• Documentation
• API
• FAQ
• Quick Reference
• Tutorial

Feedback

• press ENTER to search or ESC to cancel

Table of content

Overview
API Changes

Globals namespaces

• app
• app.command
• app.fs
• app.os
• app.pixelColor
• app.theme
• json

Constants

• Align
• AniDir
• BlendMode
• BrushPattern
• BrushType
• ColorMode
• FilterChannels
• Ink
• MouseButton
• MouseCursor
• RangeType
• SelectionMode
• SpriteSheetDataFormat
• SpriteSheetType
• WebSocketMessageType

Classes/objects

• Brush
• Cel
• Color
• ColorSpace
• Dialog
• Editor
• Events
• Frame
• GraphicsContext
• Image
• ImageSpec
• KeyEvent
• Layer
• MouseEvent
• Palette
• Plugin
• Point
• Properties
• Range
• Rectangle
• Selection
• Site
• Size
• Slice
• Sprite
• Tag
• Tile
• Tileset
• Timer
• Tool
• TouchEvent
• Version
• WebSocket
• Window

---

Edit page

Image

An image object. Each cel has one image (all linked cels share the same image).

An image can be associated to a specific cel or can be alone:

• When you create an image with Image(), the image will be alone so they don't generate undo information.
• A Cel.image is an image associated to a cel, some functions will generate undoable actions.

Image()

local image = Image(width, height [, colorMode])
local image = Image(spec)
local image = Image(sprite)
local image = Image(otherImage)
local image = Image(otherImage, rectangle)
local image = Image{ fromFile=filename }

Creates a new image with the given width and height. The color mode is optional, RGB by default.

The spec parameter is an image specification object.

If you specify otherImage, it's equivalent to use otherImage:clone(). If you specify Image(otherImage, rectangle) this will create a copy of the specified rectangle of otherImage. Image(otherImage, rectangle) may return nil if the rectangle is empty (i.e., if its width and/or height are equal to or less than zero).

If a sprite is given, the image will be a render of the first frame of the sprite. Note: You can use Image:drawSprite() to draw other frame.

If fromFile is given, it indicates a file name (a string) and it will create an image loading the first frame of the given file.

Image:clone()

local copy = image:clone()

Creates a copy of the given image. You can use the Image(otherImage) constructor for the same behavior.

The new image is unrelated to the sprite, but it can be used to update the image in a transaction. E.g. You clone an image, modify the pixels from the copy, and then you patch the image.

Image.id

local id = image.id

Internal unique ID (an integer) assigned to this image inside the program. You can use it to identify this image in your script.

Image.version

local version = image.version

Internal version (an integer) assigned to this image inside the program. It's always incremented when the image is modified. Can be used to cache some complex calculation related to the image and recalculate it only if the version is modified (which means the pixels are modified).

Image.width

local w = image.width

Image.height

local h = image.height

Image.bounds

local rectangle = image.bounds

Returns a rectangle with the bounds of the image with origin equal to (0, 0).

Image.colorMode

local colorMode = image.colorMode

Image color mode.

Image.spec

local spec = image.spec

The specification for this image.

Image.cel

local cel = image.cel

Returns the cel to which this image belongs or nil if this image is not associated to a cel.

Image.bytes

local s = image.bytes
image.bytes = s

A byte string that contains raw image data. The layout depends on the image's color mode. It can be used with the usual string methods, or split into bytes using string.bytes. For editing the image, it's easier to use image:pixels().

The number of bytes must be equal to image.rowStride * image.height.

Image.rowStride

local numberOfBytesPerRow = image.rowStride

Number of bytes for each row in the image. Each image has a total of image.rowStride * image.height bytes. It's useful if you are going to get or set the Image.bytes property manually.

Image.bytesPerPixel

Number of bytes per pixel.

The number of visible pixels per row is Image.width, the number of bytes per row is Image.rowStride, the number of "visible bytes" is image.width * image.bytesPerPixel, and the number of "pixels per row" (visible pixels + hidden/just for alignment) is image.rowStride / image.bytesPerPixel.

