Make position of * in pointer declarations in include/ consistent #1252

No description provided.

I think we should move it the other direction, so the * is part of the type and not part of the name.

I went with the traditional convention as used in the C standard. I'd agree with you if the position of * was entirely arbitrary. But making * part of the name makes more sense in C because it applies to the variable and not the type. For example, char* foo, bar; declares foo to be a pointer to char and bar to be a char.

I think they're arguments for either side. But yeah, we should make it consistent.

utACK https://github.com/bitcoin-core/secp256k1/pull/1252/commits/3d1f430f9f32d45885b0a10b448c0f15386c423d

I would ACK the other variant as well.

I think we should move it the other direction, so the * is part of the type and not part of the name.

+1 on that. Especially in the case when this repo is a subtree in other C++ project (like Bitcoin Core), which uses the same convention.

utACK 3d1f430f9f32d45885b0a10b448c0f15386c423d. I have not verified these are the only instances where changes would need to be made.

In general, concept ACK on making these types consistent across the codebase. I have a slight preference for * hugging the variable rather than the type, because it's closer to how the C semantics for types actually work, but consistency is more important than bike shedding this if there is disagreement.

utACK 3d1f430 from me too. I also value consistency more than either specific choice.'