This seems all over the place. You're marketing this to, what, people who have experience in C and want a better experience?

Why is there a readme link to download a .zip of the repo? I know how to clone.

Why is there a link to wikipedia's section on object orientation? I know that.

I don't understand the updates section. v3 parser implemented...okay? How is that better? What?

Then this big example which looks a lot like vanilla C. Is this the sum of what Nymph offers?

What I really want to know, and what is absent is:

* What does this give me that ordinary C doesn't?

* What are the longer-term goals of this project?

* Who is this intended for?

* How compatible will this be with ordinary C? Is there an interest in, say, gradual migration if this is to take off? What's the C <=> Nymph interop to look like?

I guess this could do with more details on the Readme... An api documentation perhaps?

Making bogus accounts to upvote stuff will get your account banned on HN, as well as the bogus accounts, as well as related accounts.

This is of no use for me unfortunately. I don't miss OO syntax from C at all. OO by default leads to badly architected software, as learned emprically from many projects. I'm perfectly happy with the minimal OO support that C brings - Just pass an explicit "vtable" (struct of function pointers) in the few cases where it makes sense.

Instead of OO I would like to know if we can do something about the (mostly compile performance) problem that importing external interfaces is done by including text files, and that compilation means writing intermediate results to disk.

Also there might or might not be a nice solution for better compile time code generation (just for the rare things where it's really useful, like support for type-generic value printing when debugging).

Jonathan Blow has made some really impressive advancements with his Jai language (his own language) compiler https://www.youtube.com/watch?v=gWv_vUgbmug. Compiling a game engine and a little game (50K lines of code I think) in less than a second. Whether his ideas work out well in the end or just work for himself remains to be shown, but it's very interesting to watch the video in any case.

I think some comments in the example nymph code explaining what each line does and the differences from C could go a really long way here.

Does malloc do something different? Explain in a comment.

What are the two different calls to `print()` supposed to demonstrate? How does nymph know which one to call?

In the `obj` block, there are initializers/default values, it seems. When are those set? How do they work?

previously discussed at [0].

Suggestions for the author:

Define the grammar in BNF and the semantics in prose.

Consider using a lexical scanner generator, a parser generator, and a string library.

Don't call strlen in a loop[1], many of your for loops are O(n^2) instead of O(n).

malloc returns uninitialized memory. Calling strcat on the return value of malloc[2] is a bug.

[0] https://news.ycombinator.com/item?id=15045592

[1] https://github.com/roecrew/nymph/blob/master/nymph/helper_fu...

[2] https://github.com/roecrew/nymph/blob/master/nymph/nymph_pp....

Edited to add the last two sentences.

The code fragment

    Box **myBoxes = malloc(10*sizeof(Box));
    Box *myBox = malloc(sizeof(Box));
    myBoxes[0] = myBox;

looks like it should have used 10 times the size of a Box pointer.