In the appendix A of this paper there is an integral that the author says can be solved using any good integral table. However I cannot seem to find it on any integral table (ex: gradshteyn and ryzhik) - although it could be that I am not using the table properly as it is the first time I am doing it. I have solved it using Wolfram but it does not give the same result as in the paper.
The integral is:
$$\int{\frac{-1}{1-F \cdot\text{sech}^2(x)}dx}$$
Note that the author gives the result for F < 1 which is what I want. Maybe that is why the wolfram solution is different, I am not sure.
Factor $-\dfrac1F$ outside the integral sign. Then, since $F<1$, it follows that the $\dfrac1F$ in the integrand's
denominator can be written as $1+a^2$. At the same time, similarly to $\sec^2x=1+\tan^2x$ which
we know to hold for its trigonometric counterpart, we also have $\text{sech}^2x=1-\tanh^2x$, making
our integral $\displaystyle-\frac1F\int\frac{dx}{a^2+\tanh^2x}$ . At this point we recognize $\arctan't=\dfrac1{1+t^2}$ . Substituting
$u=\dfrac{\tanh x}a$, we finally arrive at the desired result, $-I=x+\dfrac1a\arctan\dfrac{\tanh x}a$.