[Cfp-interest 3086] Re: csinh(x + i y) - G.6.3.5 - 5th bullet point of special cases

[Jim Thomas]

Conjugate, odd, and even were used where it seemed the extra bullets required to do otherwise would expand the specification unduly. Using them was not an absolute principle. For example, a better grouping of the bullets might have seemed more important. There might be inconsistencies here that would be important to remove, but we need to ask whether a considered change would lead to an overall significantly clearer specification? 

- Jim Thomas

> On Apr 6, 2024, at 10:53 PM, Damian McGuckin <damianm at esi.com.au> wrote:
> 
> 
> The domain for
> 
> 	csinh(x + i INF)
> 
> is written to exploit the fact that the function is odd. It says
> 
> 	positive finite x
> 
> It could have ignored the fact that the function is odd and said
> 
> 	finite non-zero
> 
> The fact that the function is odd is ignored for the domain for
> 
> 	csinh(x + i NaN)
> and
> 	ctanh(x + i INF)
> and
> 	ctanh(x + i NAN)
> 
> which return the same result, NaN + i NaN
> 
> They ignore the fact that the function is odd and say
> 
> 	finite non-zero x
> 
> Should things be consistent?
> 
> I think we should exploit the fact that the function is odd but I am easy either way. But I think consistency should be pursued. I have not gone through all thororoughly but it seems to ignore that fact more often than it exploits it.
> 
> Either way, these four cases jumped out at me.
> 
> Not that the ordering of the first six cases for ccosh() and csinh() agree but they are then inconsistent with the 4th and 5th points of ctanh().
> 
> Do you agree that they need to be swapped?
> 
> I have a summary almost done but the above inconsistencies need to be addressed.
> 
> Thanks - Damian