合并两个具有不同列数和列名的csv文件

Question

我有两个csv文件，列数不同。我想通过第一列合并这些csv文件(它们有不同的列名)。

csv 1：

ID, name, A1, A2, A3
1101,台泥,29.19,2998730.0,0.23
1102,亞泥,36.06,933069.0,0.08
1103,嘉泥,23.29,132555.0,-0.17
1104,環球水泥,25.64,139041.0,0.45
1108,幸福水泥,11.8,80854.0,0.05

csv2：

NO, no_name, D1, D2, D3, D4
01003T,兆豐新光R1,14.08,14.08,13.95,14.00
01004T,土銀富邦R2,13.39,13.40,13.30,13.34
01007T,兆豐國泰R2,14.76,14.76,14.76,14.76
1101,台泥,35.45,35.45,34.80,35.15
1102,亞泥,26.45,26.50,26.30,26.50
1103,嘉泥,8.71,8.71,8.61,8.71

如果csv2 NO = csv1 ID，将D1、D2、D3、D4附加到csv1，则新的csv如下所示：

ID, name, A1, A2, A3, D1, D2, D3, D4
1101,台泥,29.19,2998730.0,0.23,35.45,35.45,34.80,35.15
1102,亞泥,36.06,933069.0,0.08,26.45,26.50,26.30,26.50
1103,嘉泥,23.29,132555.0,-0.17,8.71,8.71,8.61,8.71

我的想法是:对两个csv文件进行读行，而不是使用双循环查找第一列的相同值，而不是将第一个csv文件和第二个csv文件中的某些列组合起来，而不是创建一个合并csv文件，下面是代码：

`import csv

# readlines for the two csv file
inf = open('csv1.csv', "r").readlines()
inf1 = open('csv2.csv', "r").readlines()

data = {}

# use double loop find out the same value of first column
# combine first csv file and certain columns in second csv file
for line in inf:
    part = line.split(',')
    a = part[0]
    for line1 in inf1:
        part1 = line1.split(',')
        b = part1[0]
        if a == b in data:    
            row = line + ',' + part1[2] + ',' + part1[3] + ',' + part1[4] + ',' + part1[5]    

# create a merge csv file        
with open('m_data.csv', 'w', newline = '') as f_out:
    fieldnames = ['ID', 'name', 'A1', 'A2', 'A3', 'D1', 'D2', 'D3', 'D4']
    writer = csv.DictWriter(f_out, fieldnames = fieldnames)

    writer.writeheader()

for c1, c2, c3, c4, c5, c6, c7, c8, c9 in data.items():
    writer.writerow({'ID': c1, 'name': c2, 'A1': c3, 'A2': c4, 'A3': c5,'D1': c6, 'D2': c7, 'D3': c8, 'D4':c9})

f_out.close()`

我认为问题可能在于“将第一个csv文件和第二个csv文件中的某些列组合起来”和“创建一个合并csv文件”之间的问题，但是我不知道如何修复它，或者还有其他方法吗？

非常感谢您的帮助！

Answer 1

这一解决办法应如预期的那样发挥作用：

import csv
from collections import OrderedDict


data = OrderedDict()

with open('temp.csv', 'r', encoding='utf-8', newline='') as infile1:
    reader1 = csv.reader(infile1)
    next(reader1, None)  # Skip header.
    for row in reader1:
        data[row[0]] = row  # id as the key and row as the value.

file2_NOs = set()  # A set to store the NO fields of file2.

with open('temp2.csv', 'r', encoding='utf-8', newline='') as infile2:
    reader2 = csv.reader(infile2)
    next(reader2, None)  # Skip header.
    for row in reader2:
        file2_NOs.add(row[0])  # Add the NO fields to this set.
        if row[0] in data:  # If the NO/id exists in file1...
            data[row[0]].extend(row[2:])  # ... extend the rows.

with open('temp3.csv', 'w', encoding='utf-8', newline='') as outfile:
    writer = csv.writer(outfile)
    writer.writerow(['ID', 'name', 'A1', 'A2', 'A3', 'D1', 'D2', 'D3', 'D4'])
    for row in data.values():
        if row[0] in file2_NOs:  # Write only if the number/id exists in file 2.
            writer.writerow(row)

我首先将file1的行添加到OrderedDict中，ids作为键，行作为值。然后，我遍历第二个文件，如果NO作为dict中的键存在，则扩展该行。我还添加了一个集(file2_NOs)，我在其中添加了NO，这样我就可以检查id是否也是有效的NOs，然后如果id和NO在这两个文件中都不存在，只需写行。