Re: svds.m failures

[marco atzeri]

2011/1/30 Tatsuro MATSUOKA :

> For me svds.m test gave no error.
>
>  d:\usr\Tatsu\mingwhome\octaves\hg\octave-work\scripts/sparse\svds.m  PASS    
> 4/4
>
> I'm using the soure
>
> author  John W. Eaton <address@hidden>
> Sat Jan 29 13:05:18 2011 -0500 (79 minutes ago)
> changeset 12206 6212f303243e
> parent 12205    be90f9e4548e
>
> on default branch
>
> Tatsuro


thanks for the testing.
I found that it is related to the atlas library.

I rebuilt

$ hg tip
changeset:   12219:17d7834df7b4
tag:         tip
user:        Petr Mikulik <address@hidden>
date:        Sun Jan 30 03:20:29 2011 -0500
summary:     __fltk_ginput__.m: return keypress info


than tested both
 version 3.3.55
 version 3.5.0+

with lapack 3.2.2 both pass the tests.

test /pub/hg/octave/scripts/sparse/svds.m
PASSES 4 out of 4 tests

with atlas 3.9.33, instead I found
 version 3.3.55    PASSES 4 out of 4 tests
 version 3.5.0+    PASSES 3 out of 4 tests

>>>>> processing /pub/hg/octave/scripts/sparse/svds.m
  ***** test
 [u2,s2,v2,flag] = svds(A,k,0,opts);
 s2 = diag(s2);
 assert(flag,!1);
 assert(s2, s(k:-1:1), 1e-10);
!!!!! test failed
assert (s2,s (k:-1:1),1e-10) expected
   38.060
   38.034
   38.034
   38.015
   38.015
   38.004
   38.004
but got
   38.034
   38.034
   38.015
   38.015
   38.004
   38.004
Dimensions don't match


Marco