Fixed mistake in "Putting it all together" page in docs

[jsenn]

By submitting this pull request, I confirm that my contribution is made under the terms of the MIT license.

[parno]

Thanks for catching this! What if we used text like this?

To keep the overflow reasoning simple, we add a precondition to loop_triangle saying that the result needs to fit within a u32.

I think something like the initial clause is needed to imply that this is a somewhat artificial precondition (we're asking the caller to establish a bound on the final value, but presumably it's calling this function because it doesn't know the actual value!).

I'd also be inclined to simplify the code by replacing < 0x1_0000_0000 with <+ u32::MAX. I think that would make the intent of the property clearer.

What do you think?

[tjhance]

I agree we should use u32::MAX, it's helpful for users to know that they can use those

[jsenn]

Updated the text per @parno's suggestion and changed the recursion example to use u32::MAX throughout instead of the magic number.

This probably should've been an issue not a PR so you could word things how you want, sorry about that.

[parno]

Thanks for making the updates! I think it looks good now.

This probably should've been an issue not a PR so you could word things how you want, sorry about that.

No worries -- we appreciate contributions in all forms :)

[tjhance]

yes, PRs are a perfectly fine way to iterate on minor changes

[jaylorch]

Should this use <= u32::MAX instead of < u32::MAX?

[jsenn]

Should this use <= u32::MAX instead of < u32::MAX?

yes it should, good catch: #2256