According to the answer in What are the consequences of Diffie Hellman problem in P?, it appears we believe Discrete Logarithm and Diffie-Hellman are equivalent problems.
We also know if Discrete logarithm is easy, the factoring is easy.
Can we conclude and say breaking Diffie-Hellman yields a polynomial time algorithm for integer factoring? What would such an algorithm look like?