[Octave-bug-tracker] [bug #56933] linspace not Matlab-compatible for edge cases with -Inf or Inf as endpoint

[Rik]

URL:
  <https://savannah.gnu.org/bugs/?56933>

                 Summary: linspace not Matlab-compatible for edge cases with
-Inf or Inf as endpoint
                 Project: GNU Octave
            Submitted by: rik5
            Submitted on: Sun 22 Sep 2019 03:51:28 PM PDT
                Category: Octave Function
                Severity: 2 - Minor
                Priority: 5 - Normal
              Item Group: Matlab Compatibility
                  Status: Wont Fix
             Assigned to: None
         Originator Name: 
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: dev
        Operating System: Any

    _______________________________________________________

Details:

Matlab example #1


linspace (-Inf, Inf, 5)
ans = [-Inf, -Inf, 0, Inf, Inf]


Octave produces


ans = [-Inf, NaN, 0, NaN, Inf]


Matlab example #2


linspace (-Inf, 0, 4)
ans = [-Inf, NaN, NaN, 0]


Octave produces


ans = [-Inf, NaN, -Info, 0]


I am merely recording this.  I don't think it is necessary to have special
cases in the code in liboctave/array/xRowVector.cc to deal with this so I am
marking this bug as "Won't Fix".  But it will exist in case anyone searches
for this.





    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?56933>

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