# Solve Output by Type#

The output of the solve() function can seem very unwieldy since it may appear to arbitrarily return one of six different types of output (in addition to raising errors). The reasons for this are historical and are biased toward human interaction rather than programmatic use. The type of output will depend on the type of equation(s) (and how they are entered) and the number of symbols that are provided (and how they are provided).

>>> from sympy import sqrt, exp, solve, Symbol, Eq
>>> from sympy.abc import x, y, z, a, b


The solve() function attempts to find all values for as many symbols as possible that will make each expression given equal to zero. The format of the output can be controlled by using the dict or set keyword:

>>> solve(x - 1, dict=True)
[{x: 1}]
>>> solve([x**2 - y, x + y - 6], set=True)
([x, y], {(-3, 9), (2, 4)})


The following discussion provides an explanation for the output obtained when not using those keywords.

## Empty List#

When there is no solution, an empty list is returned.

>>> solve(sqrt(x) + 1)  # or solve(sqrt(x) + 1, dict=True)
[]
>>> solve(sqrt(x) + 1, set=True)
([x], set())


## List Of Values#

A list of values is given when the symbol to solve for was unambiguous in context because a) the equation was univariate or b) a single symbol was specified as being of interest.

>>> solve(x**2 - 4)
[-2, 2]
>>> solve(x - y - 1, x)
[y + 1]


## Single Dictionary#

A single dictionary with keys being symbols and values being the solutions for those symbols is the result when equations are passed as a list and are all linear in the symbols given. Note: such a system is automatically generated for a single equation (not passed as a list) if there is an undetermined-coefficients solution for the symbols specified. If this is not what was intended, then pass the expression in a list.

>>> solve([x + y - 2, x - y + 2], x, y)
{x: 0, y: 2}
>>> eq = a*x - 2*x + b - 5
>>> solve(eq, {a, b})  # undetermined coefficients
{a: 2, b: 5}
>>> solve([eq], {a, b})  # algebraic
{a: -b/x + (2*x + 5)/x}


## List of Tuples#

Each tuple in the list gives a solution for the symbols in the order they were given. This format is used when a) a list of equations contains at least one nonlinear equation or b) a list of symbols is given in a well defined order. (This is also the format for the tuples in the set returned when using the flag set=True.)

>>> solve(x - 1, x, y)  # more than one symbol
[(1, y)]
>>> solve([x**2], x)  # list with nonlinear equation
[(0,)]
>>> solve([x**2 - 1], x)
[(-1,), (1,)]
>>> solve([x**2 - y, x - 3], x, y)  # nonlinear and multiple symbols
[(3, 9)]


## List of Dictionaries#

The list of dictionaries is returned when the expression was not univariate or there was a nonlinear expression in a list and the order of symbols would otherwise be ambiguous because a) no symbols were passed or b) the symbols were passed as a set. (This is also the format selected with dict=True.)

>>> solve(x - y)
[{x: y}]
>>> solve([exp(x) - 1, x*(x - 1)])
[{x: 0}]
>>> system = [x + y - z, x**2 - y + z, exp(z) + 1/x + 1/y - 2]
>>> sol = solve(system[:2]); sol
[{x: -1, y: z + 1}, {x: 0, y: z}]


The dictionaries only contain values that are distinct from the keys. In the last example above, there is no key for z in the dictionary since only two of the three equations were insufficient to determine its value. These solutions can be used to eliminate those variables from the third equation, however, to give a relationship in a single variable that can be solved (perhaps numerically) to obtain a full solution with the advantage of only needing to guess a single value instead of three.

>>> from sympy import nsolve
>>> [system[-1].subs(s) for s in sol]
[exp(z) - 3 + 1/(z + 1), exp(z) + zoo + 1/z]
>>> z_eq = _[0]
>>> zsol = nsolve(z_eq, 1); zsol
0.906425478894557
>>> sol0 = {k: v.subs(z, zsol) for k, v in sol[0].items()}
>>> sol0[z] = zsol; sol0
{x: -1, y: 1.90642547889456, z: 0.906425478894557}


## Boolean or Relational#

A boolean expression is returned when a relational expression other than an Equality is given as an expression to solve. A single $$Equality$$ or a more complicated relational expression might be returned. The use of solve() here is equivalent to passing the equation set and symbols to reduce_inequalities() (and dict, set, and check flags are ignored).

>>> solve([x**2 > 4, x > 0])
(2 < x) & (x < oo)

>>> from sympy import Unequality as Ne
>>> solve([x**2 - 4, Ne(x, -2)])
Eq(x, 2)


Any returned $$Equality$$ can be converted to a dictionary:

>>> {_.lhs: _.rhs}
{x: 2}