Get the latest tech news
Adding row polymorphism to Damas-Hindley-Milner
Intro to HM
In order to unify and print rows, we’ll often deal with them in their flattened representation, which is a collapsed tuple of fields and a rest type. Pattern matching inference changed a little bit since the last post but not enough to make a big deal out of it. I thought that there was some quirk or trick in the representation that enabled row polymorphism, but that’s not the case at all(shout-out to Sinan and River for making this clear).
Or read this on Hacker News