TOP consists of a language that defines equations to be solved (see equation files), a compiler wrapper designed to compile these equation files, and a python module to actually run computations.
Compiling an equation file is performed by the compiler wrapper top-build. Equation files are attached to a given star model, you have to provide top-build with the model corresponding to the equation file with the --model= option.
| Example: | top-build --model=poly_ester eq_poly_ester |
|---|
The options you can pass to top-build are:
| --latex=FILE | enable LaTeX output to file FILE |
| --order=FILE | use the file named FILE instead of the default one to manage the order of variable and equation. |
| --model=NAME | this option is mandatory, it tells TOP which stellar model to use. |
| --cplx | forces TOP to compile with the complex version. |
| --debug | enable debug mode for the file compiled. |
In order to use your newly compiled equation file, you need to use the top python module:
import top # imports the top module
import numpy as np # imports numpy
p = top.load('eq_poly_ester') # loads your compiled equation file
p.read_dati('dati') # reads the parameter file `dati`
model = 'model/' # path to the model
shift = p.dati.shift
m = p.init_model(model) # initializes the model stored in the directory
r = p.run_arncheb(shift) # runs Arnoldi-Chebyshev method
# get the solutions
for i in range(0, r.nsol): # for all solutions
for var in p.get_vars(0): # for each variable (in the fist domain)
# plot the solution
r.plot(0, i, v) # quick plot of the solution
# plot an expression of the solution:
# get the solution
val, vec, l = r.get_sol(0, i, v)
# get the grid
radius, theta = r.get_grid()
cost = np.cos(theta)
# get a field from the model
h = m['hh']
# project the solution onto the grid
gv = top.leg.eval2d(vec, cost, l[0], 2, r.dati['m'])
# actually plot the expression
r.plot_val(gv*np.sqrt(h)**r.dati['pindex'])
For a detailed description of all functionalities available through the python module, see python API.