Re: Some demos cannot work correct on cvs source (2010-05-08)

[Hans-Bernhard B. <HBB...@t-...>]

Tatsuro MATSUOKA wrote:
> However,
> 2010-05-08  Hans-Bernhard Broeker  <br...@ph...>
> 
> 	* src/eval.c (magnitude): Make implementation robust to overflows
> 	and underflows.
> 
> the above change seems to not cause the directly cause the bug.

Actually, it does.  The crucial clue is the second of the errors you 
found.  It turns out that
	
	print laplace(0, 0, 0.5)

works, but

	print laplace(0.0, 0.0, 0.5)

reports 'undefined value'.  My re-implementation of hypot() has a 
loophole for input value {0,0} that causes it to divide by zero.  I'll 
fix it ASAP.



> 
> I do not figure out what is wrong at present.
> 
> 
> Regards
> 
> Tatsuro
> 
> --------------------------------------
> GyaO! - Anime, Dramas, Movies, and Music videos [FREE]
> http://pr.mail.yahoo.co.jp/gyao/
> 
> ------------------------------------------------------------------------------
> 
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
>