solve running forever

[AntonioJCosta]

from fractions import Fraction
from pprint import pprint
from sympy import *
import numpy as np

x = symbols('x')

g = 9.81; y0 =100; v0 = 55; m = 80; c = 15
y = y0 + (m / c) * (v0 + ((m * g) / c)) * (1 - exp(- (c / m) * x)) - ((m * g) / c) * x
print('Equation:')
print(y)

Output:

Equation:
-52.32x + 672.373333333333 - 572.373333333333exp(-0.1875*x)

When I try to solve this using solve, the program runs forever and don't stop, without any error message.

solve(y)

[praneethratna]

solve() is used to algebraically solve equations and systems of equations , that is solving for a particular symbol in the equation where the first argument is an equation equaled to 0 and second argument is the symbol that we wanted to solve for.

>>> solve(x**2 - 1, x)
[-1, 1]

Here you wanted to substitute values in y if I am right for which you can use .subs() with list of tuples of values to be substituted in the equation as argument for the required output.

>>> g = 9.81; y0 =100; v0 = 55; m = 80; c = 15
>>> y = y0 + (m / c) * (v0 + ((m * g) / c)) * (1 - exp(- (c / m) * x)) - ((m * g) / c) * x
>>> y.subs([(g,9.81),(y0,100),(v0,55),(m,80),(c,15)])
-52.32*x + 672.373333333333 - 572.373333333333*exp(-0.1875*x)

[ThePauliPrinciple]

solve() is used to algebraically solve equations and systems of equations , that is solving for a particular symbol in the equation where the first argument is an equation equaled to 0 and second argument is the symbol that we wanted to solve for.

>>> solve(x**2 - 1, x)
[-1, 1]

Here you wanted to substitute values in y if I am right for which you can use .subs() with list of tuples of values to be substituted in the equation as argument for the required output.

>>> g = 9.81; y0 =100; v0 = 55; m = 80; c = 15
>>> y = y0 + (m / c) * (v0 + ((m * g) / c)) * (1 - exp(- (c / m) * x)) - ((m * g) / c) * x
>>> y.subs([(g,9.81),(y0,100),(v0,55),(m,80),(c,15)])
-52.32*x + 672.373333333333 - 572.373333333333*exp(-0.1875*x)

subs here does nothing, since g, y0,v0,m and c are not symbols, just python objects of ints and floats.
It would be nice if the OP could describe in words what is attempted.

I would agree with the OP though that a solution should be provided, the question asked to sympy was:
solve -52.32x + 672.373333333333 - 572.373333333333exp(-0.1875*x)=0, which, given that one equation and one unknown was given should be solvable.

Even when specifying to solve for x explicitly:
solve(y,x) there seems to be no solution found, nor the algorithm exiting and telling that no solution was found, so I would think this is a bug of some sort.

When we take the exponential term to the rhs and plot both sides of the equation we can find that a solution should exist:

[praneethratna]

Even when specifying to solve for x explicitly: solve(y,x) there seems to be no solution found, nor the algorithm exiting and telling that no solution was found, so I would think this is a bug of some sort.

I agree with you , I have understood the issue in a wrong way ,so there seems to be a bug in solve.

[oscarbenjamin]

If you pass rational=False then it returns quickly:

In [7]: solve(y, rational=False)
Out[7]: [11.610838471061, 5.33333333333333⋅W(-0.184305885067861, -1) + 12.8511722731906]

The problem is that the floats nsimplify to complicated rationals:

In [8]: nsimplify(y)
Out[8]: 
                                              -3⋅x 
                                              ─────
                                                16 
  1308⋅x   672373333333333   572373333333333⋅ℯ     
- ────── + ─────────────── - ──────────────────────
    25      1000000000000        1000000000000  

Internally the slow part in solve is attempting to compute 572373333333333**279040000000000 which is a stupendously large integer. It would be better if that Pow would remain unevaluated.

There are a number of related issues but see also #6835