[bug #68604] sort() is not stable for sparse arrays containing multiple NaNs

Please use the bug tracker to post updates to a bug report.  The mailing list is intended as a read-only notification stream.  Info posted to this mailing list address won't appear in the tracker database where it is most useful.

URL:
  <https://savannah.gnu.org/bugs/?68604>

                 Summary: sort() is not stable for sparse arrays containing
multiple NaNs
                   Group: GNU Octave
               Submitter: dasergatskov
               Submitted: Sat 08 Aug 2026 10:43:25 PM UTC
                Category: Octave Function
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Incorrect Result
                  Status: None
             Assigned to: None
         Originator Name:
        Originator Email:
             Open/Closed: Open
         Discussion Lock: Unlocked
                 Release: dev
        Operating System: Any
           Fixed Release: None
         Planned Release: None


    _______________________________________________________

Follow-up Comments:


-------------------------------------------------------
Date: Sat 08 Aug 2026 10:43:25 PM UTC By: Dmitri A. Sergatskov <dasergatskov>
Currently `sort()`  does not preserve the input order of multiple NaN entries
when sorting sparse arrays. Dense arrays preserve their order, but sparse
arrays reverse them. 
For example:

octave:1>  x = [NaN, NaN, 1]
x =

   NaN   NaN     1

octave:2> [y, d_idx] = sort (x)
y =

     1   NaN   NaN

d_idx =

   3   1   2

octave:3> [s_y, s_idx] = sort (sparse(x))
s_y =

Compressed Column Sparse (rows = 1, cols = 3, nnz = 3 [100%])

  (1, 1) -> 1
  (1, 2) -> NaN
  (1, 3) -> NaN

s_idx =

   3   2   1


This becomes observable in the complex case:

octave:9> x = complex ([NaN, NaN, 1], [1, 2, 0])
x =

   NaN +   1i   NaN +   2i     1 +   0i

octave:10> y = sort(x)
y =

     1 +   0i   NaN +   1i   NaN +   2i

octave:11> s_y =full(sort(sparse(x)))
s_y =

     1 +   0i   NaN +   2i   NaN +   1i



The problem is with Sparse-c/d code, not the sort itself. I should have a
patch soon.

Dmitri.
-- 








    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?68604>

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

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQQk97aszIMMAvLLwm6qLAuaBUf3TgUCanexEQAKCRCqLAuaBUf3
Tml5AP4gVsO27rhgP7IWVr8ws2t95ELhG45mnFxT86PxDDpXfwD/ZMoqIFjrvaqD
S9wfxkEjPBugef4/rpgdgmU8/Ou4wwk=
=LGP6
-----END PGP SIGNATURE-----