سه روش سریع برای مقایسه داده ها با پایتون — راهنمای کاربردی

1import hashlib, sys
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3def md5(fname):
4    md5hash = hashlib.md5()
5    with open(fname) as handle: #opening the file one line at a time for memory considerations
6        for line in handle:
7            md5hash.update(line.encode('utf-8'))
8    return(md5hash.hexdigest())
9print('Comparing Files:',files[0],'and',files[1])
10if md5(files[0]) == md5(files[1]):
11    print('Matched')
12else:
13    print('Not Matched')

1import hashlib, sys
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3def sha1(fname):
4    sha1hash = hashlib.sha1()
5    with open(fname) as handle: #opening the file one line at a time for memory considerations
6        for line in handle:
7            sha1hash.update(line.encode('utf-8'))
8    return(sha1hash.hexdigest())
9print('Comparing Files:',files[0],'and',files[1])
10if sha1(files[0]) == sha1(files[1]):
11    print('Matched')
12else:
13    print('Not Matched')

1import sys, sqlite3, pandas as pd
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3conn = sqlite3.connect(':memory:') #we are spinning an SQL db in memory
4cur = conn.cursor()
5chunksize = 10000
6i=0
7for file in files:
8    i = i+1
9    for chunk in pd.read_csv(file, chunksize=chunksize): #load the file in chunks in case its too big
10        chunk.columns = chunk.columns.str.replace(' ', '_') #replacing spaces with underscores for column names
11        chunk.to_sql(name='file' + str(i), con=conn, if_exists='append')
12print('Comparing', files[0], 'to', files[1]) #Compare if all data from File[0] are present in File[1]
13cur.execute( '''SELECT * FROM File1
14                EXCEPT
15                SELECT * FROM File2''')
16i=0
17for row in cur:
18    print(row)
19    i=i+1
20if i==0: print('No Differences')
21print('Comparing', files[1], 'to', files[0]) #Compare if all data from File[1] are present in File[0]
22cur.execute( '''SELECT * FROM File2
23                EXCEPT
24                SELECT * FROM File1''')
25i=0
26for row in cur:
27    print(row)
28    i=i+1
29if i==0: print('No Differences')
30cur.close()

1import sys, sqlite3, pandas as pd
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3df1 = pd.read_csv(files[0])
4df2 = pd.read_csv(files[1])
5df3 = df1.equals(df2)
6print('Matches:', df3)

1import sys, sqlite3, pandas as pd
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3df1 = pd.read_csv(files[0])
4df2 = pd.read_csv(files[1])
5df3 = (df1 != df2).any(axis=None)
6print('Differences in file:', df3)
7df3 = (df1 != df2).any(1)
8ne_stacked = (df1 != df2).stack()
9changed = ne_stacked[ne_stacked]
10changed.index.names = ['id', 'col']
11print('Differences In:')
12print(changed)

1import sys, sqlite3, pandas as pd, numpy as np
2files = [sys.argv[1], sys.argv[2]] #these are the arguments we take
3df1 = pd.read_csv(files[0])
4df2 = pd.read_csv(files[1])
5df3 = df1.eq(df2)
6print(df3.all())
7#print(df3.all(axis=1))
8df4 = df3.all(axis=1)
9df4 = pd.DataFrame(df4, columns=['Columns'])
10print(df4[df4['Columns']==False])