WIP: Feature/expr sig

[dromer]

Putting it up for review, even though it's not nearly done yet.

Still needed:

• implement missing operators/functions: https://github.com/Wasted-Audio/hvcc/pull/104/files#diff-abe0967ea77175617953b26cecd4092b9c6cc7653cfcc77f682dd464b59a65f4R149
• de-duplicate expressions - use unique hashes as the function names?
• tests
  • control rate tests should be pretty easy
  • signal rate tests still need a method to either generate correct reference files or run a similarity on the output

[dromer]

@dgbillotte whenever you have time can you check some of the changes I've ported into your branch?

[dgbillotte]

I've just recently been trying to follow these again and was surprised to see my handle... I will try to check it out this weekend. Worst case, I have a week off in a couple of weeks (aug 7-11) and can definitely get my head into it then.

…

On Fri, Jun 23, 2023 at 6:19 PM dreamer ***@***.***> wrote: @dgbillotte <https://github.com/dgbillotte> whenever you have time can you check some of the changes I've ported into your branch? — Reply to this email directly, view it on GitHub <#104 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AADYHBGZQ57IV6LAITWQKO3XMY6A7ANCNFSM6AAAAAAZMYDPTU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[dromer]

@dgbillotte did you have a chance to have a look at this again? :)

I do think that we might be painting ourselves a bit into a corner here, as we won't be able to use any of the more recent features of expr~ (like signal objects and others).

However decomposing the expression into a graph might pose additional problems as we'd need to deal with proper control flow as well (particularly for control rate expressions), which is very much non-trivial

So ultimately providing a minimal way to do expr/expr~ might be preferable over trying (and failing) to really decompose the expression into a "proper" heavy graph.