It's written in my book that $g(x)$ is polynomial and if it's equal to zero at $a$, that is $ g(a) = 0 $ then we can write $g(x) = (x - a)^k\cdot g_1(x)$. Well, this proposition popped in my book without any description or derivation. Can anybody give me an intuition of what it means and derive it? Furthermore, I read this in the portion "Limits of rational functions". What this expression could mean in that context?
This follows from the Euclidean division of polynomials, which implies $$ g(x) = (x-a)q(x)+g(a) $$ Thus, $a$ is a root of $g$ iff $x-a$ divides $g$ and so $g(x)=(x-a)q(x)$.
Now repeat the argument with $q$. If $a$ is not a root of $q$, then we're done. If $a$ is a root of $q$, then $q(x)=(x-a)q_2(x)$ and so $g(x)=(x-a)^2q_2(x)$.
This process cannot go on forever because the degree of the quotient is decreasing. We end up with $g(x)=(x-a)^k q_k(x)$ with $q_k(a)\ne 0$.