Is there a way to perform a "literal replacing" in SymPy, which matches exactly the pattern?
For example, for the expression exp(I*theta)+exp(-I*theta), I would like to replace only exp(It) to zero without replacing exp(-It)
A minimal example would be

from sympy import *
f=exp(I*t)+exp(-I*t)

Clearly, subs will not work

f.subs((exp(I*t),0))

because it also replaces exp(-I*t) with infinity.

>Solution :

You can use the xreplace method:

f.xreplace({exp(I*t):0})

Returns:

exp(-t*I)

Here is the documentation for reference.