sampSize: Parameter esf and allocation ratio

[loeffju]

In the (still experimental?) function sampSize(), the parameter esf is not documented. The function calls the internal function sampSizeCore(), which has parameters for allocation ratio (alRatio) and type of desired output (Ntype), which is "total" or "arm". However, these parameters are currently not used by sampSize(). If I am not mistaken, the obtained result is the total sample size. Furthermore, the undocumented parameter esf seems to be a function of the allocation ratio, which can be observed by using the respective option in the GUI and checking the resulting code. It would be nice if the sampSize() function offered the possibility to 1) specify the allocation ratio directly, 2) specify if the output should give the sample size per arm or in total.
I am not sure how cumbersome it would be to modify the function. I had originally made a suggestion, but in the meanwhile I realized that that would only cover the case with 2 treatment arms. For cases with >2 arms, we would also need to input the number of arms as well as the information on which arms contribute to which node. For my personal use, I have written such a function, taking the basic algorithm from the Java code used in the GUI.