Hello Karol,

yes, I know padding. In my case I would like to set the numeric parameters with two variables first.

• Take the R11.3 into two variables f1 and f2 (f1 = 11, f2 = 3)
• Set the numeric(:f1,:f2) without execute statements

Now I have it realized as following:

  c_anz = substring(t_str from 2 for (position( '.' in t_str)-2)); -- precision (t_str is ’11.3’ a string)

  select value_int from p_u_isnumeric(:c_anz) into :i_anz; -- check for integer and convert into an int

  c_anz2 = substring(t_str from (position('.' in t_str)+1) for 5); -- scale

  select value_int from p_u_isnumeric(:c_anz2) into :i_anz2;

  tmp_int = cast(z_str as double precision) * power(10,i_anz2); -- set the value, round it –z_str = double precision, the value to convert into a string (2.4 into 00000000002.400)

  tmp_str =  lpad(cast(tmp_int as varchar(15)),i_anz + i_anz2,'0');

  n_str = o_str || left(tmp_str,i_anz) || '.' || right(tmp_str,i_anz2) || '#'; -- set the decimalpoint

  suspend;