aboutsummaryrefslogtreecommitdiff
path: root/src/mtx_mul.c
diff options
context:
space:
mode:
authorIOhannes m zmölnig <zmoelnig@users.sourceforge.net>2005-05-11 13:05:29 +0000
committerIOhannes m zmölnig <zmoelnig@users.sourceforge.net>2005-05-11 13:05:29 +0000
commitc74f9041a42d95190b88990e91c2587f2f57a056 (patch)
tree5679242a76c8f8a0fc59fcc044c97c01b6713471 /src/mtx_mul.c
parentf4299cb7e6a5e161aa16670acd33de1ef75fae4d (diff)
split the objects from mtx_binops into several files: mtx_add, mtx_sub, mtx_mul, mtx_pow
mtx_binops is still there for glue functions exposed some important operations on float-arrays via iemmatrix.h (mtx_doInvert, mtx_doTranspose, mtx_doMultiply) as you can easily see, they all start with "mtx_do" and a capital letter; they all return a pointer to (newly allocated) memory with the result svn path=/trunk/externals/iem/iemmatrix/; revision=2947
Diffstat (limited to 'src/mtx_mul.c')
-rw-r--r--src/mtx_mul.c361
1 files changed, 361 insertions, 0 deletions
diff --git a/src/mtx_mul.c b/src/mtx_mul.c
new file mode 100644
index 0000000..ef0f12b
--- /dev/null
+++ b/src/mtx_mul.c
@@ -0,0 +1,361 @@
+/*
+ * iemmatrix
+ *
+ * objects for manipulating simple matrices
+ * mostly refering to matlab/octave matrix functions
+ *
+ * Copyright (c) IOhannes m zmlnig, forum::fr::umlute
+ * IEM, Graz, Austria
+ *
+ * For information on usage and redistribution, and for a DISCLAIMER OF ALL
+ * WARRANTIES, see the file, "LICENSE.txt," in this distribution.
+ *
+ */
+#include "iemmatrix.h"
+
+/*
+ mtx_mul
+ mtx_*
+ mtx_.*
+ mtx_./
+
+ matrix multiplication
+
+*/
+
+t_matrixfloat*mtx_doMultiply(int rowA, t_matrixfloat*A, int colArowB, t_matrixfloat*B, int colB){
+ t_matrixfloat*result=(t_matrixfloat*)getbytes(sizeof(t_matrixfloat)*rowA*colB);
+ int r, c, n;
+ for(r=0; r<rowA; r++){
+ for(c=0; c<colB; c++){
+ t_matrixfloat sum=0.f;
+ for(n=0;n<colArowB; n++)
+ sum+=A[colArowB*r+n]*B[colB*n+c];
+ }
+ }
+ return result;
+}
+
+/* mtx_mul */
+static t_class *mtx_mul_class, *mtx_mulelement_class, *mtx_mulscalar_class;
+
+static void mtx_mul_matrix(t_mtx_binmtx *x, t_symbol *s, int argc, t_atom *argv)
+{
+ t_matrix *m=&x->m, *m2=&x->m2;
+ t_atom *ap, *ap1=argv+2, *ap2=m2->atombuffer+2;
+ int row=atom_getfloat(argv), col=atom_getfloat(argv+1);
+ int row2, col2, n, r, c;
+
+ if (!m2->atombuffer){ post("mulitply with what ?"); return; }
+ if (argc<2){ post("mtx_mul: crippled matrix"); return; }
+ if ((col<1)||(row<1)){post("mtx_mul: invalid dimensions"); return; }
+ if (col*row>argc-2){ post("sparse matrix not yet supported : use \"mtx_check\""); return; }
+
+ row2=atom_getfloat(m2->atombuffer);
+ col2=atom_getfloat(m2->atombuffer+1);
+
+ if (col!=row2) { post("mtx_mul: matrix dimensions do not match !"); return; }
+
+ adjustsize(m, row, col2);
+ ap=m->atombuffer+2;
+
+ for(r=0;r<row;r++)
+ for(c=0;c<col2;c++) {
+ t_matrixfloat sum = 0.f;
+ for(n=0;n<col;n++)sum+=(t_matrixfloat)atom_getfloat(ap1+col*r+n)*atom_getfloat(ap2+col2*n+c);
+ SETFLOAT(ap+col2*r+c,sum);
+ }
