yage.resource.image

class Image ;

A class for loading and manipulating images. Supports loading images from any channels supported by SDL_Image. Currently supports grayscale, RGB, and RGBA image data.

BUGS:
An RGB image will often be returned when loading grayscale images. Use setFormat(IMAGE_FORMAT_GRAYSCALE) to correct this.

TODO:
Add convolution support: http://www.php.net/manual/en/function.imageconvolution.php#77818

TODO:
Convert to struct?

enum Format ;

The pixel format, also the bytes-per-pixel

this(int channels, int width, int height, ubyte[] lookaside = null);

Create a new emtpy image.

Params:

int channels	number of color channels.
int width	width in pixels
int height	height in pixel

this(ubyte[] data, int channels, int width, int height = 0);

Construct from image data in memory. This does not create a copy of the data.

Params:

image	array of raw image data
int channels	number of color channels.
int width	width in pixels
int height	height in pixels, if 0 it is auto-calculated from width, channels, and data's length.

this(char[] filename);

Construct and load image data from a file.

Params:

char[] filename

absolute or relative path of an image file supported by sdl_image.

Returns:
An image with the number of channels of the source image. Paletted images are converted to nonpaletted.

Pixel!(4) bilinearFilter (float u, float v);

Get the pixel color value at the coordinates using bilinear interpolation.

See:


http:
//en.wikipedia.org/wiki/Bilinear_filtering

Params:

float u	A value betwen 0 and 1.
float v	A value betwen 0 and 1.

Image crop (int left, int top, int right, int bottom);

Crop the image. The four parameters define a box, in coordinates relative to the top left of the source image. For example, crop (0, 0, width, height) would return an exact copy of the original image.

Params:

int left	left side of the cropping box. This and the other parameters can be positive or negative.
int top	top side of the cropping box.
int right	right side of the cropping box
int bottom	bottom side of the cropping box.

Returns:
A new image of the size right-left, bottom-top

ubyte[] getData ();

Return the raw image data. The array length is always width*height*C. Row-major order is used. This means that the array contains all of row 1's pixels, followed by row 2's pixels, etc.

int getChannels ();

Get the number of color channels.

int getWidth ();

Get the width or height of the image in pixels.

int getHeight ();

ditto.

ubyte[] opIndex (int i);
ubyte[] opIndexAssign (ubyte[] val, int i);

Get or set the ith pixel in the image.

ubyte[] opIndex (int x, int y);
ubyte[] opIndexAssign (ubyte[] val, int x, int y);

Get or set the pixel at the given coordinates.

void overlay (Image img, int xoffset = 0, int yoffset = 0);

Paste another image on top of this one. This operation is performed in-place and does not generate any heap activity. This also does not perform proper overlays of images with an alpha channel.

Params:

Image img	Image to add to this one.
int xoffset	x-offset of img from this image's left side. Out of bounds values will be cropped.
int yoffset	y-offset of img from this image's top side. Out of bounds values will be cropped.

void add (Image img, int xoffset = 0, int yoffset = 0);

Overlay another image on top of this one, adding color channel values. This operation is performed in-place and does not generate any heap activity.

Params:

Image img	Image to add to this one.
int xoffset	x-offset of img from this image's left side. Out of bounds values will be cropped.
int yoffset	y-offset of img from this image's top side. Out of bounds values will be cropped.

void overlayAndColor (Image img, Color color, int xoffset = 0, int yoffset = 0);

Convert a monochrome image to color and paste it over this image. This operation is performed in-place and does not generate any heap activity. This somewhat specialized function is used to accelerate text rendering.

Params:

Image img	a monochrome image
Color color	img will be converted to an RGBA image of this color before pasting.
int xoffset	x-offset of img from this image's left side. Out of bounds values will be cropped.
int yoffset	y-offset of img from this image's top side. Out of bounds values will be cropped.

TODO:
Make the top go to the right instead of taking the bottom to the left when skewing

void* ptr ();

Return a c-style pointer to the image data. The length of the array is always width*height*channels.

Image resize (int width, int height = 0);

Resize this image using bilinear interpolation.

Params:

int width	The new width.
int height	The new height. If 0, height will be calculated automatically with aspect ratio maintained.

Returns:
A new image of the same type and of the new size, or an exact copy if the dimensions are the same.

Image subImage (int top, int left, int bottom, int right);

Return a new image that is a sub-image of this image.

TODO:
Replace this with crop. The four parameters should be in pixels. The sub-image will start with the pixels including top and left and stop just before pixels specified by bottom and right.

abstract class ImageBase ;

ubyte[] getBytes ();

Image2!(T2,C2) convert (T2 : real, int C2)();

class Image2 (T : real,int C): ImageBase;

Successor to the Image class.

TODO:
The rest of Image's functionality should be migrated to Image2 , Image deleted, and Image2 renamed as Image. @param T Type of each pixel component @param C number of channels.

this(int width, int height, T[C][] data = null);

ubyte[] getBytes ();