error in testsuite after cvs update



> From: Andreas Eder
> 
> Raymond Toy wrote (on Mon, 23 May 2005 at 11:18 -0400):
> 
> > >>>>> "Andreas" == Andreas Eder  writes:
> > 
> >     Andreas> With the new code in cvs there is no more 
> error thrown, but the
> >     Andreas> teststuite still is not happy and give sthe 
> following output:
> > 
> >     Andreas> Running tests in rtest14.mac:
> >     Andreas> ********************** Problem 111 ***************
> >     Andreas> Input:
> >     Andreas> closeto(AS_10_4_10(%i - 1), 1.0e-15)
> > 
> > 
> >     Andreas> Result:
> >     Andreas>                                                
>                         1
> >     Andreas> airy_ai(%i - 1) airy_dbi(%i - 1) - airy_bi(%i 
> - 1) airy_dai(%i - 1) - ---
> >     Andreas>                                                
>                        %pi
> > 
> >     Andreas> This differed from the expected result:
> >     Andreas> true
> > 
> > 
> >     Andreas> ********************** Problem 113 ***************
> >     Andreas> Input:
> >     Andreas> closeto(AS_10_4_10(- %i - 1), 1.0e-15)
> > 
> > 
> >     Andreas> Result:
> >     Andreas> airy_ai(- %i - 1) airy_dbi(- %i - 1) - 
> airy_bi(- %i - 1) airy_dai(- %i - 1)
> >     Andreas>                                                
>                               1
> >     Andreas>                                                
>                            - ---
> >     Andreas>                                                
>                              %pi
> > 
> > I didn't have this problem when I ran the test with cmucl.
> > What lisp are you using?  
> > 
> > Can you evaluate the results to see if close they really 
> are to zero?
> > I'm guessing 1e-15 is too tight.
> > 
> > Ray
> 
> I get these errors with cmucl (CMU Common Lisp Snapshot 
> 2005-04, running on elgin
> With core: /usr/local/lib/cmucl/lib/lisp.core
> Dumped on: Tue, 2005-04-05 16:17:40+02:00 on lorien
> See <http://www.cons.org/cmucl/>; for support information.
> Loaded subsystems:
>     Python 1.1, target Intel x86
>     CLOS based on Gerd's PCL 2004/04/14 03:32:47
> )
> 
> and sbcl (This is SBCL 0.9.0, an implementation of ANSI Common Lisp.
> )
> 
> gcl and clisp work fine.
> 
> What do you want me to test?
> 
> AS_10_4_10(%i - 1),numer; gives 9.99200722162641e-16 %i - 
> 3.33066907387547e-16
> and
>  AS_10_4_10(- %i - 1),numer; gives - 9.99200722162641e-16 %i 
> - 3.33066907387547e-16
> 
> in sbcl.
> 
> AS_10_4_10(%i - 1),numer; gives  8.88178419700125e-16 %i - 
> 4.44089209850063e-16
> and
> AS_10_4_10(- %i - 1),numer; gives - 8.88178419700125e-16 %i - 
> 4.44089209850063e-16
> in cmucl.
> 
> Yes, they are both very close to 0 but obviously bot sufficiently
> close.
> 

I will increase the tolerance for this case.   The existing values
were set by trial and error, with a small safety factor added.

I will also change the comparison function to report the numerical value.

	David


NOTICE
This e-mail and any attachments are private and confidential and may contain privileged information. If you are not an authorised recipient, the copying or distribution of this e-mail and any attachments is prohibited and you must not read, print or act in reliance on this e-mail or attachments.
This notice should not be removed.