I have the following intuition, which I am not sure about if it is true or not, since I cannot come up with a proof or with a counterexample.
Given an $f \in L^1(\mathbb{R})$, if $$\int_\mathbb{R}f \cdot \phi = 0, \forall\phi \in C_c^{\infty}(\mathbb{R})$$ then we have $f = 0$ (in the sense of equivalence classes).
Does anyone know a proof of this? And if the integral evaluates to for example $\phi(1)$ does it still hold that $f=0$?
Thanks a lot in advance!
It is easy to prove with this condition that for all $[a,b]$ (it is easy to construct a sequence $\phi_n\rightarrow 1_{[a,b]} $, with $\phi_n$ smooth with compact support, by convolution and then use dominating theorem (since $f$ is in $L^1$)) $$ \int f1_{[a,b]} =0.$$ And then since the borelian $\sigma$-algebra is generated by $\{[a,b],\ a,b \in \mathbb{R} \}$ you will have $$ \int f 1_A =\int_A f =0,\quad \forall A \in \mathcal{B}(\mathbb{R})$$ which implies easily (take $A=\{f>0\}$ and then $A'=\{f<0\}$, you will have $\lambda(A)=\lambda(A')=0$, where $\lambda$ is the Lebesgue measure) the result.