Partial parsing and exposing the next expected tokens

[aybabtme]

Hi, thanks for this project. It's very useful!

I was wondering if there's a recommended way to handle failed parses, to get a partial result and enumerate the next valid rules that would have been expected. My motivation is to build an auto-complete feature for a small language. I've been struggling to grok the internals of the Peg struct being created. Am I missing something obvious that allows what I'm looking for, or should something change in the implementation to expose the information required to implement such a functionality?

[pointlander]

Parse does return an error type that has the "deepest" parse. In a typical peg grammar the parser attempts to parse the input in several different ways, so the correct next token is not clear.

[pointlander]

You could use peg to define your own parsing language, and your parser could use levenshtein distance for parsing.