Folland claims in Lemma 8.25
If $f, g \in L^1(\mathbf{R}^d)$, then $\int f \hat g = \int \hat f g$.
Note that by the $\hat\cdot$ notation, Folland means the Fourier transform.
My question is why is this true? The proof (in Folland) appears to be just an application of Fubini Theorem. But why is it obvious that the integrand: $$ f(x) g(\theta) e^{-2\pi i \langle x, \theta \rangle} $$ is in $L^1(\mathbf{R}^d \times \mathbf{R}^d)$ if $f, g \in L^1(\mathbf{R}^d)$?
First, note that $|f \widehat{g}| \leq |f| {\cdot} |\widehat{g}|_\infty \leq |f| {\cdot} \|g\|_{L^1}$ and similiar $|g\widehat{f}| \leq |g| \|f\|_{L^1}$. Thus, both integrals exist (in the sense of Lebesgue-integrals as usual). Now, observe that $|f(x) g(\theta) e^{- 2\pi \mathrm{i}\langle x, \theta \rangle}| \leq |f(x) g(\theta)|$. And because of the theorem of Fubini (resp. Tonelli), we know that $\int \int |f(x)| {\cdot} |g(\theta)| \, \rm{d}x \, \rm{d} \theta =\|f\|_{L^1} \|g\|_{L^1} < \infty$, i.e. your function is integrable.