fastest way to parse prefix codes in python

Question

I have a string composed of binary prefix codes (huffman coding), eg

00010001101110

where (for example):

a = "000",