將包含列表的嵌套字典導入 csv

Question

我正在嘗試將以下數據導入 CSV：

{'test.foo.com': {'domain': 'foo.com','FQDN': 'test.foo.com', 'AS': 'AS1111', 'ressource_type': 'A', \
'nb_ip': '1', 'IP': '1.1.1.1', 'service': ['UNKNOWN'], 'port': '[443, 8443]'}}

我用這段代碼幾乎成功了：

#!/bin/python3

## Import ##
# Offical
import csv

### Main ###
if __name__ == '__main__':
  ## Variables
  csv_headers = ['domain', 'FQDN', 'AS', 'ressource_type', 'nb_ip', 'IP', 'service', 'port']
  final_data = {'test.foo.com': {'domain': 'foo.com','FQDN': 'test.foo.com', 'AS': 'AS1111', 'ressource_type': 'A', \
  'nb_ip': '1', 'IP': '1.1.1.1', 'service': ['UNKNOWN'], 'port': '[443, 8443]'}}

  # Open the csv file in "write mode"
  with open(file_name, mode='w') as file:
      # Prepare the writer to add a dict into the csv file
      csv_writer = csv.DictWriter(file, fieldnames=headers)

      # Write the columns header into the csv file
      csv_writer.writeheader()
      # Write the dict into the file
      for key, val in nest_dict.items():
          row = {'FQDN': key}
          row.update(val)
          csv_writer.writerow(row)

結果是：

domain,FQDN,AS,ressource_type,nb_ip,IP,service,port
foo.com,test.foo.com,AS1111,A,1,1.1.1.1,['UNKNOWN'],"[443, 8443]"

但我想：

domain,FQDN,AS,ressource_type,nb_ip,IP,service,port
foo.com,test.foo.com,AS1111,A,1,1.1.1.1,'UNKNOWN','443'
foo.com,test.foo.com,AS1111,A,1,1.1.1.1,'UNKNOWN','8443'

看到不同 ？ 我有一個“服務”列表（這里不需要處理）和一個“端口”列表。 如果“端口”列中有 1 個以上的端口，我需要為列表中的每個端口打印一個新行。

我正在努力做到這一點，因為我沒有完全理解這段代碼：

# Write the dict into the file
      for key, val in nest_dict.items():
          row = {'FQDN': key}
          row.update(val)
          csv_writer.writerow(row)

你能幫我處理一下嗎？

Answer 1

這將使用給定的數據給出所需的結果：

### Main ###
if __name__ == '__main__':
  ## Variables
  csv_headers = ['domain', 'FQDN', 'AS', 'ressource_type', 'nb_ip', 'IP', 'service', 'port']
  final_data = {'test.foo.com': {'domain': 'foo.com','FQDN': 'test.foo.com', 'AS': 'AS1111', 'ressource_type': 'A', \
  'nb_ip': '1', 'IP': '1.1.1.1', 'service': ['UNKNOWN'], 'port': '[443, 8443]'}}

  # Open the csv file in "write mode"
  with open('out.csv', mode='w') as file:
      # Prepare the writer to add a dict into the csv file
      csv_writer = csv.DictWriter(file, fieldnames=csv_headers)

      # Write the columns header into the csv file
      csv_writer.writeheader()
      # Write the dict into the file
      for key, val in final_data.items():
          row = {'FQDN': key}
          # Assume that service is always a list of one value and replace it with the one value
          # it contains.
          val['service'] = val.pop('service')[0]
          row.update(val)
          # Since the value of port is quoted it will be a string, but we wat a list. Remove the
          # value of 'port' from the dict and put it in 'port_string' (= '[443, 8443'')
          port_string = val.pop('port')
          # Remove the opening and closing brackets from the port_string (= '443, 8443').
          port_string = port_string.replace('[', '')
          port_string = port_string.replace(']', '')
          # Now we can split the string into a python list (= ['443', ' 8443'])
          port_list = port_string.split(',')
          # Write a csv row for each value in the port list
          for port in port_list:
              row['port'] = port.strip()
              csv_writer.writerow(row)

（順便說一句，原始帖子中的代碼沒有運行。此代碼包括使其運行的編輯。）

請注意，由於 'port' 的值被引用（與 'service' 的值不同）它將作為字符串讀入，因此必須首先將其轉換為列表。 如果刪除 [443, 8443] 周圍的單引號，則代碼的端口部分簡化為：

          port_list = val.pop('port')
          # Write a csv row for each value in the port list
          for port in port_list:
              row['port'] = port
              csv_writer.writerow(row)

另一個潛在的問題是“服務”。 它是一個列表，所以它可以有多個值嗎？ 如果是這樣，則需要修改代碼以解決該問題。

最后，我在此展示的代碼可能更加 Python 化，但希望確保它對於初學者來說盡可能具有可讀性。 一旦它完全按需要工作，它就可以變得更加 Pythonic。

Answer 2

所以這是最終的代碼。

不確定這是否是最好的方法，但對我來說看起來已經足夠了。

#!/bin/python3

## Import ##
# Offical
import csv

### Main ###
if __name__ == '__main__':
  ## Variables
  csv_headers = ['domain', 'FQDN', 'AS', 'ressource_type', 'nb_ip', 'IP', 'service', 'port']
  final_data = {'test.foo.com': {'domain': 'foo.com','FQDN': 'test.foo.com', 'AS': 'AS1111', 'ressource_type': 'A', \
  'nb_ip': '1', 'IP': '1.1.1.1', 'service': ['UNKNOWN'], 'port': '[443, 8443]'}}

# Open the csv file in "write mode"
with open(file_name, mode='w') as file:
    # Prepare the writer to add a dict into the csv file
    csv_writer = csv.DictWriter(file, fieldnames=csv_headers)

    # Write the columns header into the csv file
    csv_writer.writeheader()

    for key, val in final_data.items():
        # ?
        row = {'FQDN': key}
        # Update the row with all columns values
        row.update(val)
        # If service contains multiple elements it will transform the list into a string with each string separate by a space
        # If service contains just one element, it will transform the list into a string (no space) added
        row['service'] = ' '.join(val['service'])

        # Write a row for each value in the port list
        for port in val['port']:
            row['port'] = port
            csv_writer.writerow(row)

結果輸出：

domain,FQDN,AS,ressource_type,nb_ip,IP,service,port
foo.com,test.foo.com,AS1111,A,1,1.1.1.1,'UNKNOWN','443'
foo.com,test.foo.com,AS1111,A,1,1.1.1.1,'UNKNOWN','8443'

不要投票我的答案，我把它用於知識目的。 所有獎勵都應該發給@bartonstanley