(Required) The width of the source image in samples.

(Required) The height of the source image in samples.

(Required) The number of bits used to represent each color component.

(Optional) true if the image should be treated as a mask, false otherwise.

(Optional) true if interpolation is performed, false otherwise. Interpolation attempts to smooth transitions between sample values.

true if decode is used, false otherwise.

(Optional) An array of numbers specifying the mapping from sample values in the image to values appropriate for the current color space.

An ASAtom representing the color space name.

true if the color space is indexed, false otherwise.

(Optional) This is used if isIndexed is true. Colors are specified by integers in the range 0 to hival.

(Required for images, not allowed for image masks) A Cos object of the color space used for the image samples.

The length of the sample data in bytes.

The number of components in colorSpace. For instance, comps is 3 for an RGB color space.

gray

1

RGB

3

CMYK

4

Lab

3

The image whose lookup table is obtained.

(Filled by the method) An array for the returned color space information.

The length of data in bytes.

The image whose attributes are obtained.

(Filled by the method) A pointer to a PDImageAttrs structure containing the image attributes.

It must be sizeof(PDImageAttrs).

The value passed to the image selector as a parameter to PDImageSelectAlternate().

The matrix that will replace the imageToUserMatri (see PDImageSelGetGeoAttr()). The imageToDevMatrix is automatically calculated from newUserMatrix.

The Cos object for this image XObject's base image. Under some circumstances, PDImageSelectAlternate() can be called with an XObject type other than an image; for example, a form. Because of this, your code must check the XObject's subtype (you can use CosDictGet() to read the value of the XObject's Subtype key). If the subtype is not Image, your code must not modify the XObject, but simply return the CosObj that was passed to you.

true if printing, false if displaying.

The amount of time, in ticks, before the image selector must return control to the Acrobat viewer. This parameter is not relevant for image selectors that simply choose an existing alternate or create a new image XObject using Cos methods, but it is relevant for image selectors that create a new image XObject by calculating or reading data from a network or a slow device.

• If tickLimit is zero, the image selector must not return control to the Acrobat viewer until the selector can provide the alternate image to use.

• If tickLimit is nonzero, the image selector does not have to provide the image XObject within tickLimit, but it must raise the fileErrBytesNotReady exception if it cannot. This returns control to the Acrobat viewer and informs it that the data is not ready. The Acrobat viewer then calls the image selector periodically until the image selector returns the selected image XObject.

If true, the image data returned to the Acrobat viewer is cached for future use. If false, it is not. Pass true if you expect your image data to remain valid for at least several calls to your client. Pass false if you expect your image data to change on the next call to your client. If you create and return a Cos stream object with an external file system and the stream's data will not always be the same, you must set cacheImageP to false. If you fail to do this, the Acrobat viewer may use cached image data when it should not.

An opaque pointer containing data that must be passed in other image selector calls.

The image XObject to use. Returning a NULL Cos object (obtained using CosNewNull) tells the Acrobat viewer to skip this XObject entirely.

(Filled by the method) The value passed to the image selector as a parameter to PDImageSelectAlternate().

(Filled by the method) The destination device's color space. It will be one of the following:

If the device has some other color space or its color space cannot be determined, PDDeviceRGB is returned.

(Filled by the method) The number of bits used for each pixel. For example, a device with 8 bits red, 8 bits green, and 8 bits blue would have a bitsPerPixel of 24. If bitsPerPixelP has a value of 0 (instead of the more standard 1, 8, or 24), the number of bits per pixel on the output device could not be determined.

(Filled by the method) The output device type. It will be one of the following:

(Filled by the method) The value passed to the image selector as a parameter to PDImageSelectAlternate().

(Filled by the method) The rectangle bounding the region of the page to update. This is the intersection of the visible portion of the page, any update regions, and any clipping paths that have been explicitly set in the PDF file. Its coordinates are specified in default user space.

(Filled by the method) A matrix specifying the transformation from image space (the space in which all images are 1x1 units) to default user space (the space with 72 units per inch). It contains sufficient information for the image selector plug-in to determine the image's size and location on the page. For a normal page and image (an upright image on a non-rotated page), the following is true:

• imageToDefaultMatrixP->a is equal to the image horizontal size in 1/72 of an inch units.

• imageToDefaultMatrixP->b = 0

• imageToDefaultMatrixP->c = 0

• imageToDefaultMatrixP->d is equal to the image vertical size in 1/72 of an inch units.

• imageToDefaultMatrixP->h is equal to the left edge of image.

• imageToDefaultMatrixP->v is equal to the bottom edge of the image.

In other words, this matrix provides the image's height, width, and position on the page, all in units of points (compare to imageToDefaultMatrixP). The intersection of the rectangle obtained by transforming a 1x1 unit rectangle (the image) through imageToDeviceMatrixP and the updateBBoxP rectangle is the region of the image that is actually drawn. This is the region of the image for which data is required.

(Filled by the method) A matrix specifying the transformation from image space (the space in which all images are 1x1 unit) to device space (the space in which one unit is one pixel). This matrix provides the image's height, width, and position on the page, all in pixels (compare to imageToDefaultMatrixP).