The full matrix vector multiplication routine using the full 4x4 values. Also changes the secret w component in the vec3 to w which is only really eve

Used to multiply direction vectors or when you do not what your vector to be translated. Implicitely makes the matrix a 3x3

fast 0..255 * 0..255 => 0..255 rounded multiplication

---------------------------------PLANE------------------------------------// Returns a negative value if not aligned in the same direction of plane no

take a -128..127 value and stbi__clamp it and convert to 0..255

returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow

initialize a callback-based context

initialize a memory-decode context

if there's a pending marker from the entropy stream, return that otherwise, fetch from the stream and get a marker. if there's no marker, return 0xff,

returns 1 if the product is valid, 0 on overflow. negative factors are considered invalid.

Tiling has invalidated my bilinear filtering code but it is here for posterity

decode a jpeg huffman value from the bitstream

after a restart interval, stbi__jpeg_reset the entropy decoder and the dc prediction

clean up the temporary component buffers

combined JPEG 'receive' and JPEG 'extend', since baseline always extends everything it receives.

returns 0..31 for the highest set bit

mallocs with size overflow checking

extract an arbitrarily-aligned N-bit value (N=bits) from v, and then make it 8-bits long and fractionally extend it to full full range.

returns STBI_rgb or whatever, 0 on error

Cannot use memset to clear a float since the binary Representation is more complex. We just iterate through the whole thing and explicitely set it to

Used by input to reset camera to origin in case user loses their bearings

return 1 if the sum is valid, 0 on overflow. negative terms are considered invalid.

Taken from Mesa implementation of GLU library

create the png data from post-deflated data

this function is designed to support animated gifs, although stb_image doesn't support it two back is the image from two frames ago, used for a very s

decode one 64-entry block--

get some unsigned bits

set up the kernels