extends Node2D class_name Canvas var layers := [] var current_layer_index := 0 var trans_background : ImageTexture var location := Vector2.ZERO var size := Vector2(64, 64) var frame := 0 var frame_button : VBoxContainer var frame_texture_rect : TextureRect var previous_mouse_pos := Vector2.ZERO var mouse_inside_canvas := false #used for undo var sprite_changed_this_frame := false #for optimization purposes var is_making_line := false var is_making_selection := "None" var line_2d : Line2D # Called when the node enters the scene tree for the first time. func _ready() -> void: Global.can_draw = false #Background trans_background = ImageTexture.new() trans_background.create_from_image(load("res://Transparent Background.png"), 0) #The sprite itself if layers.empty(): var sprite := Image.new() sprite.create(size.x, size.y, false, Image.FORMAT_RGBA8) sprite.lock() var tex := ImageTexture.new() tex.create_from_image(sprite, 0) #Store [Image, ImageTexture, Layer Name, Visibity boolean] layers.append([sprite, tex, "Layer 0", true]) generate_layer_panels() frame_button = load("res://FrameButton.tscn").instance() frame_button.name = "Frame_%s" % frame frame_button.get_node("FrameButton").frame = frame frame_button.get_node("FrameID").text = str(frame + 1) Global.frame_container.add_child(frame_button) frame_texture_rect = Global.find_node_by_name(frame_button, "FrameTexture") frame_texture_rect.texture = layers[0][1] #ImageTexture current_layer_index camera_zoom() # warning-ignore:unused_argument func _process(delta) -> void: sprite_changed_this_frame = false update() var mouse_pos := get_local_mouse_position() - location var mouse_pos_floored := mouse_pos.floor() var mouse_pos_ceiled := mouse_pos.ceil() var current_mouse_button := "None" var current_action := "None" if Input.is_mouse_button_pressed(BUTTON_LEFT): current_mouse_button = "left_mouse" current_action = Global.current_left_tool elif Input.is_mouse_button_pressed(BUTTON_RIGHT): current_mouse_button = "right_mouse" current_action = Global.current_right_tool if visible: if !point_in_rectangle(mouse_pos, location, location + size): if !Input.is_mouse_button_pressed(BUTTON_LEFT) && !Input.is_mouse_button_pressed(BUTTON_RIGHT): if mouse_inside_canvas: mouse_inside_canvas = false Global.cursor_position_label.text = "[%sx%s]" % [size.x, size.y] else: Global.cursor_position_label.text = "[%sx%s] %s, %s" % [size.x, size.y, mouse_pos_floored.x, mouse_pos_floored.y] #Handle current tool match current_action: "Pencil": var current_color : Color if current_mouse_button == "left_mouse": current_color = Global.left_color_picker.color elif current_mouse_button == "right_mouse": current_color = Global.right_color_picker.color pencil_and_eraser(mouse_pos, current_color, current_mouse_button) "Eraser": pencil_and_eraser(mouse_pos, Color(0, 0, 0, 0), current_mouse_button) "Fill": if point_in_rectangle(mouse_pos, location, location + size) && Global.can_draw && Global.has_focus && Global.current_frame == frame: var current_color : Color if current_mouse_button == "left_mouse": current_color = Global.left_color_picker.color elif current_mouse_button == "right_mouse": current_color = Global.right_color_picker.color flood_fill(mouse_pos, layers[current_layer_index][0].get_pixelv(mouse_pos), current_color) "RectSelect": #Check SelectionRectangle.gd for more code on Rectangle Selection if Global.can_draw && Global.has_focus && Global.current_frame == frame: #If we're creating a new selection if Global.selected_pixels.size() == 0 || !point_in_rectangle(mouse_pos_floored, Global.selection_rectangle.polygon[0] - Vector2.ONE, Global.selection_rectangle.polygon[2]): if Input.is_action_just_pressed(current_mouse_button): Global.selection_rectangle.polygon[0] = mouse_pos_floored Global.selection_rectangle.polygon[1] = mouse_pos_floored Global.selection_rectangle.polygon[2] = mouse_pos_floored Global.selection_rectangle.polygon[3] = mouse_pos_floored is_making_selection = current_mouse_button