Why is this not correct? (codeeval challenge)PYTHON

This code works for me:

import sys


def main(name_file):
    _file = open(name_file, 'r')
    text = ""
    while True:
        try:
            line = _file.next()
            disordered_line, numbers_string = line.split(';')
            numbers_list = map(int, numbers_string.strip().split(' '))
            missing_number = sum(xrange(sorted(numbers_list)[0],sorted(numbers_list)[-1]+1)) - sum(numbers_list)
            if missing_number == 0:
                missing_number = len(disordered_line)
            numbers_list.append(missing_number)
            disordered_list = disordered_line.split(' ')
            string_position = zip(disordered_list, numbers_list)
            ordered = sorted(string_position, key = lambda x: x[1])
            text +=  " ".join([x[0] for x in ordered])
            text += "\n"
        except StopIteration:
            break
    _file.close()
    print text.strip()

if __name__ == '__main__':
    main(sys.argv[1])

I'll try to explain my code step by step so maybe you can see the difference between your code and mine one:

while True 

A loop that breaks when there are no more lines.

try:

I put the code inside a try and catch the StopIteracion exception, because this is raised when there are no more items in a generator.

line = _file.next()

Use a generator, so that way you do not put all the lines in memory from once.

disordered_line, numbers_string = line.split(';')

Get the unordered phrase and the numbers of every string's position.

numbers_list = map(int, numbers_string.strip().split(' '))

Convert every number from string to int

missing_number = sum(xrange(sorted(numbers_list)[0],sorted(numbers_list)[-1]+1)) - sum(numbers_list)

Get the missing number from the serial of numbers, so that missing number is the position of the last string in the phrase.

if missing_number == 0:
    missing_number = len(unorder_line)

Check if the missing number is equal to 0 if so then the really missing number is equal to the number of the strings that make the phrase.

numbers_list.append(missing_number)

Append the missing number to the list of numbers.

disordered_list = disordered_line.split(' ')

Conver the disordered phrase into a list.

string_position = zip(disordered_list, numbers_list)

Combine every string with its respective position.

ordered = sorted(string_position, key = lambda x: x[1])

Order the combined list by the position of the string.

text +=  " ".join([x[0] for x in ordered])

Concatenate the ordered phrase, and the reamining code it's easy to understand.

UPDATE

By looking at your code here is my opinion tha might solve your problem.

split already returns a list so you do not have to loop over the splitted content to add that content to another list.

So these six lines:

for l in test.strip().split(";"):
    saver.append(l)
for i in saver[0].split(" "):
    textList.append(i)
for j in saver[1].split(" "):
    positionList.append(j)

can be converted into three:

splitted_test = test.strip().split(';')
textList = splitted_test[0].split(" ")
positionList = map(int, splitted_test[1].split(" "))

In this line positionList = map(int, splitted_test[0].split(" ")) You already convert numbers into int, so you save these two lines:

for i in range(0,len(positionList)):
    positionList[i]=int(positionList[i])

The next lines:

accomodator=[None]*len(textList)

for n in range(1,len(textList)):
    if n not in positionList:
         accomodator[n]=textList[len(textList)-1]
         exists=n       

can be converted into the next four:

missing_number = sum(xrange(sorted(positionList)[0],sorted(positionList)[-1]+1)) - sum(positionList)
if missing_number == 0:
    missing_number = len(textList)
    positionList.append(missing_number)

Basically what these lines do is calculate the missing number in the serie of numbers so the len of the serie is the same as textList.

The next lines:

for item in positionList:
    accomodator[item-1]=textList[counter]
    counter+=1
    if counter>item:
       accomodator[exists-1]=textList[counter]
for word in accomodator:
   text+=str(word) + " "

Can be replaced by these ones:

string_position = zip(textList, positionList) 
ordered = sorted(string_position, key = lambda x: x[1])
text +=  " ".join([x[0] for x in ordered])
text += "\n"

From this way you can save, lines and memory, also use xrange instead of range.

Maybe the factors that make your code pass partially could be:

1. Number of lines of the script
2. Number of time your script takes.
3. Number of memory your script uses.

What you could do is:

1. Use Generators. #You save memory
2. Reduce for's, this way you save lines of code and time.
3. If you think something could be made it easier, do it.
4. Do not redo the wheel, if something has been already made it, use it.