel(maxi(xs - 1, 0), ys + 1, d if (dh and !fh and !db) else e)
|
|
scaled.set_pixel(
|
|
xs, ys + 1, h if (fh and !bf and !dh and !eg) or (dh and !fh and !db and !ei) else e
|
|
)
|
|
scaled.set_pixel(xs + 1, ys + 1, f if (fh and !bf and !dh) else e)
|
|
|
|
return scaled
|
|
|
|
|
|
func rotxel(sprite: Image, angle: float, pivot: Vector2) -> void:
|
|
if is_zero_approx(angle) or is_equal_approx(angle, TAU):
|
|
return
|
|
if is_equal_approx(angle, PI / 2.0) or is_equal_approx(angle, 3.0 * PI / 2.0):
|
|
nn_rotate(sprite, angle, pivot)
|
|
return
|
|
if is_equal_approx(angle, PI):
|
|
sprite.rotate_180()
|
|
return
|
|
|
|
var aux := Image.new()
|
|
aux.copy_from(sprite)
|
|
var ox: int
|
|
var oy: int
|
|
for x in sprite.get_size().x:
|
|
for y in sprite.get_size().y:
|
|
var dx := 3 * (x - pivot.x)
|
|
var dy := 3 * (y - pivot.y)
|
|
var found_pixel := false
|
|
for k in range(9):
|
|
var modk := -1 + k % 3
|
|
var divk := -1 + int(k / 3)
|
|
var dir := atan2(dy + divk, dx + modk)
|
|
var mag := sqrt(pow(dx + modk, 2) + pow(dy + divk, 2))
|
|
dir -= angle
|
|
ox = roundi(pivot.x * 3 + 1 + mag * cos(dir))
|
|
oy = roundi(pivot.y * 3 + 1 + mag * sin(dir))
|
|
|
|
if sprite.get_width() % 2 != 0:
|
|
ox += 1
|
|
oy += 1
|
|
|
|
if (
|
|
ox >= 0
|
|
&& ox < sprite.get_width() * 3
|
|
&& oy >= 0
|
|
&& oy < sprite.get_height() * 3
|
|
):
|
|
found_pixel = true
|
|
break
|
|
|
|
if !found_pixel:
|
|
sprite.set_pixel(x, y, Color(0, 0, 0, 0))
|
|
continue
|
|
|
|
var fil: int = oy % 3
|
|
var col: int = ox % 3
|
|
var index: int = col + 3 * fil
|
|
|
|
ox = roundi((ox - 1) / 3.0)
|
|
oy = roundi((oy - 1) / 3.0)
|
|
var p: Color
|
|
if ox == 0 || ox == sprite.get_width() - 1 || oy == 0 || oy == sprite.get_height() - 1:
|
|
p = aux.get_pixel(ox, oy)
|
|
else:
|
|
var a := aux.get_pixel(ox - 1, oy - 1)
|
|
var b := aux.get_pixel(ox, oy - 1)
|
|
var c := aux.get_pixel(ox + 1, oy - 1)
|
|
var d := aux.get_pixel(ox - 1, oy)
|
|
var e := aux.get_pixel(ox, oy)
|
|
var f := aux.get_pixel(ox + 1, oy)
|
|
var g := aux.get_pixel(ox - 1, oy + 1)
|
|
var h := aux.get_pixel(ox, oy + 1)
|
|
var i := aux.get_pixel(ox + 1, oy + 1)
|
|
|
|
match index:
|
|
0:
|
|
p = (
|
|
d
|
|
if (
|
|
similar_colors(d, b)
|
|
&& !similar_colors(d, h)
|
|
&& !similar_colors(b, f)
|
|
)
|
|
else e
|
|
)
|
|
1:
|
|
p = (
|
|
b
|
|
if (
|
|
(
|
|
similar_colors(d, b)
|
|
&& !similar_colors(d, h)
|
|
&& !similar_colors(b, f)
|
|
&& !similar_colors(e, c)
|
|
)
|
|
|| (
|
|
similar_colors(b, f)
|
|
&& !similar_colors(d, b)
|
|
&& !similar_colors(f, h)
|
|
&& !similar_colors(e, a)
|
|
)
|
|
)
|
|
else e
|
|
)
|
|
2:
|
|
p = (
|
|
f
|
|
if (
|
|
similar_colors(b, f)
|
|
&& !similar_colors(d, b)
|
|
&& !similar_colors(f, h)
