Find & replace string in multiple csv files under series of subfolders

Question

Folder structure

|FolderA
|-Foldera1
|      A_a1_001.csv
|      A_a1_002.csv
|-Folderb1
|      A_b1_001.csv
|      A_b1_002.csv

CSV format

a_a1_001.csv

XXX,XXX,XXX,XXX
123,456,789,AAA

a_a1_002.csv

XXX,XXX,XXX,XXX
376,452,953,AAA

Request

Find all the csv files in the root to find&replace certain string, and save it back as the same path & filename.

Code

import OS
texttofind='AAA'
texttoreplace='BBB'
sourcepath=os.walk(r"/Users/Gens/Documents/FolderA/")
for file in sourcepath:
    inputfile=os.walk(r"/Users/Gens/Documents/FolderA/")
    with open (inputfile,'r') as inputfile:
        filedata=inputfile.read()
        freq=0
        freq=filedata.count(texttofind)
    destinationpath=os.walk(r"/Users/Gens/Documents/FolderA/")
    filedata=filedata.replace(texttofind,texttoreplace)
    with open(destinationpath,'w')as file:
        file.write(filedata)
    file.close()

Questions

It shows:

with open (inputfile,'r') as inputfile:
TypeError: expected str, bytes or os.PathLike object, not generator

Please help me to solve this. Thanks!

Asked By: Gen1224

||

Source

Answer 1

os.walk() returns a 3-tuple

So, walk the tree finding files that end with .csv. Open in mode r+, read entire content and just do a string replace. Seek to BOF, write the (possibly) modified text then truncate.

from os import walk
from os.path import join

FIND = 'AAA'
REPLACE = 'BBB'
ROOT = '/Users/Gens/Documents/FolderA/'

for dirpath, _, filenames in walk(ROOT):
    for filename in filenames:
        if filename.endswith('.csv'):
            with open(join(dirpath, filename), 'r+') as csv: # note the open mode
                text = csv.read().replace(FIND, REPLACE)
                # seek to BOF
                csv.seek(0)
                csv.write(text)
                # truncate is necessary in case the output is smaller than the original input
                csv.truncate()

Answered By: OldBill