如何使用Python用随机数替换文件中的数字？

Question

I have a file that contains numbers in this format:

78 23 69 26 56 59 74 45 94 28 37 
62 52 84 27 12 95 86 86 12 89 92
43 84 88 22 31 25 80 40 59 32 98

(All the numbers are on a single wrapped line in notepad++, and it contains 1,5k of 2 digit number sets, with space in between)

What I want to do is randomize some of the numbers everytime I run the Python code, so the second .tmp file will be unique, but keeping the same format.

So I tried this and worked, but using static numbers: 12 as search and 55 as target.

infile = open('file.txt', 'r')
outfile = open('file.txt.tmp', 'w')
for line in infile:
    outfile.write(line.replace('12', '55'))

infile.colse()
outfile.colse()

However, for a better randomization, what I want to do is to use random numbers between 10-99 instead of static numbers like 12 and 55.

So what I tried to do (and failed) is replacing the static 12 and 55 numbers to randomized ones like that:

randnum1 = randint(10,99)
randnum2 = randint(10,99)

infile = open('file.txt', 'r')
outfile = open('file.txt.tmp', 'w')
for line in infile:
    outfile.write(line.replace(randnum1, randnum2))

And I get this error:

Traceback (most recent call last):
  File "<pyshell#579>", line 2, in <module>
    outfile.write(line.replace(randnum1, randnum2))
TypeError: Can't convert 'int' object to str implicitly

Answer 1

randint gives an int , which needs to be converted to str .

Try outfile.write(line.replace(str(randnum1), str(randnum2)))

As simple as that :)

Answer 2

The error exactly says what the problem is: TypeError: Can't convert 'int' object to str implicitly . It is issued because randnum1 and randnum2 are int s and not str s.

You must convert them to str by calling str(randnum1) and str(randnum2) , eg like this:

randnum1 = randint(10,99)
randnum2 = randint(10,99)
randnum1 = str(randnum1)
randnum2 = str(randnum2)

infile = open('file.txt', 'r')
outfile = open('file.txt.tmp', 'w')
for line in infile:
    outfile.write(line.replace(randnum1, randnum2))

---

Note: it is not recommended to use one variable name multiple times with multiple value types as it makes the code less readable. However, in this case, it is re-used once so it does not harm the readability very much.

Answer 3

Just in case you find it useful, you could take the following approach. This first reads your single line in and splits it into a list. It then picks 10 random entries from the list and replaces the entry with a new random number between 10 and 99 . Lastly it writes the new data back to the file.

from random import randint

with open('input.txt') as f_input:
    data = f_input.readline().split()
    entries = len(data) - 1

for _ in xrange(10):
    data[randint(0, entries)] = str(randint(10, 99))

with open('input.txt', 'w') as f_output:
    f_output.write(' '.join(data))