Dense, incomprehensible Joycean tract that will keep the professors busy for centuries.
A NaNoGenMo 2016 project.
License: Riverrun is released under an MIT License. It utilizes the Olipy library, which is released under the GNU Public License v3.
UlyssesTextToData.py - this script processes the Gutenberg Ulysses text and parses it into a data file that is readable by the Olipy library.
This process changes for some chapters, as (for example) Penelope (Ch. 18) consists of 5 extremely long run-on sentences, and must be split up in a different way. Similarly, Aeolus (Ch. 7) must be parsed to separate newspaper headlines from the rest of the text.
This file defines a method for standard chapters, and a method for each chapter that requires its own special treatment.
QueneauNovel.py - uses a queneau generator to create a new novel. Because the queneau generator uses the original novel as the corpus, it works much better when generating a finished product smaller than the original corpus.
MarkovNovel.py - uses a Markov generator to create a new novel. This is much less sensible, has many more grammatical mistakes, but generates much more interesting content.
MakeHTML.py - a very, very simple script to dump out the contents of a novel (in a Unicode text file) to an HTML file. This also correctly handles Unicode symbols and translates them into the corresponding HTML codes.