|
|
)
|
|
else e
|
|
)
|
|
3:
|
|
p = (
|
|
d
|
|
if (
|
|
(
|
|
similar_colors(d, h)
|
|
&& !similar_colors(f, h)
|
|
&& !similar_colors(d, b)
|
|
&& !similar_colors(e, a)
|
|
)
|
|
|| (
|
|
similar_colors(d, b)
|
|
&& !similar_colors(d, h)
|
|
&& !similar_colors(b, f)
|
|
&& !similar_colors(e, g)
|
|
)
|
|
)
|
|
else e
|
|
)
|
|
4:
|
|
p = e
|
|
5:
|
|
p = (
|
|
f
|
|
if (
|
|
(
|
|
similar_colors(b, f)
|
|
&& !similar_colors(d, b)
|
|
&& !similar_colors(f, h)
|
|
&& !similar_colors(e, i)
|
|
)
|
|
|| (
|
|
similar_colors(f, h)
|
|
&& !similar_colors(b, f)
|
|
&& !similar_colors(d, h)
|
|
&& !similar_colors(e, c)
|
|
)
|
|
)
|
|
else e
|
|
)
|
|
6:
|
|
p = (
|
|
d
|
|
if (
|
|
similar_colors(d, h)
|
|
&& !similar_colors(f, h)
|
|
&& !similar_colors(d, b)
|
|
)
|
|
else e
|
|
)
|
|
7:
|
|
p = (
|
|
h
|
|
if (
|
|
(
|
|
similar_colors(f, h)
|
|
&& !similar_colors(f, b)
|
|
&& !similar_colors(d, h)
|
|
&& !similar_colors(e, g)
|
|
)
|
|
|| (
|
|
similar_colors(d, h)
|
|
&& !similar_colors(f, h)
|
|
&& !similar_colors(d, b)
|
|
&& !similar_colors(e, i)
|
|
)
|
|
)
|
|
else e
|
|
)
|
|
8:
|
|
p = (
|
|
f
|
|
if (
|
|
similar_colors(f, h)
|
|
&& !similar_colors(f, b)
|
|
&& !similar_colors(d, h)
|
|
)
|
|
else e
|
|
)
|
|
sprite.set_pixel(x, y, p)
|
|
|
|
|
|
func fake_rotsprite(sprite: Image, angle: float, pivot: Vector2) -> void:
|
|
if is_zero_approx(angle) or is_equal_approx(angle, TAU):
|
|
return
|
|
if is_equal_approx(angle, PI / 2.0) or is_equal_approx(angle, 3.0 * PI / 2.0):
|
|
nn_rotate(sprite, angle, pivot)
|
|
return
|
|
if is_equal_approx(angle, PI):
|
|
sprite.rotate_180()
|
|
return
|
|
var selected_sprite := scale_3x(sprite)
|
|
nn_rotate(selected_sprite, angle, pivot * 3)
|
|
selected_sprite.resize(
|
|
selected_sprite.get_width() / 3, selected_sprite.get_height() / 3, Image.INTERPOLATE_NEAREST
|
|
)
|
|
sprite.blit_rect(selected_sprite, Rect2(Vector2.ZERO, selected_sprite.get_size()), Vector2.ZERO)
|
|
|
|
|
|
func nn_rotate(sprite: Image, angle: float, pivot: Vector2) -> void:
|
|
if is_zero_approx(angle) or is_equal_approx(angle, TAU):
|
|
return
|
|
if is_equal_approx(angle, PI):
|
|
sprite.rotate_180()
|
|
return
|
|
var aux := Image.new()
|
|
aux.copy_from(sprite)
|
|
var angle_sin := sin(angle)
|
|
var angle_cos := cos(angle)
|
|
for x in range(sprite.get_width()):
|
|
for y in range(sprite.get_height()):
|
|
var ox := (x - pivot.x) * angle_cos + (y - pivot.y) * angle_sin + pivot.x
|
|
var oy := -(x - pivot.x) * angle_sin + (y - pivot.y) * angle_cos + pivot.y
|
|
if ox >= 0 && ox < sprite.get_width() && oy >= 0 && oy < sprite.get_height():
|
|
sprite.set_pixel(x, y, aux.get_pixel(ox, oy))
|
|
else:
|
|
sprite.set_pixel(x, y, Color(0, 0, 0, 0))
