vpClassifyScanline is used to incrementally compute a preclassified volume by classifying one scanline of
voxels and storing the result in the current preclassified volume.
The voxels argument is a 1D array containing voxels in the format previously specified with
vpSetVoxelSize and vpSetVoxelField. vpVolumeSize must be called to declare the size of the volume before
the first call to vpClassifyScanline, and the length of the 1D voxel array must be equal to the length of
the X dimension of the volume. The 1D voxel array may be initialized with vpScanlineNormals or a user-
defined function.
The voxels in the array are classified using the current opacity transfer function (see
vpSetClassifierTable(3)) and appended to the preclassified volume (see vpClassifyVolume(3)). Scanlines
must be processed in Z-major order: all of the scanlines in the Z=0 slice are processed from Y=0 onward,
then the Z=1 slice is processed, and so on. When the last scanline in the volume has been processed the
preclassified data structure is ready to be used for rendering. Rendering is not possible until all
scanlines have been processed.
When the first scanline of a volume is passed to vpClassifyScanline, any existing preclassified volume
data is destroyed.