CODEWITHSUNDEEP
pythoncsv
Trying_hard
Trying_hard

Reputation: 9501

Comparing two csv files and getting difference

I have two csv file I need to compare and then spit out the differnces:

CSV FORMAT:

 Name   Produce   Number
 Adam   Apple     5
 Tom    Orange    4
 Adam   Orange    11

I need to compare the two csv files and then tell me if there is a difference between Adams apples on sheet and sheet 2 and do that for all names and produce numbers. Both CSV files will be formated the same.

Any pointers will be greatly appreciated

Upvotes: 8

Views: 31581

Answers (6)

Aakash Gupta
Aakash Gupta

Reputation: 776

I have used csvdiff 

$pip install csvdiff
$csvdiff --style=compact col1 a.csv b.csv

Link to package on pypi

I found this link useful

Upvotes: 8

octopusgrabbus
octopusgrabbus

Reputation: 10695

If you want to use Python's csv module along with a function generator, you can use nested looping and compare large .csv files. The example below compares each row using a cursory comparision:

import csv

def csv_lazy_get(csvfile):
    with open(csvfile) as f:
        r = csv.reader(f)
        for row in r:
            yield row

def csv_cmp_lazy(csvfile1, csvfile2):
    gen_2 = csv_lazy_get(csvfile2)

    for row_1 in csv_lazy_get(csvfile1):
        row_2 = gen_2.next()

        print("row_1: ", row_1)
        print("row_2: ", row_2)

        if row_2 == row_1:
            print("row_1 is equal to row_2.")
        else:
            print("row_1 is not equal to row_2.")

    gen_2.close()

Upvotes: 1

Jon Clements
Jon Clements

Reputation: 142256

If your CSV files aren't so large they'll bring your machine to its knees if you load them into memory, then you could try something like:

import csv
csv1 = list(csv.DictReader(open('file1.csv')))
csv2 = list(csv.DictReader(open('file2.csv')))
set1 = set(csv1)
set2 = set(csv2)
print set1 - set2 # in 1, not in 2
print set2 - set1 # in 2, not in 1
print set1 & set2 # in both

For large files, you could load them into a SQLite3 database and use SQL queries to do the same, or sort by relevant keys and then do a match-merge.

Upvotes: 5

Hugh Bothwell
Hugh Bothwell

Reputation: 56714

import csv

def load_csv_to_dict(fname, get_key, get_data):
    with open(fname, 'rb') as inf:
        incsv = csv.reader(inf)
        incsv.next()  # skip header
        return {get_key(row):get_data(row) for row in incsv}

def main():
    key = lambda r: tuple(r[0:2])
    data = lambda r: int(r[2])
    f1 = load_csv_to_dict('file1.csv', key, data)
    f2 = load_csv_to_dict('file2.csv', key, data)

    f1keys = set(f1.iterkeys())
    f2keys = set(f2.iterkeys())

    print("Keys in file1 but not file2:")
    print(", ".join(str(a)+":"+str(b) for a,b in (f1keys-f2keys)))

    print("Keys in file2 but not file1:")
    print(", ".join(str(a)+":"+str(b) for a,b in (f2keys-f1keys)))

    print("Differing values:")
    for k in (f1keys & f2keys):
        a,b = f1[k], f2[k]
        if a != b:
            print("{}:{} {} <> {}".format(k[0],k[1], a, b))

if __name__=="__main__":
    main()

Upvotes: 1

ChrisP
ChrisP

Reputation: 5952

Here a start that does not use difflib. It is really just a point to build from because maybe Adam and apples appear twice on the sheet; can you ensure that is not the case? Should the apples be summed, or is that an error?

import csv
fsock = open('sheet.csv','rU')
rdr = csv.reader(fsock)
sheet1 = {}
for row in rdr:
    name, produce, amount = row
    sheet1[(name, produce)] = int(amount) # always an integer?
fsock.close()
# repeat the above for the second sheet, then compare

You get the idea?

Upvotes: 0

SomeKittens
SomeKittens

Reputation: 39522

One of the best utilities for comparing two different files is diff.

See Python implementation here: Comparing two .txt files using difflib in Python

Upvotes: 1

Related Questions