3 Cache-image Snip

NOTE: This library is deprecated. Use racket/gui instead. This library will no longer be public in a future release; much of it will be available privately to continue to support the implementation of htdp/image, but the other exported functions here are not useful and have names that confusingly match unrelated other libraries.

The mrlib/cache-image-snip library provides the core data structure for DrRacket’s "image.rkt" teachpack. Images in the "image.rkt" teachpack are instances of the cache-image-snip% class.

The library also defines a new type, argb, that represents a bitmap, but with alpha values. It has a maker, two selectors, and a predicate.

class
cache-image-snip% : class?

superclass: image-snip%

The cache-image-snip% class is a subclass of image-snip% simply so that its instances can be compared with image-snip% using equal?. All image-snip% functionality is overridden or ignored.

method
(send a-cache-image-snip equal-to? snip
equal?) → boolean?
  snip : (is-a?/c image-snip%)
  equal? : (any/c any/c . -> . boolean?)
Calls the other-equal-to? method of snip if it is also a cache-image-snip% instance, otherwise calls the other-equal-to? of a-cache-image-snip.
method
(send a-cache-image-snip get-argb) → argb?
Returns a pixel array for this image, forcing it to be computed.
method
(send a-cache-image-snip get-argb-proc)
→ (argb? exact-integer? exact-integer? . -> . void?)
Returns a procedure that fills in an argb with the contents of this image at the given offset
method
(send a-cache-image-snip get-argb/no-compute)
→ (or/c false/c argb?)
Returns a pixel array for this image or #f if it has not been computed yet.
method
(send a-cache-image-snip get-bitmap)
→ (or/c false/c (is-a?/c bitmap%))
Overrides get-bitmap in image-snip%.
Builds (if not yet built) a bitmap corresponding to this snip and returns it.
If the width or the height of the snip is 0, this method return #f.
method
(send a-cache-image-snip get-dc-proc)
→ (or/c false/c ((is-a?/c dc<%>) real? real? -> void?))
Either returns false, or a procedure that draws the contents of this snip into a dc.
method
(send a-cache-image-snip get-pinhole) →
real? real?
Returns the pinhole coordinates for this image, counting from the top-left of the image.
method
(send a-cache-image-snip get-size)
→
exact-nonnegative-integer?
exact-nonnegative-integer?
Returns the width and height for the image.
method
(send a-cache-image-snip other-equal-to? snip
equal?) → boolean?
  snip : (is-a?/c image-snip%)
  equal? : (any/c any/c . -> . boolean?)
Overrides other-equal-to? in image-snip%.
Refines the comparison of other-equal-to? in image-snip% to exactly match alpha channels.

value
snip-class : (is-a?/c snip-class%)

This snipclass is used for saved cache image snips.

procedure
(make-argb vectorof width height) → argb?
  vectorof : byte?
  width : exact-nonnegative-integer?
  height : exact-nonnegative-integer?

Constructs a new argb value with the given width and height, using the data in the vector. The vector has four entries for each pixel, an alpha, red, green, and blue value. The pixels are specified in row-major order, so that the pixel at location (x,y) comes from vector entry (4*(x+(width*y))).

procedure
(argb-vector argb) → (vectorof byte?)
argb : argb?

Extracts the vector from argb. The resulting vector has entries in row-major order, so that the data for the pixel at (x,y) winds up in four vector entries beginning at (4*(x+(width*y))).