使用Python编辑CSV文件的特定位置

Question

I have a CSV file that looks as such:

header1, header2, header3, header4
1, 2, 3, abc
4, 5, 6, abc
7, 8, 9, abc

My goal is to only change the values marked as "abc". I want the file to look like this when I'm done:

header1, header2, header3, header4
1, 2, 3, test
4, 5, 6, test
7, 8, 9, test

The code I have come up with so far is as follows:

import csv

with open('test2-2.csv', 'w') as csvout:
    write=csv.writer(csvout, delimiter=',', lineterminator='\r')
    with open('test2.csv', 'rb') as csvfile:
        read=csv.reader(csvfile, delimiter=',')
        for row in read:    
            row[3]="test"
            write.writenow(row)

The problem with this code is that it overwrites the "header4" location as well, so it comes out looking like this:

header1, header2, header3, test
1, 2, 3, test
4, 5, 6, test
7, 8, 9, test

Is there any way to specify the index locations I want to change?

Thanks for any help!

Here is the new code that works with this example:

import csv

with open('test2-2.csv', 'w') as csvout:
    write=csv.writer(csvout, delimiter=',', lineterminator='\r')
    with open('test.csv', 'rb') as csvfile:
        read=csv.reader(csvfile, delimiter=',')
        header=next(read)
        write.writerow(header)
        for row in read:    
            row[3]="test"
            write.writerow(row)

Answer 1

If you're just looking for a way to treat the header row separately, that's pretty easy: Just process the first row separately.

with open('test2-2.csv', 'w') as csvout:
    write=csv.writer(csvout, delimiter=',', lineterminator='\r')
    with open('test2.csv', 'rb') as csvfile:
        read=csv.reader(csvfile, delimiter=',')
        header = next(read)
        write.writerow(header)
        for row in read:
            row[3]="test"
            write.writenow(row)

(Note that this still has the same bugs are your original code—at least one typo, writenow , a missing skipinitialspace=True , etc.)

---

If you want to change things by some different rule, just write a different transformation. As long as you can describe it in English, you should be able to convert it to Python pretty easily.

For example, if you wanted to change any 6 in any column into test :

for row in read:
    row = ('test' if col == '6' else col for col in row)
    write.writenow(row)

Or, if you wanted to change column 2 of row 3 into test :

for i, row in enumerate(read):
    if i == 3:
        row[2] = 'test'
    write.writenow(row)

Or, if you wanted to change any 6 in column 2 of any row:

for i, row in enumerate(read):
    if row[2] == '6':
        row[2] = 'test'
    write.writenow(row)

… and so on.

Answer 2

Just wrap your replacing code with an if block, and use enumerate() (see #added comments):

import csv

with open('test2-2.csv', 'w') as csvout:
    write=csv.writer(csvout, delimiter=',', lineterminator='\r')
    with open('test2.csv', 'rb') as csvfile:
        read=csv.reader(csvfile, delimiter=',')
        for row in enumerate(read): #modified
            if row[0] >= 1: #added
                row[1][3]="test"
                write.writenow(row)