In the answer to my question here, it is asserted that when $f:[0, \infty)\to\mathbb{R}$ is a function defined as $f(x) = \sqrt{x}$, it is valid & precise to write: $$\lim_{x\to 0}f(x) = 0$$ because in the answer to my question here, $\lim_{x\to c}f(x) = L$ only requires that $c$ be a limit point. This is in agreement with Rudin's definition of $\lim_{x\to p}f(x) = q$.
Now when discussing discontinuity, Rudin asserts that $\lim_{t\to x}f(t)$ exists if and only if $f(x+) = f(x-) = \lim_{t\to x}f(t)$. But, as far as I can understand from Rudin's definition of $f(x-)$, which is the left-hand limit, when $f(x) = \sqrt{x}$, $f(0-)$ does not exist, and therefore, $\lim_{t\to 0}f(t)$ does not exist as well. But, this contradicts the previous paragraph that says that $\lim_{t\to 0} \sqrt{x}$ exists because $0$ is a limit point.
What do I miss?