How to access grid coordinates when defining the evolution_rate of a customed PDE class?

[mynanshan]

This might be a dumb question, but what is the internal approach to get the values of the grid?

Specifically, when writing the member function evolution_rate of a customer PDE class, I need to use the values of the grid. Although I can just calculate the grid values in advance outside the PDE class, I think there should be a way to access the grid values from the argument state or anything internal.

[david-zwicker]

I'm a not entirely sure what you mean by "values of the grid", but you can access the grid object using state.grid and if you need direct access to the field values at the grid cells, you can access the numpy array state.data. Does this help?

[mynanshan]

Sorry that I have not made it clear. I was meant to say the coordinates of the grid. Suppose the pde is c(x). Then, if I got it right, state.data should be the value of c evaluated on the grid, but I want to use the discrete values of x.

[david-zwicker]

You can use state.grid.axes_coords, state.grid.cell_coords, or state.grid.coordinate_arrays; see the documentation for the exact details.