[Octave-bug-tracker] [bug #65291] [octave forge] (mapping) degrees2dms produces incorrect results

[Philip Nienhuis]

Follow-up Comment #16, bug #65291 (group octave):

In a spell between other work I just tried this with your new degrees2dms.m:

>> degrees2dms ([1 : 0.6 : 5]')
ans =
    1.0000         0         0
    1.0000   36.0000    0.0000
    2.0000   12.0000    0.0000
    2.0000   47.0000   60.0000
    3.0000   23.0000   60.0000
    4.0000         0         0
    4.0000   35.0000   60.0000

where minute values still can be 60 and don't overflow to minutes.
Or,

>> format long
>> degrees2dms ([1 : 0.6 : 5]')
ans =
   1.000000000000000e+00                       0                       0
   1.000000000000000e+00   3.600000000000000e+01   4.263256414560601e-13
   2.000000000000000e+00   1.200000000000000e+01   6.394884621840902e-13
   2.000000000000000e+00   4.700000000000000e+01   5.999999999999915e+01
   3.000000000000000e+00   2.300000000000000e+01   5.999999999999957e+01
   4.000000000000000e+00                       0                       0
   4.000000000000000e+00   3.500000000000000e+01   5.999999999999872e+01

To have proper output, seconds >= 59.5 (rather than >= 60.0-eps) need to
overflow to minutes. IIRC that's also what I did for angl2str.m (patch
#9953).

Yet I wonder if we should hunt down this all the way or just leave it as-is.
It's not pretty but still mostly bug-for-bug compatible with Matlab (although
just a tiny bit better :-) see last-but-one row):

>> degrees2dms ([1 : 0.6 : 5])
Warning: Reshaping input into 7-by-1 column vector.  Output will be 7-by-3. 
> In degrees2dms (line 57) 
ans =
    1.0000         0         0
    1.0000   36.0000    0.0000
    2.0000   12.0000    0.0000
    2.0000   47.0000   60.0000
    3.0000   23.0000   60.0000
    3.0000   59.0000   60.0000
    4.0000   35.0000   60.0000

(where you can also see that Matlab warns about a wrong dimension of the input
vector while Octave doesn't.)



    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?65291>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/