Global.selected_pixels.clear() else: if is_making_selection != "None": #If we're making a new selection... var start_pos = Global.selection_rectangle.polygon[0] if start_pos != mouse_pos_floored: var end_pos := Vector2(mouse_pos_ceiled.x, mouse_pos_ceiled.y) if mouse_pos.x < start_pos.x: end_pos.x = mouse_pos_ceiled.x - 1 if mouse_pos.y < start_pos.y: end_pos.y = mouse_pos_ceiled.y - 1 Global.selection_rectangle.polygon[1] = Vector2(end_pos.x, start_pos.y) Global.selection_rectangle.polygon[2] = end_pos Global.selection_rectangle.polygon[3] = Vector2(start_pos.x, end_pos.y) if !is_making_line: previous_mouse_pos = mouse_pos previous_mouse_pos.x = clamp(previous_mouse_pos.x, location.x, location.x + size.x) previous_mouse_pos.y = clamp(previous_mouse_pos.y, location.y, location.y + size.y) else: line_2d.set_point_position(1, mouse_pos) if is_making_selection != "None": #If we're making a selection if Input.is_action_just_released(is_making_selection): #Finish selection when button is released var start_pos = Global.selection_rectangle.polygon[0] var end_pos = Global.selection_rectangle.polygon[2] if start_pos.x > end_pos.x: var temp = end_pos.x end_pos.x = start_pos.x start_pos.x = temp if start_pos.y > end_pos.y: var temp = end_pos.y end_pos.y = start_pos.y start_pos.y = temp Global.selection_rectangle.polygon[0] = start_pos Global.selection_rectangle.polygon[1] = Vector2(end_pos.x, start_pos.y) Global.selection_rectangle.polygon[2] = end_pos Global.selection_rectangle.polygon[3] = Vector2(start_pos.x, end_pos.y) for xx in range(start_pos.x, end_pos.x): for yy in range(start_pos.y, end_pos.y): if xx >= location.x && xx < size.x && yy >= location.y && yy < size.y: Global.selected_pixels.append(Vector2(xx, yy)) is_making_selection = "None" if sprite_changed_this_frame: update_texture(current_layer_index) func update_texture(layer_index : int) -> void: layers[layer_index][1].create_from_image(layers[layer_index][0], 0) get_layer_container(layer_index).get_child(0).get_child(1).texture = layers[layer_index][1] #This code is used to update the texture in the animation timeline frame button #but blend_rect causes major performance issues on large images var whole_image := Image.new() whole_image.create(size.x, size.y, false, Image.FORMAT_RGBA8) for layer in layers: whole_image.blend_rect(layer[0], Rect2(position, size), Vector2.ZERO) layer[0].lock() var whole_image_texture := ImageTexture.new() whole_image_texture.create_from_image(whole_image, 0) frame_texture_rect.texture = whole_image_texture func get_layer_container(layer_index : int) -> PanelContainer: for container in Global.vbox_layer_container.get_children(): if container is PanelContainer && container.i == layer_index: return container return null func _draw() -> void: draw_texture_rect(trans_background, Rect2(location, size), true) #Draw transparent background #Onion Skinning #Past if Global.onion_skinning_past_rate > 0: var color : Color if Global.onion_skinning_blue_red: color = Color.blue else: color = Color.white for i in range(1, Global.onion_skinning_past_rate + 1): if Global.current_frame >= i: for texture in Global.canvases[Global.current_frame - i].layers: color.a = 0.6/i draw_texture(texture[1], location, color) #Future if Global.onion_skinning_future_rate > 0: var color : Color if Global.onion_skinning_blue_red: color = Color.red else: color = Color.white for i in range(1, Global.onion_skinning_future_rate + 1): #print(i) if Global.current_frame < Global.canvases.size() - i: for texture in Global.canvases[Global.current_frame + i].layers: color.a = 0.6/i draw_texture(texture[1], location, color) #Draw current frame layers for texture in layers: if texture[3]: #if it's visible draw_texture(texture[1], location) if Global.tile_mode: draw_texture(texture[1], Vector2(location.x, location.y + size.y)) #Down draw_texture(texture[1], Vector2(location.x - size.x, location.y + size.y)) #Down Left draw_texture(texture[1], Vector2(location.x - size.x, location.y)) #Left draw_texture(texture[1], location - size) #Up left draw_texture(texture[1], Vector2(location.x, location.y - size.y)) #Up draw_texture(texture[1], Vector2(location.x + size.x, location.y - size.y)) #Up right draw_texture(texture[1], Vector2(location.x + size.x, location.y)) #Right draw_texture(texture[1], location + size) #Down right #Idea taken from flurick (on GitHub) if Global.draw_grid: for x in size.x: draw_line(Vector2(x, location.y), Vector2(x, size.y), Color.black, true) for y in size.y: draw_line(Vector2(location.x, y), Vector2(size.x, y), Color.black, true) #Draw rectangle to indicate the pixel currently being hovered on var mouse_pos := get_local_mouse_position() - location if point_in_rectangle(mouse_pos, location, location + size): mouse_pos = mouse_pos.floor() if Global.left_square_indicator_visible: var start_pos_x = mouse_pos.x - (Global.left_brush_size >> 1) var start_pos_y = mouse_pos.y - (Global.left_brush_size >> 1) draw_rect(Rect2(start_pos_x, start_pos_y, Global.left_brush_size, Global.left_brush_size), Color.blue, false) if Global.right_square_indicator_visible: var start_pos_x = mouse_pos.x - (Global.right_brush_size >> 1) var start_pos_y = mouse_pos.y - (Global.right_brush_size >> 1) draw_rect(Rect2(start_pos_x, start_pos_y, Global.right_brush_size, Global.right_brush_size), Color.red, false) func generate_layer_panels() -> void: for child in Global.vbox_layer_container.get_children(): if child is PanelContainer: child.queue_free() current_layer_index = layers.size() - 1 if layers.size() == 1: Global.remove_layer_button.disabled = true Global.remove_layer_button.mouse_default_cursor_shape = Control.CURSOR_FORBIDDEN else: Global.remove_layer_button.disabled = false Global.remove_layer_button.mouse_default_cursor_shape = Control.CURSOR_POINTING_HAND for i in range(layers.size() -1, -1, -1): var layer_container = load("res://LayerContainer.tscn").instance() #layer_names.insert(i, "Layer %s" % i) layers[i][2] = "Layer %s" % i layer_container.i = i #layer_container.get_child(0).get_child(2).text = layer_names[i] layer_container.get_child(0).get_child(2).text = layers[i][2] layers[i][3] = true #set visible layer_container.get_child(0).get_child(1).texture = layers[i][1] Global.vbox_layer_container.add_child(layer_container) func camera_zoom() -> void: #Set camera offset to the center of canvas Global.camera.offset = size / 2 #Set camera zoom based on the sprite size var bigger = max(size.x, size.y) var zoom_max := Vector2(bigger, bigger) * 0.01 if zoom_max > Vector2.ONE: Global.camera.zoom_max = zoom_max else: Global.camera.zoom_max = Vector2.ONE Global.camera.zoom = Vector2(bigger, bigger) * 0.002 Global.zoom_level_label.text = "Zoom: x%s" % [stepify(1 / $"../Camera2D".zoom.x, 0.01)] func pencil_and_eraser(mouse_pos : Vector2, color : Color, current_mouse_button : String) -> void: if Input.is_key_pressed(KEY_SHIFT): if !is_making_line: line_2d = Line2D.new() line_2d.width = 0.5 line_2d.default_color = Color.darkgray line_2d.add_point(previous_mouse_pos) line_2d.add_point(mouse_pos) add_child(line_2d) is_making_line = true else: var brush_size := 1 if current_mouse_button == "left_mouse": brush_size = Global.left_brush_size elif current_mouse_button == "right_mouse": brush_size = Global.right_brush_size if is_making_line: fill_gaps(mouse_pos, color, brush_size) is_making_line = false line_2d.queue_free() else: if point_in_rectangle(mouse_pos, location, location + size): mouse_inside_canvas = true #Draw draw_pixel(mouse_pos, color, brush_size) fill_gaps(mouse_pos, color, brush_size) #Fill the gaps #If mouse is not inside bounds but it used to be, fill the gaps elif point_in_rectangle(previous_mouse_pos, location, location + size): fill_gaps(mouse_pos, color, brush_size) func draw_pixel(pos : Vector2, color : Color, brush_size : int) -> void: if Global.can_draw && Global.has_focus && Global.current_frame == frame: #If there is a selection and current pixel