Inputing floats, integers, or equations in raw_input to define a variable

Question

I've written this program so solve two equations based on values defined by the user. The constants kx and ky, I've defined as floats. For the range - variables start and end - I would like the user to either enter a number, or something like 6 * np.pi (6Pi). As it is now, I get the following error. How can I define this variable to allow users to enter multiple types of inputs? Thanks!

    Traceback (most recent call last):
  File "lab1_2.py", line 11, in <module>
    x = np.linspace(start, end, 256, endpoint=True)
  File "/Library/Frameworks/Python.framework/Versions/7.3/lib/python2.7/site-  packages/numpy/core/function_base.py", line 80, in linspace
    step = (stop-start)/float((num-1))
  TypeError: unsupported operand type(s) for -: 'str' and 'float'

Here's the code:

from pylab import *
import numpy as np

kx = float(raw_input("kx: "))
ky = float(raw_input("ky: "))

print "Define the range of the output:"
start = float(raw_input("From: "))
end = float((raw_input("To: "))

x = np.linspace(start, end, 256, endpoint=True)
y = np.linspace(start, end, 256, endpoint=True)

dz_dx = ((1 / 2.0) * kx * np.exp((-kx * x) / 2)) * ((2 * np.cos(kx *x)) - (np.sin(kx * x)))
dz_dy = ((1 / 2.0) * ky * np.exp((-ky * y) / 2)) * ((2 * np.cos(ky *y)) - (np.sin(ky * y)))

plot(x, dz_dx, linewidth = 1.0)
plot(y, dz_dy, linewidth = 1.0)
grid(True)


show()

Answer 1

You'll need to either parse the string yourself (the ast module would probably be useful), or use eval:

>>> s = '6*pi'
>>> eval(s,{'__builtins__': None, 'pi': np.pi})
18.84955592153876

Note that there are some nasty things that users can do with eval. My solution protects you from most of them, but not all -- pre-checking the string to make sure that there aren't any __ will make it even safer (that eliminates all of the vulnerabilities that I know of, but there could be others)

Answer 2

I would like the user to either enter a number, or something like 6 * np.pi (6Pi).

To implement this you need to do two things :

• First,check whether your input is an int or float

  This is simple. You can check for this by multiple isinstance checks

• If not, then you need to execute the input as a command and save the output in a variable

  Once you are sure it is a command, then you can use the exec or eval command to execute the input and store the output in a variable.

  Example : exec("""v = 6*2""") or v = eval("6*2") both of which assign 12 to v

Answer 3

I'm using Python 2.7 and Ecplise workspace. This source code is working fine for me, by-using raw_input.

import time

startTime = time.clock()
print "{0:*^80}".format(" Begins ")

name = raw_input('Please give the name : ')
print "Name is : ".ljust(40, '.'), name
print "Length of name : ".ljust(40, '.'), len(name)

print
radius = float(raw_input('Please enter the radius of circle : '))
print "Radius is : ".ljust(40, '.'),radius
diameter = float(2 * radius)
print "Diameter of circle is : ".ljust(40, '.'), diameter

print "\n{0:*^80}".format(" End ")

print "\nElapsed time : ", (time.clock() - startTime)*1000000, "microseconds"

---

Here, input :
name : "Sir! Guido von Rossum" and
radius : 1.1

Output :

************************************ Begins ************************************
Please give the name : Sir! Guido von Rossum
Name is : .............................. Sir! Guido von Rossum
Length of name : ....................... 21

Please enter the radius of circle : 1.1
Radius is : ............................ 1.1
Diameter of circle is : ................ 2.2

************************************* End **************************************

Elapsed time :  3593748.49903 microseconds

Thought, might be it helps you or someone else.
Thanks