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), m->row*m->col+2, m->atombuffer);
+}
+
+static void mtx_mul_float(t_mtx_binmtx *x, t_float f)
+{
+ t_matrix *m=&x->m, *m2=&x->m2;
+ t_atom *ap, *ap2=m2->atombuffer+2;
+ int row2, col2, n;
+
+ if (!m2->atombuffer){ post("mulitply with what ?"); return; }
+
+ row2=atom_getfloat(m2->atombuffer);
+ col2=atom_getfloat(m2->atombuffer+1);
+ adjustsize(m, row2, col2);
+ ap=m->atombuffer+2;
+
+ n=row2*col2;
+
+ while(n--){
+ SETFLOAT(ap, f*atom_getfloat(ap2++));
+ ap++;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), m->row*m->col+2, m->atombuffer);
+}
+
+static void mtx_mulelement_matrix(t_mtx_binmtx *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int row=atom_getfloat(argv++);
+ int col=atom_getfloat(argv++);
+ t_atom *m;
+ t_atom *m2 = x->m2.atombuffer+2;
+ int n = argc-2;
+
+ if (argc<2){ post("mtx_mul: crippled matrix"); return; }
+ if ((col<1)||(row<1)) { post("mtx_mul: invalid dimensions"); return; }
+ if (col*row>argc-2){ post("sparse matrix not yet supported : use \"mtx_check\""); return; }
+ if (!(x->m2.col*x->m2.row)) {
+ adjustsize(&x->m, row, col);
+ matrix_set(&x->m, 0);
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+ return;
+ }
+ if ((col!=x->m2.col)||(row!=x->m2.row)){ post("matrix dimension do not match"); /* LATER SOLVE THIS */ return; }
+
+ adjustsize(&x->m, row, col);
+ m = x->m.atombuffer+2;
+
+ while(n--){
+ t_float f = atom_getfloat(argv++)*atom_getfloat(m2++);
+ SETFLOAT(m, f);
+ m++;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+}
+
+static void mtx_mulscalar_matrix(t_mtx_binscalar *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int n=argc-2;
+ t_atom *m;
+ t_float factor = x->f;
+ int row=atom_getfloat(argv++);
+ int col=atom_getfloat(argv++);
+
+ if (argc<2){
+ post("mtx_mul: crippled matrix");
+ return;
+ }
+ adjustsize(&x->m, row, col);
+ m = x->m.atombuffer+2;
+
+ while(n--){
+ m->a_type = A_FLOAT;
+ (m++)->a_w.w_float = atom_getfloat(argv++)*factor;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+}
+static void mtx_mulscalar_list(t_mtx_binscalar *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int n=argc;
+ t_atom *m;
+ t_float factor = x->f;
+ adjustsize(&x->m, 1, argc);
+ m = x->m.atombuffer;
+
+ while(n--){
+ m->a_type = A_FLOAT;
+ (m++)->a_w.w_float = atom_getfloat(argv++)*factor;
+ }
+ outlet_list(x->x_obj.ob_outlet, gensym("list"), argc, x->m.atombuffer);
+}
+
+static void *mtx_mul_new(t_symbol *s, int argc, t_atom *argv)
+{
+ if (argc>1) post("mtx_mul : extra arguments ignored");
+ if (argc) {
+ t_mtx_binscalar *x = (t_mtx_binscalar *)pd_new(mtx_mulscalar_class);
+ floatinlet_new(&x->x_obj, &x->f);
+ x->f = atom_getfloatarg(0, argc, argv);
+ outlet_new(&x->x_obj, 0);
+ return(x);
+ } else {
+ if (s->s_name[4]=='.') {
+ /* element mul */
+
+ t_matrix *x = (t_matrix *)pd_new(mtx_mulelement_class);
+ inlet_new(&x->x_obj, &x->x_obj.ob_pd, gensym("matrix"), gensym(""));
+ outlet_new(&x->x_obj, 0);
+ x->col = x->row = 0;
+ x->atombuffer = 0;
+ return(x);
+ } else {
