I was wondering if others could review my proof the following integral converges and tell me if it's ok? W is the Lambert-W function.
$$-\int_0^{1}\frac{\text{W}[-\log(\sin(z))]}{\log[\sin(z)]}dz$$
A plot of the function from $(0,\pi)$ is shown below.
Proof:
Near zero, $\sin(z)\sim z$ so we can consider $$\int_0^{1}\frac{\text{W}[-\log(z)]}{\log[z]}dz$$ It suffices then to show $$\lim_{z\to 0^+} \frac{\text{W}(-log(z))}{log(z)}=0 $$ Since we have the indeterminant form $\frac{\infty}{\infty}$, applying L'Hopital's rule: $$ \frac{\frac{d}{dz}\text{W}(-log(z))}{\frac{d}{dz}\log(z)}=\frac{\text{W}(-\log(z))}{\log(z)(1+\text{W}(-\log(z))} $$ and this is where my logic is a bit unorthodox: We still have the indeterminant form $\frac{\infty}{\infty}$ however, the denominator approaches $\infty+\infty$ so the expression becomes much smaller than $1/z$ so that I may conclude the limit is zero.
Is this rigorous enough or is there a more appropriate way to prove the limit is zero?

I don't really see how you conclude the limit is zero from where you left off. There appears to be a lot of handwaving involved. Alternatively, multiply the integrand by $-1$. One can then see the integrand is then the inverse of $y=\arcsin(x^{1/x})$. Letting $y\to0^+$, one can see that $x^{1/x}=\exp(\ln(x)/x)\to0$, so that $\ln(x)/x\to-\infty$, which clearly happens as $x\to0^+$. So the limit exists and is zero.