binary模块的match 与 matches函数

import re import random #TODO : Change support for 0b to 0h inputs. with open('tb_systolic_array.sv', 'r') as file: content = file.read() """ A function to convert from float to binary """ def float_to_binary(num : float, precision: int = 23) -> str: whole, decimal = str(num).split(".") whole = int(whole) decimal = float(f"0.{decimal}") #Convert the whole number part into binary whole_bin = bin(whole).replace("0b", "") decimal_bin = [] while decimal and len(decimal_bin) < precision: decimal *= 2 bit = int(decimal) decimal_bin.append(str(bit)) decimal -= bit return whole_bin + "." + "".join(decimal_bin) """ Normalizing the floating point number """ def normalize(num : str) -> list: temp = num.split(".") temp = int(temp[0]) exp = -1 while temp != 0: temp //= 10 exp += 1 num = num.replace(".","") num = num[1:24] exp += 127 exp = bin(exp).replace("0b", "") return [num, exp] def float_to_fp32(num : float) -> int: sign_bit = 0 if (num < 0): sign_bit = 1 y = normalize(float_to_binary(num)) return int(str(sign_bit) + str(y[1]) + y[0]) if __name__ == "__main__": # Find all occurrences of in_a and in_b assignments matches = re.findall(r"in_a = 32'b[01]+; in_b = 32'b[01]+;", content) for match in matches: r1 = float_to_fp32(random.uniform(0, 100)) r2 = float_to_fp32(random.uniform(0, 50)) in_a_value, in_b_value = r1, r2 new_assignment = f"in_a = 32'b{in_a_value}; in_b = 32'b{in_b_value};" content = content.replace(match,new_assignment,1) # Write the modified content back to the file with open('tb_systolic_array_2.sv', 'w') as file: file.write(content)解释上述代码