The really unfortunate thing about that post is it completely fails to understand the essential part of solving the TSP this way, which are the lazily-added sub

The really unfortunate thing about that post is it completely fails to understand the essential part of solving the TSP this way, which are the lazily-added subtour elimination constraints. I don't think the blog author at Forio really understood the code they were using, hopefully they will update it one day!