|
|
|
|
|
|
## Compares two colors, and returns [code]true[/code] if the difference of these colors is
|
|
## less or equal to the tolerance [param tol]. [param tol] is in the range of 0-1.
|
|
func similar_colors(c1: Color, c2: Color, tol := 0.392157) -> bool:
|
|
return (
|
|
absf(c1.r - c2.r) <= tol
|
|
&& absf(c1.g - c2.g) <= tol
|
|
&& absf(c1.b - c2.b) <= tol
|
|
&& absf(c1.a - c2.a) <= tol
|
|
)
|
|
|
|
|
|
# Image effects
|
|
func center(indices: Array) -> void:
|
|
var project := Global.current_project
|
|
Global.canvas.selection.transform_content_confirm()
|
|
var redo_data := {}
|
|
var undo_data := {}
|
|
project.undos += 1
|
|
project.undo_redo.create_action("Center Frames")
|
|
for frame in indices:
|
|
# Find used rect of the current frame (across all of the layers)
|
|
var used_rect := Rect2i()
|
|
for cel in project.frames[frame].cels:
|
|
if not cel is PixelCel:
|
|
continue
|
|
var cel_rect := cel.get_image().get_used_rect()
|
|
if cel_rect.has_area():
|
|
used_rect = used_rect.merge(cel_rect) if used_rect.has_area() else cel_rect
|
|
if not used_rect.has_area():
|
|
continue
|
|
|
|
# Now apply centering
|
|
var offset: Vector2i = (0.5 * (project.size - used_rect.size)).floor()
|
|
for cel in project.frames[frame].cels:
|
|
if not cel is PixelCel:
|
|
continue
|
|
var cel_image := (cel as PixelCel).get_image()
|
|
var tmp_centered := project.new_empty_image()
|
|
tmp_centered.blend_rect(cel.image, used_rect, offset)
|
|
var centered := ImageExtended.new()
|
|
centered.copy_from_custom(tmp_centered, cel_image.is_indexed)
|
|
centered.add_data_to_dictionary(redo_data, cel_image)
|
|
cel_image.add_data_to_dictionary(undo_data)
|
|
Global.undo_redo_compress_images(redo_data, undo_data)
|
|
project.undo_redo.add_undo_method(Global.undo_or_redo.bind(true))
|
|
project.undo_redo.add_do_method(Global.undo_or_redo.bind(false))
|
|
project.undo_redo.commit_action()
|
|
|
|
|
|
func scale_project(width: int, height: int, interpolation: int) -> void:
|
|
var redo_data := {}
|
|
var undo_data := {}
|
|
for f in Global.current_project.frames:
|
|
for i in range(f.cels.size() - 1, -1, -1):
|
|
var cel := f.cels[i]
|
|
if not cel is PixelCel:
|
|
continue
|
|
var cel_image := (cel as PixelCel).get_image()
|
|
var sprite := _resize_image(cel_image, width, height, interpolation) as ImageExtended
|
|
sprite.add_data_to_dictionary(redo_data, cel_image)
|
|
cel_image.add_data_to_dictionary(undo_data)
|
|
|
|
general_do_and_undo_scale(width, height, redo_data, undo_data)
|
|
|
|
|
|
func _resize_image(
|
|
image: Image, width: int, height: int, interpolation: Image.Interpolation
|
|
) -> Image:
|
|
var new_image: Image
|
|
if image is ImageExtended:
|
|
new_image = ImageExtended.new()
|
|
new_image.is_indexed = image.is_indexed
|
|
new_image.copy_from(image)
|
|
new_image.select_palette("", false)
|
|
else:
|
|
new_image = Image.new()
|
|
new_image.copy_from(image)
|
|
if interpolation == Interpolation.SCALE3X:
|
|
var times := Vector2i(
|
|
ceili(width / (3.0 * new_image.get_width())),