+ t_mtx_binmtx *x = (t_mtx_binmtx *)pd_new(mtx_mul_class);
+ inlet_new(&x->x_obj, &x->x_obj.ob_pd, gensym("matrix"), gensym(""));
+ outlet_new(&x->x_obj, 0);
+ x->m.col = x->m.row = x->m2.col = x->m2.row = 0;
+ x->m.atombuffer = x->m2.atombuffer = 0;
+ return (x);
+ }
+ }
+}
+
+void mtx_mul_setup(void)
+{
+ mtx_mul_class = class_new(gensym("mtx_mul"), (t_newmethod)mtx_mul_new, (t_method)mtx_binmtx_free,
+ sizeof(t_mtx_binmtx), 0, A_GIMME, 0);
+ class_addcreator((t_newmethod)mtx_mul_new, gensym("mtx_*"), A_GIMME,0);
+ class_addmethod(mtx_mul_class, (t_method)mtx_mul_matrix, gensym("matrix"), A_GIMME, 0);
+ class_addmethod(mtx_mul_class, (t_method)mtx_bin_matrix2, gensym(""), A_GIMME, 0);
+ class_addfloat (mtx_mul_class, mtx_mul_float);
+ class_addbang (mtx_mul_class, mtx_binmtx_bang);
+
+ mtx_mulelement_class = class_new(gensym("mtx_.*"), (t_newmethod)mtx_mul_new, (t_method)mtx_binmtx_free,
+ sizeof(t_mtx_binmtx), 0, A_GIMME, 0);
+ class_addmethod(mtx_mulelement_class, (t_method)mtx_mulelement_matrix, gensym("matrix"), A_GIMME, 0);
+ class_addmethod(mtx_mulelement_class, (t_method)mtx_bin_matrix2, gensym(""), A_GIMME, 0);
+ class_addfloat (mtx_mulelement_class, mtx_mul_float);
+ class_addbang (mtx_mulelement_class, mtx_binmtx_bang);
+
+ mtx_mulscalar_class = class_new(gensym("mtx_mul"), 0, (t_method)mtx_binscalar_free,
+ sizeof(t_mtx_binscalar), 0, 0);
+ class_addcreator(0, gensym("mtx_*"), 0, 0);
+ class_addcreator(0, gensym("mtx_.*"), 0, 0);
+ class_addmethod(mtx_mulscalar_class, (t_method)mtx_mulscalar_matrix, gensym("matrix"), A_GIMME, 0);
+ class_addlist (mtx_mulscalar_class, mtx_mulscalar_list);
+ class_addbang (mtx_mulscalar_class, mtx_binscalar_bang);
+
+ class_sethelpsymbol(mtx_mul_class, gensym("iemmatrix/mtx_binops"));
+ class_sethelpsymbol(mtx_mulelement_class, gensym("iemmatrix/mtx_binops"));
+ class_sethelpsymbol(mtx_mulscalar_class, gensym("iemmatrix/mtx_binops"));
+}
+
+
+/* mtx_div */
+static t_class *mtx_divelement_class, *mtx_divscalar_class;
+
+static void mtx_divelement_matrix(t_mtx_binmtx *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int row=atom_getfloat(argv++);
+ int col=atom_getfloat(argv++);
+ t_atom *m;
+ t_atom *m2 = x->m2.atombuffer+2;
+ int n = argc-2;
+
+ if (argc<2){ post("mtx_div: crippled matrix"); return; }
+ if ((col<1)||(row<1)) { post("mtx_div: invalid dimensions"); return; }
+ if (col*row>argc-2){ post("sparse matrix not yet supported : use \"mtx_check\""); return; }
+ if (!(x->m2.col*x->m2.row)) {
+ adjustsize(&x->m, row, col);
+ matrix_set(&x->m, 0);
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+ return;
+ }
+ if ((col!=x->m2.col)||(row!=x->m2.row)){ post("matrix dimension do not match"); /* LATER SOLVE THIS */ return; }
+
+ adjustsize(&x->m, row, col);
+ m = x->m.atombuffer+2;
+
+ while(n--){
+ t_float f = atom_getfloat(argv++)/atom_getfloat(m2++);
+ SETFLOAT(m, f);
+ m++;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+}
+static void mtx_divelement_float(t_mtx_binmtx *x, t_float f)
+{
+ t_matrix *m=&x->m, *m2=&x->m2;
