Midterm week and next steps

Hello! This is my final post before the end of midterm evaluation and I would like to show what I've done during these days and which are the next steps. After the previous post I wrote some of the options that the user can provide using odeset. I wrote a function to allow the user to pass the modified Jacobian through a function that returns DF/DY and DF/DYP or through a cell array (if the Jacobian is constant) and I completed options MaxStep, InitialStep, MaxOrder (commit ea1a311). I started with mentors to write function decic, which computes initial conditions of the problem F(t,x,x') considering some values of x and x' fixed by the user. This function will be submitted as a patch in the next weeks. I tried to build Octave with KLU module included (it is an optional part of IDA which depends on SuiteSparse) but I had some problems with CMake that I hope to solve during the next days. KLU module will be used to solve sparse problems in ode15i, when a sparse Jacobian is supplied. Once I will be able to write the "sparse part" of ode15i, I will show some efficiency tests between my .oct file, Sundials' MEX and C file and ode15i in Matlab. After completing this part, I will try to improve the quality of the code: - writing a class whose methods will be able to call the function supplied by the user without global pointers and perform more efficient data conversion (avoiding loops) - Refactoring the code, putting all the checks in an other function For the examples I provided in the previous post, consider commit 4f60e96.