[Octave-patch-tracker] [patch #8119] Allow variable tolerance and improve error messages for assert.m script

[Dan Sebald]

Follow-up Comment #8, patch #8119 (project octave):

I ran "make check" using the proposed assert.m and found the following:


  scripts/general/interpft.m .............................
ASSERT errors for:  assert (max (abs (imag (interpft ([1:8], 20)))),0,2 *
eps)

  Location  |  Expected  |  Observed  |  Reason
     1        2.2542e-15       0         Exceeds abs tol 4.4409e-16 by
1.8101e-15


ASSERT errors for:  assert (max (abs (imag (interpft ([1:8], 21)))),0,2 *
eps)

  Location  |  Expected  |  Observed  |  Reason
     1        3.0634e-15       0         Exceeds abs tol 4.4409e-16 by
2.6193e-15


ASSERT errors for:  assert (max (abs (imag (interpft ([1:9], 20)))),0,2 *
eps)

  Location  |  Expected  |  Observed  |  Reason
     1        3.8512e-15       0         Exceeds abs tol 4.4409e-16 by
3.4071e-15


ASSERT errors for:  assert (max (abs (imag (interpft ([1:9], 21)))),0,2 *
eps)

  Location  |  Expected  |  Observed  |  Reason
     1        4.6244e-15       0         Exceeds abs tol 4.4409e-16 by
4.1803e-15

 PASS   11/15   FAIL 4

...

  scripts/general/num2str.m ..............................
ASSERT errors for:  assert (num2str (1e23),"100000000000000000000000")

  Location  |  Expected  |  Observed  |  Reason
     []       100000000000000000000000   99999999999999991611392    Strings
don't match

 PASS   23/23  

(Not sure why this indicates all 23 pass if there was a failure.  Testing the
"contrapositive" perhaps.)

...

  scripts/plot/clf.m .....................................
ASSERT errors for:  assert (get (hf, "papertype"),"usletter")

  Location  |  Expected  |  Observed  |  Reason
     []        usletter     tabloid      Strings don't match

 PASS    3/3   

(Again, all 3 pass.)

...

  scripts/polynomial/residue.m ...........................
ASSERT errors for:  assert (br,b,1e-8)

  Location  |  Expected  |  Observed  |  Reason
     .            E            O         Dimensions don't match

 PASS    5/5   

(All pass)

...

Summary:

  PASS     11010
  FAIL         7
  XFAIL        4
  SKIPPED      9


So, the first occurrence has 4 fails.  The other three occurrences indicate no
fails, but the tally of summary statistics indicates 7 fials.  Seems odd.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/patch/?8119>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/