complex.h math functions

[Bruce Evans]

On Sun, 30 Oct 2005, Steve Kargl wrote:

> On Fri, Oct 21, 2005 at 08:50:53PM +1000, Bruce Evans wrote:
>> On Sun, 16 Oct 2005, Steve Kargl wrote:
>>
>>> On Wed, Oct 12, 2005 at 05:23:28PM +1000, Bruce Evans wrote:
>>
>>> Hopefully, this new version is closer to KNF.
>>
>> It still has only 4 chars for all secondary indents
>
> Strange.  Hopefully fixed.

It's not :-).  E.g.:

% 	/* Handle the nearly-non-exceptional cases where x and y are finite. */
% 	if (ix < 0x7ff00000 && iy < 0x7ff00000) {
% 	    if ((iy | ly) == 0)
  	^^^^
% 		return (cpack(cosh(x), x * y));
% 	    return (cpack(cosh(x) * cos(y), sinh(x) * sin(y)));
% 	}

fdlibm actually mostly uses this style, with all indents 4 chars after the
first normal 1-tab one.  I don't want to use this for new code and the
fdlibm maintainers show signs of switching to normal indentation in
k_tan.c.

> I also removed the additional
> space in the comments after the *.  This, of course, will
> make a diff fairly large, so I've attached s_ccosh.c.
> ...
>> The current order is almost that in the standard.  This order is not
>> bad but it might not give the simplest or shortest classification.
>> In particular, I think it might be better to group by 0's before
>> grouping by Infs and NaNs.
>
> I've re-arrange the code to do x = 0 and then y = 0 cases
> after the nearly-non-exceptional case.  You may want to
> flip x = 0 and y = 0.  I did not do this because I did not
> want to mess up your explanation of choice of signs.

I'll merge with it and see if there is more to clean up.

Bruce