Image:clear()

image:clear([ bounds, color ])

Clears the image pixels in a given bounds with a the given color. The entire image is erased by default if no bounds present. image.spec.transparentColor if no color specified.

Image:drawPixel()

image:drawPixel(x, y, color)

Sets the pixel in the xy-coordinate to the given integer pixel value. The xy-coordinate is relative to the image, so pixel (0, 0) is the first pixel at the top-left coorner in the image, not in the sprite canvas.

Warning: This method doesn't create undo information, you should clone the image and then patch it to get proper undo/redo information.

Image:getPixel()

local pixelValue = image:getPixel(x, y)

Returns a integer pixel value for the given xy-coordinate related to the "Image" itself. The returned value will depend on the color mode of the image.

Do NOT confuse with the absolute xy-coordinate in the Canvas. In the following example, we show the differences between getPixel(), Cel.bounds and Cel.position:

Image:drawImage()

destinationImage:drawImage(sourceImage [, position, opacity, blendMode] )

Copies/draws the given sourceImage image over destinationImage. If position is a point, it will draw the sourceImage in the given position (Point(0, 0) as default). opacity is an integer (0 to 255) that defines a general opacity level of the sourceImage (255 by default). blendMode is the blending method to compose the sourceImage over the destinationImage (BlendMode.NORMAL as default).

Warning: If the image is associated with a Cel, this method generates undo information, so you could use it as an individual operation or in a transaction.

Image:drawSprite()

destinationImage:drawSprite(sourceSprite, frameNumber, [, position ] )

Draws the given sourceSprite frame number into the destinationImage. If position is a point, it will draw the sourceSprite in the given position.

Warning: If the image is associated with a Cel, this method generates undo information, so you could use it as an individual operation or in a transaction.

Image:isEqual()

if imageA:isEqual(imageB) then
  print("Both images are equal")
end

Returns true if both images looks the same (spec is equal and all pixels are the same).

Image:isEmpty()

local result = image:isEmpty()

Returns true if all pixels in the image are equal to the transparent color.

Image:isPlain()

local result = image:isPlain(color)

Returns true if all pixels in the image are equal to the given color (which can be a pixel color or a color).

Image:pixels()

iterator = image:pixels()
iterator = image:pixels(rectangle)

Returns a pixel iterator for the whole image or the given rectangle. Then you can use the iterator to read or change pixel values in this way:

for it in image:pixels() do
  local pixelValue = it() -- get pixel
  it(pixelValue)          -- set pixel
  print(it.x, it.y)       -- get pixel x,y coordinates
end

A pixelValue! can be interpreted with the app.pixelColor functions.

Image:putPixel()

Same as Image:drawPixel().

Image:putImage()

Now it's Image:drawImage(). This function is deprecated and was renamed to avoid confusion with the behavior cel.image.

Image:putSprite()

Now it's Image:drawSprite() to match Image:drawImage() name.

Image:saveAs()

image:saveAs(filename)
image:saveAs{ filename=string,
              palette=Palette }

Saves the image as a sprite in the given filename.

To save an indexed image we need to specify a palette parameter (a Palette object). If we don't specify a palette parameter, in case that the image is related to a Cel, we'll use the palette of the Cel's Sprite.

Image:resize()

image:resize(width, height)
image:resize{ width=integer, height=integer, ... }
image:resize{ size=Size(width, height), ... }
image:resize{ ..., method='bilinear' | 'rotsprite' }
image:resize{ ..., pivot=Point(x, y) }

Resizes the image. The pivot is Point(0, 0) by default (i.e. the image right and bottom sides will be increased, and the top-left corner will be in the same place).

Image:shrinkBounds()

local rectangle = image:shrinkBounds()
local rectangle = image:shrinkBounds(refColor)

Returns the shrunken bounds (a rectangle) of the image removing all the empty space of borders using the mask color or the given reference color in refColor.