Rudin has gone through complex scenarios to prove this theorem, but isn't the following correct?
By contradiction, let $\exists p\in \mathbb{Q}; p^2=2$. Take $p=a/b$ where $a,b\in \mathbb{Z}$ and $(a,b)=1$. So, $a^2/b^2=2\Rightarrow a^2=2b^2\Rightarrow a=\sqrt{2}b$, yielding $\sqrt{2}b\in \mathbb{R}\backslash\mathbb{Q}\Rightarrow a\in\mathbb{R}\backslash\mathbb{Q}$, contradicting $a\in \mathbb{Z}$. $\square$