con-nf VS lean4-metaprogramming-book

But like, you can look at what parts of Mathlib this development imports, it's mainly stuff imported by files in this subdirectory https://github.com/leanprover-community/con-nf/tree/main/Con... , and it's pretty basic things: the definition of a permutation, a cardinal number etc. Almost all of these are things that would feature in the first one or two years of an undergraduate math degree (from just quickly scanning it, the most advanced thing I could see is the definition of cofinality of ordinals). It seems practically impossible to me that someone would make a mistake when e.g. defining what a group is, in a way subtle enough to later break this advanced theorem. If you think that people could mess up that, then all of math would be in doubt.

Lean4 manages to pull off changing the parser on the fly at compile time. You can add new productions, add new syntax node types, and add new tokens. Then define macros or code to process the additional syntax. Here is a sample I found that adds a simple JSX-like syntax starting around line 93 and then uses it at line 169:

https://github.com/leanprover/lean4/blob/master/tests/playgr...

I believe most of the language is defined this way, although it is pre-compiled.

For more details see the lean4 metaprogramming book: https://github.com/arthurpaulino/lean4-metaprogramming-book