如何比较python3中的两个csv文件？

我想比较file1.csv列A与file2.csv列B，file1.csv列B与file2.csv列c，file1.csv列c与file2.csv列A

import pandas as pd
df = pd.read_csv('table.csv', delimiter=',')
list1 = [list(row) for row in df.values]
#print(list_of_rows)

df1 = pd.read_csv('output_csv_file.csv', delimiter=',')
list2 = [list(row) for row in df1.values]
#print(list_of_rows1,"********************8")

new_items = []
for item in list2:
   if not any(x[1] == item[1] for x in list1):
       if not any(x[2] != item[2] for x in list1):       
           new_items.append(item)

示例输入文件和我添加的预期输出文件请指导我

file1                                                       

col_A    clo_B                    clo_C          col_D            

12345    2020-09-08 09:10:20      10.100.90.89   15



file2                                                       

col_A               clo_B          clo_C                      col_D   col_E    col_F     

10.100.90.89        21301          2020-09-08 09:10:20         15      1       6

10.100.90.89        12345          2020-09-08 09:10:20         15      1       6

10.100.90.89        12345          2020-06-08 09:10:20         15      1       6

Output:

col_A               clo_B          clo_C                      col_D   col_E    col_F     

10.100.90.89        21301          2020-09-08 09:10:20         15      1       6

10.100.90.89        12345          2020-06-08 09:10:20         15      1       6

---

如果您有数据帧：

list1 = [
    ['12345', '2020-09-08 09:10:20', '10.100.90.89', '15']
]

list2 = [
    ['10.100.90.89','21301','2020-09-08 09:10:20','15','1','6'],
    ['10.100.90.89','12345','2020-09-08 09:10:20','15','1','6'],
    ['10.100.90.89','12345','2020-06-08 09:10:20','15','1','6'],
]

df1 = pd.DataFrame(list1, columns=['col_A','col_B','col_C','col_D'])
df2 = pd.DataFrame(list2, columns=['col_A','col_B','col_C','col_D','col_E','col_F'])

print(df1)
print(df2)

---

---

然后：

将删除重复项：

          col_A  col_B                col_C col_D col_E col_F
0  10.100.90.89  21301  2020-09-08 09:10:20    15     1     6
2  10.100.90.89  12345  2020-06-08 09:10:20    15     1     6

---

你能编辑你的问题并把样本数据和预期输出放在那里吗？我编辑了请检查@Andrejkeslycan你能把CSV文件的内容样本和预期输出放在那里吗？我添加了请检查并帮助我@Andrejkeslycan你能解释规则吗，为什么最终输出有两行？有必要提及列名吗？@pythonnewbie根据您的示例，csv文件1的列顺序与csv文件2的列顺序不同，因此我对它们进行了相应的重新排序

tmp = pd.DataFrame(df1[['col_C', 'col_A', 'col_B', 'col_D']])
tmp.columns = ['col_A','col_B','col_C','col_D']

out = df2.merge(tmp, on=['col_A', 'col_B', 'col_C', 'col_D'], how='left', indicator=True)
out = out[out['_merge'] == 'left_only'].drop(columns='_merge')
print(out)

          col_A  col_B                col_C col_D col_E col_F
0  10.100.90.89  21301  2020-09-08 09:10:20    15     1     6
2  10.100.90.89  12345  2020-06-08 09:10:20    15     1     6