/* copyleft (c) 2003 forum::für::umläute -- IOhannes m zmölnig @ IEM * based on d_array.c from pd: * Copyright (c) 1997-1999 Miller Puckette and others. * For information on usage and redistribution, and for a DISCLAIMER OF ALL * WARRANTIES, see the file, "LICENSE.txt," in this distribution. */ /* tab16read, tab16read4, tab16write */ #include "iem16_table.h" /* ---------- tab16read4: control, non-interpolating ------------------------ */ static t_class *tab16read4_class; typedef struct _tab16read4{ t_object x_obj; t_symbol *x_arrayname; } t_tab16read4; static void tab16read4_float(t_tab16read4 *x, t_float f){ t_table16 *a; int npoints; t_iem16_16bit *vec; if (!(a = (t_table16 *)pd_findbyclass(x->x_arrayname, table16_class))) error("%s: no such array", x->x_arrayname->s_name); else if (!table16_getarray16(a, &npoints, &vec)) error("%s: bad template for tab16read4", x->x_arrayname->s_name); else if (npoints < 4) outlet_float(x->x_obj.ob_outlet, 0); else if (f <= 1) outlet_float(x->x_obj.ob_outlet, vec[1]); else if (f >= npoints - 2) outlet_float(x->x_obj.ob_outlet, vec[npoints - 2]); else { int n = f; float a, b, c, d, cminusb, frac; t_iem16_16bit *fp; if (n >= npoints - 2) n = npoints - 3; fp = vec + n; frac = f - n; a = fp[-1]; b = fp[0]; c = fp[1]; d = fp[2]; cminusb = c-b; outlet_float(x->x_obj.ob_outlet, b + frac * ( cminusb - 0.5f * (frac-1.) * ( (a - d + 3.0f * cminusb) * frac + (b - a - cminusb)))); } } static void tab16read4_set(t_tab16read4 *x, t_symbol *s){ x->x_arrayname = s; } static void *tab16read4_new(t_symbol *s){ t_tab16read4 *x = (t_tab16read4 *)pd_new(tab16read4_class); x->x_arrayname = s; outlet_new(&x->x_obj, gensym("float")); return (x); } void tab16read4_setup(void){ tab16read4_class = class_new(gensym("tab16read4"), (t_newmethod)tab16read4_new, 0, sizeof(t_tab16read4), 0, A_DEFSYM, 0); class_addfloat(tab16read4_class, (t_method)tab16read4_float); class_addmethod(tab16read4_class, (t_method)tab16read4_set, gensym("set"), A_SYMBOL, 0); }