Seite - 121 - in Programming for Computations – Python - A Gentle Introduction to Numerical Simulations with Python 3.6, Band Second Edition

5.4 MakingOurOwnModule 121 position = y(v_initial, t) # compute all positions (over T) plt.plot(t, position, ’b--’) plt.xlabel(’Time (s)’) plt.ylabel(’Vertical position (m)’) plt.show() return if __name__ == ’__main__’: application() The code in application represents the main program and should be under- standable from what we have learned previously. Note that, like we have done here, it is usually a good idea to print some information about how the program works. Testing Asasimple testof thecodeinvertical_motion.py,wemightcompare theoutput tohandcalculations,aswedidbefore. InChap.6,however,wewill learn howtesting ought to be donevia dedicated test functions.These test functionsmay berunindifferentways.Onealternative,however, is to includeanoptionwithin the testblock,whichallowstheuser torunthroughthe test functionswheneverwanted, butmoreabout that later. PlacingImport Statements in Our Module Note that ifwehaveimportstatements inourmodule, it ispossible toruninto troubleifwedonotplace themat the topof thefile (whichisaccordingto the general recommendation). With the following sketchy example module, it will work fine to import some_function in another program and use it (since, when importing some_function, the importofnumpy isdone). import numpy as np def some_function(n): a = np.zeros(n) ... return r def application(): .... n = 10 r = some_function(n) ... return if __name__ == ’__main__’: application() One choice that would not work in the same way, however, would be to instead have the import statement import numpy as np after if __name__ == ’__main__’: . Then, this import statement would not be run ifsome_function is imported foruse inanotherprogram.