Nested list doesn't work properly

Question

import re
def get_number(element):
    re_number = re.match("(\d+\.?\d*)", element)

    if re_number:
        return float(re_number.group(1))
    else:
        return 1.0

def getvalues(equation):
    elements = re.findall("([a-z0-9.]+)", equation)
    return [get_number(element) for element in elements]


eqn = []
eqn_no = int(raw_input("Enter the number of equations: "))

for i in range(eqn_no):
    eqn.append(getvalues(str(raw_input("Enter Equation %d: " % (i+1)))))
print "Main Matrix: "
for i in range((eqn_no)):
    for j in range((eqn_no+1)):
        print "\t%f" %(eqn[i][j]),
    print
print
equation=[]
equation=eqn
for k in range((eqn_no-1)):
    for i in range((k+1),eqn_no):
        for j in range((eqn_no+1)):
            if(eqn[i][j]!=0):
                eqn[i][j]=eqn[i][j]-(eqn[k][j]*(equation[i][k]/eqn[k][k]))



    print "Matrix After %d step: " %(k+1)
    for i in range(eqn_no):
            for j in range((eqn_no+1)):
                print "\t%f"%(eqn[i][j]),
                equation[i][j]=eqn[i][j];

            print
    print

for input:

25x+5y+z=106.8
64x+8y+z=177.2
144x+12y+z=279.2

output is:

Main Matrix: 
    25.000000   5.000000    1.000000    106.800000
    64.000000   8.000000    1.000000    177.200000
    144.000000  12.000000   1.000000    279.200000

Matrix After 1 step: 
    25.000000   5.000000    1.000000    106.800000
    0.000000    8.000000    1.000000    177.200000
    0.000000    12.000000   1.000000    279.200000

Matrix After 2 step: 
    25.000000   5.000000    1.000000    106.800000
    0.000000    8.000000    1.000000    177.200000
    0.000000    0.000000    1.000000    279.200000

But it should be like

Main Matrix: 
    25.000000   5.000000    1.000000    106.800000
    64.000000   8.000000    1.000000    177.200000
    144.000000  12.000000   1.000000    279.200000

Matrix After 1 step: 
    25.000000   5.000000    1.000000    106.800000
    0.000000    -4.80000    -1.56000    -96.208000
    0.000000    -16.8000    -4.76000    -335.968000

Matrix After 2 step: 
    25.000000   5.000000    1.000000    106.800000
    0.000000    -4.80000    -1.56000    -96.208000
    0.000000    0.000000    0.699999    0.759981

First of all this is a partial code for solving root of n number of equations using Naive Guass elemination method. Does anyone have any idea why the hell on earth is this happening? Why the zero parts are changing and others aren't? I have done this code in c++ and it works there perfectly but here I'm facing many problem. Maybe I'm newbie to python. I'm using python 2.7.....

Answer 1

This line:

equation=eqn

doesn't do what you think it does. Try this instead:

import copy
equation=copy.deepcopy(eqn)

Python assignment isn't a copy operation, but rather a binding operation. The line you have means "bind the name equation to the same object to which eqn is currently bound."

Your algorithm requires that equation and eqn are distinct objects.

Answer 2

I think the problem is the assignment equation = eqn. Since eqn is a list, it is a mutable and thus passed as a reference, when you assign a mutable, the variable actually contains a pointer to that object. This means that equation and eqn are the same list.

You should

from copy import deepcopy
equation = deepcopy(eqn)

You need deepcopy instead of copy because you have a list of lists, also the inner list needs to be copied.