I took a quick pass at constructing a single HTML file version. However, the HTML markup is very old and needs some love.

Here's a simple slicer that converts to a PPM image stream.
I tested briefly piping to ffplay - from the ffmpeg project.
Note that it ignores aspect ratio.

#include <stdio.h>
#include <stdlib.h>

#include "r4_image.h"

struct S_IMAGEHDR header;

int main(int argc, char **argv)
{
  if (1 != fread(&header, sizeof(header), 1, stdin))
  {
    fprintf(stderr, "%s: failed to read header\n", argv[0]);
    return 1;
  }
  if (header.magic != R4_IMAGE_ID)
  {
    fprintf(stderr, "%s: header magic != Ray4\n", argv[0]);
    return 1;
  }
  if (header.version != 1)
  {
    fprintf(stderr, "%s: header version != 1\n", argv[0]);
    return 1;
  }
  if (header.bitsperpixel != 24)
  {
    fprintf(stderr, "%s: header bitsperpixel != 24\n", argv[0]);
    return 1;
  }
  int width  = header.last[0] - header.first[0] + 1;
  int height = header.last[1] - header.first[1] + 1;
  int depth  = header.last[2] - header.first[2] + 1;
  int bytes = 3 * width * height;
  char *buffer = malloc(bytes);
  if (! buffer)
  {
    fprintf(stderr, "%s: memory allocation failed\n", argv[0]);
  }
  for (int frame = 0; frame < depth; ++frame)
  {
    if (0 > printf("P6\n%d %d\n255\n", width, height))
    {
      fprintf(stderr, "%s: write error\n", argv[0]);
      free(buffer);
      return 1;
    }
    if (1 != fread(buffer, bytes, 1, stdin))
    {
      fprintf(stderr, "%s: read error\n", argv[0]);
      free(buffer);
      return 1;
    }
    if (1 != fwrite(buffer, bytes, 1, stdout))
    {
      fprintf(stderr, "%s: write error\n", argv[0]);
      free(buffer);
      return 1;
    }
  }
  free(buffer);
  return 0;
}

ray4 parser always says unexpected 0xfe, which comes from ReadChar returning -2 (the code for nothing in the un-read buffer).

Commit ed1f0fb did something bogus it seems, it needs to actually getc somewhere.

It might be useful to embed the images in the HTML file directly, rather than fetching out of a directory. That would yield a single HTML file that contains the entire thesis that people could download. Good excuse to investigate binary data encoding. For example, can the images be placed anywhere in the HTML file and then referenced in the body of the thesis?

In file r4/src/r4_vec.c:

short V4_Normalize (Vector4 V)
{
    double norm; 

    norm = V4_Dot (V,V);

    if (norm < 1e-30) return 0;

    V4_Scalar (V, /=, norm);

    return 1;
}

Taking norm as dot product causes invalid lighting.

Sqrt of dot product should be taken

What's the output? Possibly a special 3D image file, plus a viewer that shows you a specific slice. Or generate movies. For the first iteration, do something crude and simple.

Most of the READMEs are independent and thus there's no cohesive presentation of the project as a whole. Revisit the project structure and provide a better document collection for the whole and for each component.

As a CMake neophyte I'm not sure this is the best solution but it works for me:

diff --git a/ray4/CMakeLists.txt b/ray4/CMakeLists.txt
index 79cf8c9..a05d896 100644
--- a/ray4/CMakeLists.txt
+++ b/ray4/CMakeLists.txt
@@ -21,3 +21,4 @@ set ( sources_ray4
 )
 
 add_executable (ray4 ${sources_ray4})
+target_link_libraries(ray4 -lm)

See figure 4.7a.

Also, use element id's rather than dummy elements.

These include:

• ray4/craig/README.txt
• ray4/ray4.doc
• thesis/PostScript/figures/README.txt
• wire4/psplot.txt
• wire4/wire4.doc