+ t_atom *ap, *ap2=m2->atombuffer+2;
+ int row2, col2, n;
+
+ if (!m2->atombuffer){ post("divide by what ?"); return; }
+
+ row2=atom_getfloat(m2->atombuffer);
+ col2=atom_getfloat(m2->atombuffer+1);
+ adjustsize(m, row2, col2);
+ ap=m->atombuffer+2;
+
+ n=row2*col2;
+
+ while(n--){
+ SETFLOAT(ap, f/atom_getfloat(ap2++));
+ ap++;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), m->row*m->col+2, m->atombuffer);
+}
+static void mtx_divscalar_matrix(t_mtx_binscalar *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int n=argc-2;
+ t_atom *m;
+ t_float factor = 1.0/x->f;
+ int row=atom_getfloat(argv++);
+ int col=atom_getfloat(argv++);
+
+ if (argc<2){
+ post("mtx_div: crippled matrix");
+ return;
+ }
+ adjustsize(&x->m, row, col);
+ m = x->m.atombuffer+2;
+
+ while(n--){
+ m->a_type = A_FLOAT;
+ (m++)->a_w.w_float = atom_getfloat(argv++)*factor;
+ }
+
+ outlet_anything(x->x_obj.ob_outlet, gensym("matrix"), argc, x->m.atombuffer);
+}
+static void mtx_divscalar_list(t_mtx_binscalar *x, t_symbol *s, int argc, t_atom *argv)
+{
+ int n=argc;
+ t_atom *m;
+ t_float factor = 1.0/x->f;
+
+ adjustsize(&x->m, 1, argc);
+ m = x->m.atombuffer;
+
+ while(n--){
+ m->a_type = A_FLOAT;
+ (m++)->a_w.w_float = atom_getfloat(argv++)*factor;
+ }
+
+ outlet_list(x->x_obj.ob_outlet, gensym("list"), argc, x->m.atombuffer);
+}
+
+static void *mtx_div_new(t_symbol *s, int argc, t_atom *argv)
+{
+ if (argc>1) post("mtx_div : extra arguments ignored");
+ if (argc) {
+ /* scalar division */
+ t_mtx_binscalar *x = (t_mtx_binscalar *)pd_new(mtx_divscalar_class);
+ floatinlet_new(&x->x_obj, &x->f);
+ x->f = atom_getfloatarg(0, argc, argv);
+ outlet_new(&x->x_obj, 0);
+ return(x);
+ } else {
+ /* element division */
+ t_matrix *x = (t_matrix *)pd_new(mtx_divelement_class);
+ inlet_new(&x->x_obj, &x->x_obj.ob_pd, gensym("matrix"), gensym(""));
+ outlet_new(&x->x_obj, 0);
+ x->col = x->row = 0;
+ x->atombuffer = 0;
+ return(x);
+ }
+}
+
+void mtx_div_setup(void)
+{
+ mtx_divelement_class = class_new(gensym("mtx_./"), (t_newmethod)mtx_div_new, (t_method)mtx_binmtx_free,
+ sizeof(t_mtx_binmtx), 0, A_GIMME, 0);
+ class_addmethod(mtx_divelement_class, (t_method)mtx_divelement_matrix, gensym("matrix"), A_GIMME, 0);
+ class_addmethod(mtx_divelement_class, (t_method)mtx_bin_matrix2, gensym(""), A_GIMME, 0);
+ class_addfloat (mtx_divelement_class, mtx_divelement_float);
+ class_addbang (mtx_divelement_class, mtx_binmtx_bang);
+
+ mtx_divscalar_class = class_new(gensym("mtx_./"), 0, (t_method)mtx_binscalar_free,
+ sizeof(t_mtx_binscalar), 0, 0);
+ class_addmethod(mtx_divscalar_class, (t_method)mtx_divscalar_matrix, gensym("matrix"), A_GIMME, 0);
+ class_addlist (mtx_divscalar_class, mtx_divscalar_list);
+ class_addbang (mtx_divscalar_class, mtx_binscalar_bang);
+
+ class_sethelpsymbol(mtx_divelement_class, gensym("iemmatrix/mtx_binops"));
+ class_sethelpsymbol(mtx_divscalar_class, gensym("iemmatrix/mtx_binops"));
+}
+
+void iemtx_mul_setup(void)
+{
+ mtx_mul_setup();
+ mtx_div_setup();
+}