is not in it var west_limit := location.x var east_limit := location.x + size.x var north_limit := location.y var south_limit := location.y + size.y if Global.selected_pixels.size() != 0: west_limit = max(west_limit, Global.selection_rectangle.polygon[0].x) east_limit = min(east_limit, Global.selection_rectangle.polygon[2].x) north_limit = max(north_limit, Global.selection_rectangle.polygon[0].y) south_limit = min(south_limit, Global.selection_rectangle.polygon[2].y) var start_pos_x = pos.x - (brush_size >> 1) var start_pos_y = pos.y - (brush_size >> 1) for cur_pos_x in range(start_pos_x, start_pos_x + brush_size): #layers[current_layer_index][0].set_pixel(cur_pos_x, pos.y, color) for cur_pos_y in range(start_pos_y, start_pos_y + brush_size): if point_in_rectangle(Vector2(cur_pos_x, cur_pos_y), Vector2(west_limit - 1, north_limit - 1), Vector2(east_limit, south_limit)): if layers[current_layer_index][0].get_pixel(cur_pos_x, cur_pos_y) != color: #don't draw the same pixel over and over layers[current_layer_index][0].set_pixel(cur_pos_x, cur_pos_y, color) #layers[current_layer_index][0].set_pixelv(pos, color) sprite_changed_this_frame = true func point_in_rectangle(p : Vector2, coord1 : Vector2, coord2 : Vector2) -> bool: return p.x > coord1.x && p.y > coord1.y && p.x < coord2.x && p.y < coord2.y #Bresenham's Algorithm #Thanks to https://godotengine.org/qa/35276/tile-based-line-drawing-algorithm-efficiency func fill_gaps(mouse_pos : Vector2, color : Color, brush_size : int) -> void: var previous_mouse_pos_floored = previous_mouse_pos.floor() var mouse_pos_floored = mouse_pos.floor() mouse_pos_floored.x = clamp(mouse_pos_floored.x, location.x - 1, location.x + size.x) mouse_pos_floored.y = clamp(mouse_pos_floored.y, location.y - 1, location.y + size.y) var dx := int(abs(mouse_pos_floored.x - previous_mouse_pos_floored.x)) var dy := int(-abs(mouse_pos_floored.y - previous_mouse_pos_floored.y)) var err := dx + dy var e2 := err << 1 #err * 2 var sx = 1 if previous_mouse_pos_floored.x < mouse_pos_floored.x else -1 var sy = 1 if previous_mouse_pos_floored.y < mouse_pos_floored.y else -1 var x = previous_mouse_pos_floored.x var y = previous_mouse_pos_floored.y while !(x == mouse_pos_floored.x && y == mouse_pos_floored.y): draw_pixel(Vector2(x, y), color, brush_size) e2 = err << 1 if e2 >= dy: err += dy x += sx if e2 <= dx: err += dx y += sy #Thanks to https://en.wikipedia.org/wiki/Flood_fill func flood_fill(pos : Vector2, target_color : Color, replace_color : Color) -> void: pos = pos.floor() var pixel = layers[current_layer_index][0].get_pixelv(pos) if target_color == replace_color: return elif pixel != target_color: return else: var west_limit := location.x var east_limit := location.x + size.x var north_limit := location.y var south_limit := location.y + size.y if Global.selected_pixels.size() != 0: west_limit = max(west_limit, Global.selection_rectangle.polygon[0].x) east_limit = min(east_limit, Global.selection_rectangle.polygon[2].x) north_limit = max(north_limit, Global.selection_rectangle.polygon[0].y) south_limit = min(south_limit, Global.selection_rectangle.polygon[2].y) if !point_in_rectangle(pos, Vector2(west_limit - 1, north_limit - 1), Vector2(east_limit, south_limit)): return var q = [pos] for n in q: var west : Vector2 = n var east : Vector2 = n while west.x >= west_limit && layers[current_layer_index][0].get_pixelv(west) == target_color: west += Vector2.LEFT while east.x < east_limit && layers[current_layer_index][0].get_pixelv(east) == target_color: east += Vector2.RIGHT for px in range(west.x + 1, east.x): var p := Vector2(px, n.y) draw_pixel(p, replace_color, 1) var north := p + Vector2.UP var south := p + Vector2.DOWN if north.y >= north_limit && layers[current_layer_index][0].get_pixelv(north) == target_color: q.append(north) if south.y < south_limit && layers[current_layer_index][0].get_pixelv(south) == target_color: q.append(south) func _on_Timer_timeout() -> void: Global.can_draw = true