python如何读取tsv文件，清理它并另存为新文件？

Question

I want to remove all punctuations from column 4 of my tsv file and then save the entire file. This is my code:

import csv
import string

exclude = set(string.punctuation)

with open("test1") as tsvfile:
    tsvreader = csv.reader(tsvfile, delimiter="\t")
    for line in tsvreader:
        line[4] = ''.join(ch for ch in line[4] if ch not in exclude)
    tsvfile.close()

The code above works fine, but my file did not save with the changes i made. How can i save the changes within the old file?

Answer 1

You are not writing any changes, you are simply changing each fifth element in each row and doing nothing with it, if you want to change the original file you can write to a tempfile and do a shutil.move to replace the original file with the updated temp:

import string

exclude = string.punctuation
from tempfile import NamedTemporaryFile
from shutil import move

with open("test1") as tsvfile, NamedTemporaryFile(dir=".",delete=False) as t:
    tsvreader = csv.reader(tsvfile, delimiter="\t")
    temp = csv.writer(t,delimiter="\t")
    for row in tsvreader:
        row[4] = row[4].strip(exclude)
        temp.writerow(row)

move(t.name,"test1")

If you want to create a new file instead of updating the original you just need to open a new file and write each cleaned row:

with open("test1") as tsvfile, open("out","w") as  t:
    tsvreader = csv.reader(tsvfile, delimiter="\t")
    temp = csv.writer(t,delimiter="\t")
    for row in tsvreader:
        row[4] = row[4].strip(exclude)
        temp.writerow(row)

To strip punctuation str.strip(exclude) will be sufficient. If you want to remove from anywhere you can go back to ''.join([ch for ch in line[4] if ch not in exclude]) but if you were removing from anywhere then you should use str.translate :

 row[4] = row[4].translate(None,exclude) 

If you want to add a space:

from string import maketrans
tbl = maketrans(exclude," "*len(exclude))

....
row[4] = row[4].translate(tbl) 

Lastly if you actually mean the fourth column then it would be row[3] not row[4]

Answer 2

You say that you want a new file, so you will need to open a second file and write the cleaned rows to it:

import csv
import string

exclude = string.punctuation

with open("test1") as tsvfile, open('out.csv') as outfile:
    tsvreader = csv.reader(tsvfile, delimiter="\t")
    tsvwriter = csv.writer(outfile, delimiter="\t")
    for row in tsvreader:
        row[4] = row[4].translate(None, string.punctuation)
        tsvwriter.writerow(row)

This uses str.translate() to remove all unwanted punctuation characters from the column. The above is for Python 2. For Python 3 use this:

row[4] = row[4].translate({ord(c): None for c in string.punctuation})