> On Apr 15, 2016, at 11:14 AM, Erica Sadun <[email protected]> wrote:
>
> On Apr 15: 2016: at 11:56 AM: Joe Groff via swift-evolution
> <[email protected] <mailto:[email protected]>> wrote:
>>
>>
>>> On Apr 15: 2016: at 8:20 AM: Stephen Canon via swift-evolution
>>> <[email protected] <mailto:[email protected]>> wrote:
>>>
>>> “e” is a great bike-shedding example. While it would definitely allow a
>>> fluent style if you know what you’re looking at: I worry a little bit about
>>> readability of `Float.e` or `.e` in source. Most programmers are at least
>>> passingly familiar with pi: but that familiarity doesn’t necessarily extend
>>> to e. IDEs and docstrings make this palatable: of course.
>>
>> It's also questionable to me how often `e` itself is interesting independent
>> of an exp() or log() operation.
>
> I'm sure it gets used but I'd imagine it would be at an order of magnitude or
> more less than PI. Quick gist count:
>
> M_E: 109
> M_LOG2E: 23
> M_LOG10E: 19
> M_LN2: 24
> M_LN10: 10
> M_PI: 2,231
> M_PI_2: 255
> M_PI_4: 54
> M_1_PI: 27
> M_2_PI: 25
> M_2_SQRTPI: 23
> M_SQRT2: 31
> M_SQRT1_2: 26
>
> That said, I use M_PI_4 a lot because it's really handy for sample code. I'm
> one of the
> outlier users whose use is not reflective of this quick check.
M_PI_2 and M_PI_4 are interesting cases; they date back to a time when
compilers couldn’t be trusted to constant-fold computations like M_PI / 2.
Since Swift quite reliably does this transformation, I would prefer to simply
use the more explicit .pi / 2. How does that strike you as a user?
– Steve
_______________________________________________
swift-evolution mailing list
[email protected]
https://lists.swift.org/mailman/listinfo/swift-evolution
