Re: AW: Re: Large test regression ?

[John W. Eaton]

On 27-Feb-2009, Kai Habel wrote:

| address@hidden schrieb:
| > -- Ursprüngl. Mitteil. --
| > Betreff:    Re: Large test regression ?
| > Von:        Jaroslav Hajek <address@hidden>
| > Datum:              26.02.2009 19:55
| >
| > On Thu, Feb 26, 2009 at 6:02 PM, Kai Habel <address@hidden> wrote:
| >   
| >> I see currently test failures all over the place. Can anyone confirm
| >> this or is the problem on my side.
| >>
| >> I have :
| >>
| >> Summary:
| >>
| >>  PASS   5381
| >>  FAIL    112
| >>
| >> There was 1 expected failure (see fntests.log for details).
| >>
| >> For example:
| >>     
| >>>>>>> processing /home/kai/hg-octave/octave/src/DLD-FUNCTIONS/det.cc
| >>>>>>>               
| >>  ***** assert(det (single([1, 2; 3, 4])), single(-2), 10 * eps ('single'));
| >> !!!!! test failed
| >> invalid conversion from scalar to float scalar>>>>> processing
| >> /home/kai/hg-octave/octave/src/DLD-FUNCTIONS/dispatch.cc
| >>     
| >>>>>>> processing /home/kai/hg-octave/octave/src/DLD-FUNCTIONS/dlmread.cc
| >>>>>>>               
| >> This is with:
| >>
| >> changeset:   8877:2c8b2399247b
| >> tag:         tip
| >>
| >>
| >> Kai
| >>
| >>     
| >
| > Did you rerun configure prior to rebuilding? There was an new source
| > recently added to OPERATORS/, which seems as likely cause of your
| > problem.
| >
| > I am not at my PC at the moment. I am pretty sure that I did a complete 
rebuild including ./autogen.sh and ./configure after I noticed the problem the 
first time. I will try again tomorrow.
| > Kai
| >
| >   
| It seems something went wrong yesterday. I had to "reclone" the
| repository today. But I still get three  failures.
| 
| >>>>> processing /home/kai/hg-octave/octave/scripts/audio/wavwrite.m
|   ***** test
|  A = [1:10; 1:10]/10;
|  wavwrite("a.wav", A);
|  [B, samples_per_sec, bits_per_sample] = wavread("a.wav");
|  assert(A,B, 10^(-4));
|  assert(samples_per_sec, 8000);
|  assert(bits_per_sample, 16);
| !!!!! test failed
| A(I): Index exceeds matrix dimension.  ***** test
|  A=[1:10; 1:10] / 10;
|  wavwrite("a.wav", A, 4000);
|  [B, samples_per_sec, bits_per_sample] = wavread("a.wav");
|  assert(A,B, 10^(-4));
|  assert(samples_per_sec, 4000);
|  assert(bits_per_sample, 16);
| !!!!! test failed
| A(I): Index exceeds matrix dimension.>>>>> processing
| /home/kai/hg-octave/octave/scripts/deprecated/intersection.m
| ...
| >>>>> processing /home/kai/hg-octave/octave/scripts/general/interpn.m
|   ***** test
|  [x,y,z] = ndgrid(0:2);
|  f = x+y+z;
|  assert (interpn(x,y,z,f,[.5 1.5],[.5 1.5],[.5 1.5]), [1.5, 4.5])
|  assert (interpn(x,y,z,f,[.51 1.51],[.51 1.51],[.51 1.51],'nearest'),
| [3, 6])
|  assert (interpn(x,y,z,f,[.5 1.5],[.5 1.5],[.5 1.5],'spline'), [1.5, 4.5])
|  assert (interpn(x,y,z,f,x,y,z), f)
|  assert (interpn(x,y,z,f,x,y,z,'nearest'), f)
|  assert (interpn(x,y,z,f,x,y,z,'spline'), f)
| !!!!! test failed
| A(I): Index exceeds matrix dimension.>>>>> processing
| /home/kai/hg-octave/octave/scripts/general/isa.m
| >>>>> processing /home/kai/hg-octave/octave/scripts/general/isscalar.m
| 
| This is with:
| address@hidden:~/hg-octave/octave> hg tip
| changeset:   8883:7de0992eb123
| tag:         tip
| user:        Jaroslav Hajek <address@hidden>
| date:        Fri Feb 27 08:06:25 2009 +0100
| summary:     add nargin check to strsplit

That should be fixed by the next changeset in the archive, which,
although it is mostly a style fix for strsplit.m, also includes the
change

  --- a/src/symtab.h    Fri Feb 27 08:06:25 2009 +0100
  +++ b/src/symtab.h    Fri Feb 27 03:31:41 2009 -0500
  @@ -2064,6 +2064,8 @@

          sr.force_variable (context);
         }
  +    else
  +      p->second.force_variable (context);
     }

     octave_value& do_varref (const std::string& name, context_id context)


This really should have been part of my previous change

  changeset:   8881:8ed42c679af5
  user:        John W. Eaton <address@hidden>
  date:        Thu Feb 26 13:58:47 2009 -0500
  summary:     after defining a script or function, clear local variables 
created for parsing


but was left out because I forgot to do a qrefresh before committing
the change.

Sorry for the sloppiness,

jwe