|
|
ceili(height / (3.0 * new_image.get_height()))
|
|
)
|
|
for _j in range(maxi(times.x, times.y)):
|
|
new_image.copy_from(scale_3x(new_image))
|
|
new_image.resize(width, height, Image.INTERPOLATE_NEAREST)
|
|
elif interpolation == Interpolation.CLEANEDGE:
|
|
var gen := ShaderImageEffect.new()
|
|
gen.generate_image(new_image, clean_edge_shader, {}, Vector2i(width, height), false)
|
|
elif interpolation == Interpolation.OMNISCALE and omniscale_shader:
|
|
var gen := ShaderImageEffect.new()
|
|
gen.generate_image(new_image, omniscale_shader, {}, Vector2i(width, height), false)
|
|
else:
|
|
new_image.resize(width, height, interpolation)
|
|
if new_image is ImageExtended:
|
|
new_image.on_size_changed()
|
|
return new_image
|
|
|
|
|
|
## Sets the size of the project to be the same as the size of the active selection.
|
|
func crop_to_selection() -> void:
|
|
if not Global.current_project.has_selection:
|
|
return
|
|
var redo_data := {}
|
|
var undo_data := {}
|
|
Global.canvas.selection.transform_content_confirm()
|
|
var rect: Rect2i = Global.canvas.selection.big_bounding_rectangle
|
|
# Loop through all the cels to crop them
|
|
for cel in Global.current_project.get_all_pixel_cels():
|
|
var cel_image := cel.get_image()
|
|
var tmp_cropped := cel_image.get_region(rect)
|
|
var cropped := ImageExtended.new()
|
|
cropped.copy_from_custom(tmp_cropped, cel_image.is_indexed)
|
|
cropped.add_data_to_dictionary(redo_data, cel_image)
|
|
cel_image.add_data_to_dictionary(undo_data)
|
|
|
|
general_do_and_undo_scale(rect.size.x, rect.size.y, redo_data, undo_data)
|
|
|
|
|
|
## Automatically makes the project smaller by looping through all of the cels and
|
|
## trimming out the pixels that are transparent in all cels.
|
|
func crop_to_content() -> void:
|
|
Global.canvas.selection.transform_content_confirm()
|
|
var used_rect := Rect2i()
|
|
for f in Global.current_project.frames:
|
|
for cel in f.cels:
|
|
if not cel is PixelCel:
|
|
continue
|
|
var cel_used_rect := cel.get_image().get_used_rect()
|
|
if cel_used_rect == Rect2i(0, 0, 0, 0): # If the cel has no content
|
|
continue
|
|
|
|
if used_rect == Rect2i(0, 0, 0, 0): # If we still haven't found the first cel with content
|
|
used_rect = cel_used_rect
|
|
else:
|
|
used_rect = used_rect.merge(cel_used_rect)
|
|
|
|
# If no layer has any content, just return
|
|
if used_rect == Rect2i(0, 0, 0, 0):
|
|
return
|
|
|
|
var width := used_rect.size.x
|
|
var height := used_rect.size.y
|
|
var redo_data := {}
|
|
var undo_data := {}
|
|
# Loop through all the cels to trim them
|
|
for cel in Global.current_project.get_all_pixel_cels():
|
|
var cel_image := cel.get_image()
|
|
var tmp_cropped := cel_image.get_region(used_rect)
|
|
var cropped := ImageExtended.new()
|
|
cropped.copy_from_custom(tmp_cropped, cel_image.is_indexed)
|
|
cropped.add_data_to_dictionary(redo_data, cel_image)
|
|
cel_image.add_data_to_dictionary(undo_data)
|
|
|
|
general_do_and_undo_scale(width, height, redo_data, undo_data)
