Just scan the books, OCR the scans, and remove the copyrighted information that the publisher added to the codes. You then have a perfectly legal, digital copy. Nobody does this because the codes change over time and there's no way to protect your work. It would be freely available to everyone as soon as the first copy hits an open server. It would also be nearly worthless as soon as the new version comes out (do you want to be the attorney that shows up to court without knowing the new changes?).