We define for $v \in L^1(\mathbb{R})$ $$ Var_{\mathbb{R}}(v) := \mathrm{sup}\left \lbrace\int_{\mathbb{R}} u(x)g'(x)~\mathrm{d}x: g \in C_0^\infty(\mathbb{R}),~ \lvert g(x) \rvert \leq 1\text{ for all }x \in \mathbb{R} \right \rbrace. $$ If $u \in L^1(\mathbb{R}) \cap C^1(\mathbb{R})$ and $Var_{\mathbb{R}}(u) < \infty$, is it true that $$ \int_{\mathbb{R}} \lvert v'(x) \rvert~\mathrm{d}x < \infty \quad ? $$
I am sure that it is, but I can not quite find a rigorous proof. I also know that this holds for bounded domains. I also know that $$ \int_{\mathbb{R}} v(x) g'(x)~\mathrm{d}x = - \int_{\mathbb{R}} v'(x) g(x)~\mathrm{d}x. $$ Maybe approximating $-\mathrm{sign}(u')$ could help, but I do not see how. Mabye Riesz Representation Theorem can be helpful...
I am grateful for help or a reference. I have found some posts on SE that could not really help me. Maybe it is not even true...