Replace a column in a csv file using python?

Question

How to replace a column in csv file with a condition in python ?

My csv file contains:

 34,2016-06-14 13:37:10,,1,,,t
 34,2016-06-15 13:48:20,,1,,,f
 34,2016-06-16 13:55:34,,1,,,t
 34,2016-06-17 13:48:40,,1,,,f
 34,2016-06-18 13:41:27,,1,,,t

I want to change the last column according to a condition. That is I want to change "t" to "1" and "f" to "0" using python program.

I have seen many examples but not with this condition and examples are little bit confusing.

I tried a little bit but it is not working. I also don't know if it is correct method or not

f = open('/home/rck/Desktop/attendance/month.csv','rb')
fo = open('/home/rck/Desktop/attendance/demo.csv','wb')

for line in f:
    bits = line.split(',')

if bits[5] == '"t"':
    bits[5] = '"1"'    
elif bits[5] == '"f"':
    bits[5] = '"0"'
fo.write( ','.join(bits) )

f.close()
fo.close()

How can I achieve this?

Thanks....

Answer 1

Looking at your example there are 7 elements to the bits array, so you need to test bits[6] not bits[5] . If you know you always want the last one, bits[-1] would work.

Also, I don't think your quoting '"t"' is correct. I fared better with 't' .

Finally, the last element (for me anyway) contains the newline character. So to achieve a match I changed the tests to this form:

    if bits[-1] == 't\n':

This then worked for me. rafalmp's answer works as it does a search and replace on the whole line looking for the character 't' followed by a newline, which it is.

One design difference however is that when using code based on your example, if the code encounters an unknown character (not a 't' or an 'f') it would leave the character unaltered. rafalmp's code would always change it to a '0' if it were not a 't'. to have the same effect in your example, change the elif line to an else with no supplemental test.

Answer 2

You can use regular expression to do the job:

import re

for line in f:
    line = re.sub('\w$', lambda x: '"1"' if x.group(0) == 't' else '"0"', line)
    fo.write(line)

Answer 3

Unless you insist on python , it would easily be done with awk as following:

awk '{$7= $7=="t" ? "1" : "0"}1' FS=',' OFS=',' month.csv

where replace 7th column with 0/1 using ternary and FS (field separator) and OFS (output field separator) is , .