1 files changed, 47 insertions, 0 deletions
diff --git a/mycobject.c b/mycobject.c
new file mode 100644
index 0000000..0020de5
--- /dev/null
+++ b/mycobject.c
@@ -0,0 +1,47 @@
+/* code for "mycobject" pd class.  This takes two messages: floating-point
+numbers, and "rats", and just prints something out for each message. */
+
+#include "m_pd.h"
+
+    /* the data structure for each copy of "mycobject".  In this case we
+    on;y need pd's obligatory header (of type t_object). */
+typedef struct mycobject
+{
+  t_object x_ob;
+} t_mycobject;
+
+    /* this is called back when mycobject gets a "float" message (i.e., a
+    number.) */
+void mycobject_float(t_mycobject *x, t_floatarg f)
+{
+    post("mycobject: %f", f);
+}
+
+    /* this is called when mycobject gets the message, "rats". */
+void mycobject_rats(t_mycobject *x)
+{
+    post("mycobject: rats");
+}
+
+    /* this is a pointer to the class for "mycobject", which is created in the
+    "setup" routine below and used to create new ones in the "new" routine. */
+t_class *mycobject_class;
+
+    /* this is called when a new "mycobject" object is created. */
+void *mycobject_new(void)
+{
+    t_mycobject *x = (t_mycobject *)pd_new(mycobject_class);
+    post("mycobject_new");
+    return (void *)x;
+}
+
+    /* this is called once at setup time, when this code is loaded into Pd. */
+void mycobject_setup(void)
+{
+    post("mycobject_setup");
+    mycobject_class = class_new(gensym("mycobject"), (t_newmethod)mycobject_new, 0,
+    	sizeof(t_mycobject), 0, 0);
+    class_addmethod(mycobject_class, (t_method)mycobject_rats, gensym("rats"), 0);
+    class_addfloat(mycobject_class, mycobject_float);
+}
+