[reltuk]

Similar arguments are also seen in https://graphitemaster.github.io/aau/.

When using unsigned ints for backward iteration, I'm partial to looping as:

  for (i = size - 1; i < size; i--) {
    ...
  }

since it has so much symmetry with forward iteration.

[10000truths]

I would rather just use a plain for loop and manipulate the index in the loop body:

  for(uint32_t i = 0; i < length; ++i) {
      i = (length - 1) - i;
      // ...
  }

Much easier for me to understand, making it less likely to mess anything up.

[edflsafoiewq]

You'll need a new variable there, but otherwise yeah.

[WalterBright]

Even easier (for D):

    foreach (i; 0 .. length)

D takes care of selecting the correct type for `i`.

[phkahler]

I'd argue that is a terrible way to write it since it depends on the rollover to terminate. Using i >= 0 is less mental work but also requires that i be signed.

[ghoward]

Rollover on unsigned ints is well-defined in C.

[hbn]

I think their point is that it's less well defined in the average human's brain

When most people see 0 - 1 they don't immediately think 11111111111111111111111111111111

[ghoward]

We're talking about C, though. I hope the average human is not writing C.

[jcelerier]

the average person writing C is absolutely an average human in terms of relationship with mathematics

[xdavidliu]

if we used i >= 0 and someone were to come in and change "int i" to "unsigned i" or "size_t i", the result would be catastrophic

[Leherenn]

The same argument can be applied to GP's code with the added bonus it will cause UB at some point.

[JonChesterfield]

`for (unsigned i = n; i-->0;) {}`

Probably formatted i-- > 0. Concise and reads (once used to it anyway) something like i decrementing to zero.