Mercurial > octave
view test/build_bc_overloads_expected.m @ 17691:8a54a481ecb5
Massive speed improvement in sparse min/max functions for calls like
'max(s,[],2)' (bug #40268)
* dSparse.cc (SparseMatrix SparseMatrix::max (Array<octave_idx-type>&, int)) :
Remove triple loop by using a local buffer
(SparseMatrix SparseMatrix::min (Array<octave_idx-type>&, int)) : ditto
* CSparse.cc (SparseComplexMatrix SparseComplexMatrix::max (
Array<octave_idx-type>&, int)) : Remove triple loop by using a local buffer
(SparseComplexMatrix SparseComplexMatrix::min (Array<octave_idx-type>&,
int)) : ditto
author | David Bateman <dbateman@free.fr> |
---|---|
date | Fri, 18 Oct 2013 23:44:44 +0200 |
parents | b1283d4c06c2 |
children |
line wrap: on
line source
% this script is intended to be Matlab compatible % first, run the script % % ./build_bc_overloads_tests.sh overloads_only % % to generate the overloaded functions. % ex.double = 1; ex.single = single (1); ex.logical = true; ex.char = 'char'; ex.int8 = int8 (1); ex.int16 = int16 (1); ex.int32 = int32 (1); ex.int64 = int64 (1); ex.uint8 = uint8 (1); ex.uint16 = uint16 (1); ex.uint32 = uint32 (1); ex.uint64 = uint64 (1); ex.cell = {}; ex.struct = struct (); ex.function_handle = @numel; f = fieldnames (ex); n = numel (f); fid = fopen ('bc_overloads_expected','w'); fid for i = 1:n for j = 1:n s = tbcover (ex.(f{i}), ex.(f{j})); fprintf (fid, '%s %s %s\n', f{i}, f{j}, s); end end fclose (fid)