NAME

Tk_FindPhoto, Tk_PhotoPutBlock, Tk_PhotoPutZoomedBlock, Tk_PhotoGetImage, Tk_PhotoBlank, Tk_PhotoExpand, Tk_PhotoGetSize, Tk_PhotoSetSize -
manipulate the image data stored in a photo image.

SYNOPSIS

#include <tk.h>

Tk_PhotoHandle
Tk_FindPhoto(interp, imageName)

int
Tk_PhotoPutBlock(interp, handle, blockPtr, x, y, width, height, \
compRule)

int
Tk_PhotoPutZoomedBlock(interp, handle, blockPtr, x, y, width, height, \
zoomX, zoomY, subsampleX, subsampleY, compRule)

int
Tk_PhotoGetImage(handle, blockPtr)

void
Tk_PhotoBlank(handle)

int
Tk_PhotoExpand(interp, handle, width, height)

void
Tk_PhotoGetSize(handle, widthPtr, heightPtr)

int
Tk_PhotoSetSize(interp. handle, width, height)

ARGUMENTS

DESCRIPTION

Tk_FindPhoto returns an opaque handle that is used to identify a particular photo image to the other procedures. The parameter is the name of the image, that is, the name specified to the image create photo command, or assigned by that command if no name was specified.

Tk_PhotoPutBlock is used to supply blocks of image data to be displayed. The call affects an area of the image of size width x height pixels, with its top-left corner at coordinates (x,y). All of width, height, x, and y must be non-negative. If part of this area lies outside the current bounds of the image, the image will be expanded to include the area, unless the user has specified an explicit image size with the -width and/or -height widget configuration options (see photo(n)); in that case the area is silently clipped to the image boundaries.

The block parameter is a pointer to a Tk_PhotoImageBlock structure, defined as follows:

typedef struct {
    unsigned char *pixelPtr;
    int width;
    int height;
    int pitch;
    int pixelSize;
    int offset[4];
} Tk_PhotoImageBlock;

The pixelPtr field points to the first pixel, that is, the top-left pixel in the block. The width and height fields specify the dimensions of the block of pixels. The pixelSize field specifies the address difference between two horizontally adjacent pixels. Often it is 3 or 4, but it can have any value. The pitch field specifies the address difference between two vertically adjacent pixels. The offset array contains the offsets from the address of a pixel to the addresses of the bytes containing the red, green, blue and alpha (transparency) components. These are normally 0, 1, 2 and 3, but can have other values, e.g., for images that are stored as separate red, green and blue planes.

The compRule parameter to Tk_PhotoPutBlock specifies a compositing rule that says what to do with transparent pixels. The value TK_PHOTO_COMPOSITE_OVERLAY says that the previous contents of the photo image should show through, and the value TK_PHOTO_COMPOSITE_SET says that the previous contents of the photo image should be completely ignored, and the values from the block be copied directly across. The behavior in Tk8.3 and earlier was equivalent to having TK_PHOTO_COMPOSITE_OVERLAY as a compositing rule.

The value given for the width and height parameters to Tk_PhotoPutBlock do not have to correspond to the values specified in block. If they are smaller, Tk_PhotoPutBlock extracts a sub-block from the image data supplied. If they are larger, the data given are replicated (in a tiled fashion) to fill the specified area. These rules operate independently in the horizontal and vertical directions.

Tk_PhotoPutBlock normally returns TCL_OK, though if it cannot allocate sufficient memory to hold the resulting image, TCL_ERROR is returned instead and, if the interp argument is non-NULL, an error message is placed in the interpreter's result.

Tk_PhotoPutZoomedBlock works like Tk_PhotoPutBlock except that the image can be reduced or enlarged for display. The subsampleX and subsampleY parameters allow the size of the image to be reduced by subsampling. Tk_PhotoPutZoomedBlock will use only pixels from the input image whose X coordinates are multiples of subsampleX, and whose Y coordinates are multiples of subsampleY. For example, an image of 512x512 pixels can be reduced to 256x256 by setting subsampleX and subsampleY to 2.

The zoomX and zoomY parameters allow the image to be enlarged by pixel replication. Each pixel of the (possibly subsampled) input image will be written to a block zoomX pixels wide and zoomY pixels high of the displayed image. Subsampling and zooming can be used together for special effects.

Tk_PhotoGetImage can be used to retrieve image data from a photo image. Tk_PhotoGetImage fills in the structure pointed to by the blockPtr parameter with values that describe the address and layout of the image data that the photo image has stored internally. The values are valid until the image is destroyed or its size is changed. Tk_PhotoGetImage returns 1 for compatibility with the corresponding procedure in the old photo widget.

Tk_PhotoBlank blanks the entire area of the photo image. Blank areas of a photo image are transparent.

Tk_PhotoExpand requests that the widget's image be expanded to be at least width x height pixels in size. The width and/or height are unchanged if the user has specified an explicit image width or height with the -width and/or -height configuration options, respectively. If the image data are being supplied in many small blocks, it is more efficient to use Tk_PhotoExpand or Tk_PhotoSetSize at the beginning rather than allowing the image to expand in many small increments as image blocks are supplied.

Tk_PhotoExpand normally returns TCL_OK, though if it cannot allocate sufficient memory to hold the resulting image, TCL_ERROR is returned instead and, if the interp argument is non-NULL, an error message is placed in the interpreter's result.

Tk_PhotoSetSize specifies the size of the image, as if the user had specified the given width and height values to the -width and -height configuration options. A value of zero for width or height does not change the image's width or height, but allows the width or height to be changed by subsequent calls to Tk_PhotoPutBlock, Tk_PhotoPutZoomedBlock or Tk_PhotoExpand.

Tk_PhotoSetSize normally returns TCL_OK, though if it cannot allocate sufficient memory to hold the resulting image, TCL_ERROR is returned instead and, if the interp argument is non-NULL, an error message is placed in the interpreter's result.

Tk_PhotoGetSize returns the dimensions of the image in *widthPtr and *heightPtr.

PORTABILITY

In Tk 8.3 and earlier, Tk_PhotoPutBlock and Tk_PhotoPutZoomedBlock had different signatures. If you want to compile code that uses the old interface against 8.4 without updating your code, compile it with the flag -DUSE_COMPOSITELESS_PHOTO_PUT_BLOCK. Code linked using Stubs against older versions of Tk will continue to work.

In Tk 8.4, Tk_PhotoPutBlock, Tk_PhotoPutZoomedBlock, Tk_PhotoExpand and Tk_PhotoSetSize did not take an interp argument or return any result code. If insufficient memory was available for an image, Tk would panic. This behaviour is still supported if you compile your extension with the additional flag -DUSE_PANIC_ON_PHOTO_ALLOC_FAILURE. Code linked using Stubs against older versions of Tk will continue to work.

CREDITS

The code for the photo image type was developed by Paul Mackerras, based on his earlier photo widget code.

KEYWORDS

photo, image