Python : Compare two csv files and print out differences

Question

I need to compare two CSV files and print out differences in a third CSV file. In my case, the first CSV is a old list of hash named old.csv and the second CSV is the new list of hash which contains both old and new hash.

Here is my code :

import csv
t1 = open('old.csv', 'r')
t2 = open('new.csv', 'r')
fileone = t1.readlines()
filetwo = t2.readlines()
t1.close()
t2.close()

outFile = open('update.csv', 'w')
x = 0
for i in fileone:
    if i != filetwo[x]:
        outFile.write(filetwo[x])
    x += 1
outFile.close()

The third file is a copy of the old one and not the update. What's wrong ? I Hope you can help me, many thanks !!

PS : i don't want to use diff

Answer 1

You may find this package useful (csv-diff):

pip install csv-diff

Once installed, you can run it from the command line:

csv-diff one.csv two.csv --key=id

Answer 2

It feels natural detecting differences using sets.

#!/usr/bin/env python3

import sys
import argparse
import csv


def get_dataset(f):
    return set(map(tuple, csv.reader(f)))


def main(f1, f2, outfile, sorting_column):
    set1 = get_dataset(f1)
    set2 = get_dataset(f2)
    different = set1 ^ set2

    output = csv.writer(outfile)

    for row in sorted(different, key=lambda x: x[sorting_column], reverse=True):
        output.writerow(row)


if __name__ == '__main__':
    parser = argparse.ArgumentParser()

    parser.add_argument('infile', nargs=2, type=argparse.FileType('r'))
    parser.add_argument('outfile', nargs='?', type=argparse.FileType('w'), default=sys.stdout)
    parser.add_argument('-sc', '--sorting-column', nargs='?', type=int, default=0)

    args = parser.parse_args()

    main(*args.infile, args.outfile, args.sorting_column)

Answer 3

I assumed your new file was just like your old one, except that some lines were added in between the old ones. The old lines in both files are stored in the same order.

Try this :

with open('old.csv', 'r') as t1:
    old_csv = t1.readlines()
with open('new.csv', 'r') as t2:
    new_csv = t2.readlines()

with open('update.csv', 'w') as out_file:
    line_in_new = 0
    line_in_old = 0
    while line_in_new < len(new_csv) and line_in_old < len(old_csv):
        if old_csv[line_in_old] != new_csv[line_in_new]:
            out_file.write(new_csv[line_in_new])
        else:
            line_in_old += 1
        line_in_new += 1

• Note that I used the context manager with and some meaningful variable names, which makes it instantly easier to understand. And you don't need the csv package since you're not using any of its functionalities here.
• About your code, you were almost doing the right thing, except that _you must not go to the next line in your old CSV unless you are reading the same thing in both CSVs. That is to say, if you find a new line, keep reading the new file until you stumble upon an old one and then you'll be able to continue reading.

UPDATE: This solution is not as pretty as Chris Mueller's one which is perfect and very Pythonic for small files, but it only reads the files once (keeping the idea of your original algorithm), thus it can be better if you have larger file.

Answer 4

with open('first_test_pipe.csv', 'r') as t1, open('validation.csv', 'r') as t2:
    filecoming = t1.readlines()
    filevalidation = t2.readlines()

for i in range(0,len(filevalidation)):
    coming_set = set(filecoming[i].replace("\n","").split(","))
    validation_set = set(filevalidation[i].replace("\n","").split(","))
    ReceivedDataList=list(validation_set.intersection(coming_set))
    NotReceivedDataList=list(coming_set.union(validation_set)- 
    coming_set.intersection(validation_set))
    print(NotReceivedDataList)

Answer 5

import pandas as pd
import sys
import csv

def dataframe_difference(df1: pd.DataFrame, df2: pd.DataFrame, csvfile, which=None):
    """Find rows which are different between two DataFrames."""
    comparison_df = df1.merge(
        df2,
        indicator=True,
        how='outer'
    )
    if which is None:
        diff_df = comparison_df[comparison_df['_merge'] != 'both']
    else:
        diff_df = comparison_df[comparison_df['_merge'] == which]
    diff_df.to_csv(csvfile)
    return diff_df


if __name__ == '__main__':
    df1 = pd.read_csv(sys.argv[1], sep=',')    
    df2 = pd.read_csv(sys.argv[2], sep=',')

    df1.sort_values(sys.argv[3])
    df2.sort_values(sys.argv[3])
    #df1.drop(df1.columns[list(map(int, sys.argv[4].split()))], axis = 1, inplace = True)
    #df2.drop(df2.columns[list(map(int, sys.argv[4].split()))], axis = 1, inplace = True)

    print(dataframe_difference(df1, df2, sys.argv[5]))

to use run:

python3 script.py file1.csv file2.csv some_common_header_to_sort_each_file output_file.csv

In case you want to drop any columns from comparasion, uncomment df.drop part and run

python3 script.py file1.csv file2.csv some_common_header_to_sort_each_file "x y z..." output_file.csv

where x,y,z are the column numbers to drop, index starts from 0.

Answer 6

Thanks to @vishnoo-rath's comment under one of the above answers, for providing a link to the following page : https://github.com/simonw/csv-diff#as-a-python-library

from csv_diff import load_csv, compare
diff = compare(
    load_csv(open("one.csv"), key="id"),
    load_csv(open("two.csv"), key="id")
)
print(diff)