Re: Handling branch cuts for hypergeometric functions

Richard Fateman wrote:
> Robert Dodier wrote:
> 
>> Hello Ray, you wrote in part:
>>
>>
>>> Would it make sense to just ask the user if 0 < x < 1 and use
>>> 15.4.13, otherwise use 15.4.12?  This kind of implies to the
>>> user that x is a real number.
>>
>>
>>
>> I'm mostly against the use of interactive queries, as their
>> presence makes it difficult to write noninteractive programs
>> (e.g. batch files).
>>
> Since maxima already uses interactive queries, it would be
> more consistent to have more interactive queries.
> On the other hand, it is also consistent to have flags
> of some sort that make default assumption.  So my suggestion
> is,  if it is not too much trouble, to have a flag to set
> that controls this behavior.

Control which behavior?  Whether maxima asks?

I'm not sure what to do.  The various expressions are valid for 
different parts of the plane.  Some are for x < 0; others for 0 < x < 1. 
  How is maxima to know which one to use without asking?  Right now, I 
think maxima is using the one that is valid for the whole plane, which 
oftentimes excludes the part of the real line where real numbers would 
result.

Ray