|
|
|
|
|
|
func resize_canvas(width: int, height: int, offset_x: int, offset_y: int) -> void:
|
|
var redo_data := {}
|
|
var undo_data := {}
|
|
for cel in Global.current_project.get_all_pixel_cels():
|
|
var cel_image := cel.get_image()
|
|
var resized := ImageExtended.create_custom(
|
|
width, height, cel_image.has_mipmaps(), cel_image.get_format(), cel_image.is_indexed
|
|
)
|
|
resized.blend_rect(
|
|
cel_image, Rect2i(Vector2i.ZERO, cel_image.get_size()), Vector2i(offset_x, offset_y)
|
|
)
|
|
resized.convert_rgb_to_indexed()
|
|
resized.add_data_to_dictionary(redo_data, cel_image)
|
|
cel_image.add_data_to_dictionary(undo_data)
|
|
|
|
general_do_and_undo_scale(width, height, redo_data, undo_data)
|
|
|
|
|
|
func general_do_and_undo_scale(
|
|
width: int, height: int, redo_data: Dictionary, undo_data: Dictionary
|
|
) -> void:
|
|
var project := Global.current_project
|
|
var size := Vector2i(width, height)
|
|
var x_ratio := float(project.size.x) / width
|
|
var y_ratio := float(project.size.y) / height
|
|
|
|
var selection_map_copy := SelectionMap.new()
|
|
selection_map_copy.copy_from(project.selection_map)
|
|
selection_map_copy.crop(size.x, size.y)
|
|
redo_data[project.selection_map] = selection_map_copy.data
|
|
undo_data[project.selection_map] = project.selection_map.data
|
|
|
|
var new_x_symmetry_point := project.x_symmetry_point / x_ratio
|
|
var new_y_symmetry_point := project.y_symmetry_point / y_ratio
|
|
var new_x_symmetry_axis_points := project.x_symmetry_axis.points
|
|
var new_y_symmetry_axis_points := project.y_symmetry_axis.points
|
|
new_x_symmetry_axis_points[0].y /= y_ratio
|
|
new_x_symmetry_axis_points[1].y /= y_ratio
|
|
new_y_symmetry_axis_points[0].x /= x_ratio
|
|
new_y_symmetry_axis_points[1].x /= x_ratio
|
|
|
|
project.undos += 1
|
|
project.undo_redo.create_action("Scale")
|
|
project.undo_redo.add_do_property(project, "size", size)
|
|
project.undo_redo.add_do_property(project, "x_symmetry_point", new_x_symmetry_point)
|
|
project.undo_redo.add_do_property(project, "y_symmetry_point", new_y_symmetry_point)
|
|
project.undo_redo.add_do_property(project.x_symmetry_axis, "points", new_x_symmetry_axis_points)
|
|
project.undo_redo.add_do_property(project.y_symmetry_axis, "points", new_y_symmetry_axis_points)
|
|
Global.undo_redo_compress_images(redo_data, undo_data)
|
|
project.undo_redo.add_undo_property(project, "size", project.size)
|
|
project.undo_redo.add_undo_property(project, "x_symmetry_point", project.x_symmetry_point)
|
|
project.undo_redo.add_undo_property(project, "y_symmetry_point", project.y_symmetry_point)
|
|
project.undo_redo.add_undo_property(
|
|
project.x_symmetry_axis, "points", project.x_symmetry_axis.points
|
|
)
|
|
project.undo_redo.add_undo_property(
|
|
project.y_symmetry_axis, "points", project.y_symmetry_axis.points
|
|
)
|
|
project.undo_redo.add_undo_method(Global.undo_or_redo.bind(true))
|
|
project.undo_redo.add_do_method(Global.undo_or_redo.bind(false))
|
|
project.undo_redo.commit_action()
|