SymPy automatically processes expressions

I have been using SymPy to c开发者_运维百科onvert expressions into latex (to then be rendered by Matplotlib). e.g.

from sympy import latex, sympify
from sympy.abc import x

str = '2*x + 3*x'

TeX = latex(sympify(str))

The problem is that it automatically processes the expression, so 2*x + 3*x automatically becomes 5*x etc; which is not what I want (don't ask!).

Sympy's Add class handles the addition of symbols. You can provide a keyword argument to stop the automatic collection of terms.

from sympy import Add
from sympy.abc import x

eq = Add(2*x, 3*x, evaluate=False)

# this will print: 2*x + 3*x
print eq

This may not be exactly what you want based on your reply to phimuemue's comment.

Actually when you call sympify(str) it tries to parse expression and convert them into the default classes. In this case Add(2*x,2*x) will be called (with default parameter evalaute=True) so this became 5*x. If you want to avoid it either you have to call Add(2*x,3*x,evaluate=False) or use some global variable and check in init method of AssocOp class in core-> operation.py

i am doing this

 try:
        import __builtin__
        evaluate_expr=__builtin__.evaluate_expr
    except AttributeError ,ex:
        pass 

    if ((not options.pop('evaluate', True)) or (evaluate_expr==False)) :

**Note - sympy use caching for function , so if you call same function (say :sympy("2*x+3*x") )two times .1st time with your gloabl variable evalute=True and 2nd time with evaluate=False. in both cases becuase of caching you will get same result. So you need to update methods like (add ,mul ) in core->expr class. something like below

def __add__(self, other):
        #simplifychange:
        evaluate_expr=self.get_evaluate()
        return Add(self, other,evaluate=evaluate_expr)

But i will suggest it would be better if you dont use evaluate=False. Behavior of sympy will change dramatically. problems like i was facing in this post

>>> import re

>>> re.sub("(^|[\(\[\{\+\-\*/ ,\:=])(\-?[0-9]*\.?[0-9]+)", "\\1Dummy('\\2')", '2*x + 3*x')

"Dummy('2')*x + Dummy('3')*x"

>>> eval(_)

2⋅x + 3⋅x

>>> latex(_)

'2 x + 3 x'