解析和搜索 CSV 文件

Question

我想解析一个联系人列表 CSV 文件，如下所示：

First Name  Last Name  Full Name                            Short Name  Phone Number
Jenny       Smith      CN=Jenny Smith/OU=CORP/O=COMPANY     jesmi       6468675309
Mary        Poppin     CN=Mary Poppins/OU=STORE/O=COMPANY   mapop       7005555578
Tony        Stark      CN=Tony Stark/OU=STORE/O=COMPANY     tostar      6007777798
Peter       Parker     CN=Peter Parker/OU=NEWS/O=COMPANY    pepar       5008889090

我希望能够搜索“全名”列并选择字符串“OU=STORE”并将所有包含“OU=STORE”的行移到一边，然后将其移动到它自己的名为“store.csv”的 csv 文件中。 .csv”。然后对“OU=CORP”和“OU=NEWS”重复相同的过程。

---

这是我希望输出的样子：

一旦该过程完成，Store.csv 应该只包含此信息。

First Name  Last Name  Full Name                            Short Name  Phone Number
Mary        Poppin     CN=Mary Poppins/OU=STORE/O=COMPANY   mapop       7005555578
Tony        Stark      CN=Tony Stark/OU=STORE/O=COMPANY     tostar      6007777798

corp.csv

First Name  Last Name  Full Name                            Short Name  Phone Number
Jenny       Smith      CN=Jenny Smith/OU=CORP/O=COMPANY     jesmi       6468675309

news.csv

First Name  Last Name  Full Name                            Short Name  Phone Number    
Peter       Parker     CN=Peter Parker/OU=NEWS/O=COMPANY    pepar       5008889090

我有一个到目前为止我所做的小脚本，但我不确定最后要做什么：

import pandas as pd
import csv

#this is the source folder    
source_dir = 'C:/Users/username/documents/contacts/contactslist.csv'

#this is the folder where I want to move the parsed data.
store_target_dir = 'C:/Users/username/documents/contacts/store/'
corp_target_dir = 'C:/Users/username/documents/contacts/corp/'
news_target_dir = 'C:/Users/username/documents/contacts/news/'

col_list = ["Full Name"]

store = 'OU=STORE'
corp = 'OU=CORP'
news = 'OU=NEWS'

#When it comes time to move the data to their folders with their csv name
csvName = store_target_dir + "/" + "store.csv"
csvName2 = corp_target_dir + "/" + "corp.csv"
csvName3 = news_target_dir + "/" +"news.csv"

#opening the file
file = open(source_dir)

#reading the csv file
df = pd.read_csv(file)

Answer 1

要过滤您的 DataFrame，您可以执行以下操作：

# key is the value you are looking for, e.g. 'OU=STORE'    
indices = [key in value for value in df['File Name']]
subset = df[indices]

indices 是一个布尔列表，指示一行是否包含key

Answer 2

您可以提取OU= 值并将其添加为另一列。然后，.unique() 可用于确定 3 个可能的值，然后根据该值创建每个 CSV。例如：

import pandas as pd

df = pd.read_csv('contactslist.csv', dtype={'Phone Number': str})
df['file'] = df['Full Name'].str.extract(r'OU=(\S+)/')

for key in df['file'].unique():
    df_filtered = df.loc[df['file'] == key]
    df_filtered = df_filtered.drop(['file'], axis=1)
    df_filtered.to_csv(f"{key}.csv", index=False)      

Answer 3

尽量不对代码做太多修改，解决方案可能如下所示：

import os, os.path
import csv

#this is the source folder
original_contacts_filename = r'C:\Users\username\documents\contacts\contactslist.csv'

#this is the folder where I want to move the parsed data.
store_target_dir = r'C:\Users\username\documents\contacts\store'
corp_target_dir = r'C:\Users\username\documents\contacts\corp'
news_target_dir = r'C:\Users\username\documents\contacts\news'

os.makedirs(store_target_dir, exist_ok=True)
os.makedirs(corp_target_dir, exist_ok=True)
os.makedirs(news_target_dir, exist_ok=True)

store = 'OU=STORE'
corp = 'OU=CORP'
news = 'OU=NEWS'

#When it comes time to move the data to their folders with their csv name
csv_name = os.path.join(store_target_dir, "store.csv")
csv_name2 = os.path.join(corp_target_dir, "corp.csv")
csv_name3 = os.path.join(news_target_dir, "news.csv")

with (
    open(original_contacts_filename, newline='') as original_contacts_file,
    open(csv_name, mode='w', newline='') as csv_file, 
    open(csv_name2, mode='w', newline='') as csv_file2, 
    open(csv_name3, mode='w', newline='') as csv_file3):

    original_contacts = csv.DictReader(original_contacts_file)
    
    store_destination = csv.writer(csv_file)
    corp_destination = csv.writer(csv_file2)
    news_destionation = csv.writer(csv_file3)

    output_headers = ('First Name', 'Last Name', 'Full Name', 'Short Name', 'Phone Number')
    store_destination.writerow(output_headers)
    corp_destination.writerow(output_headers)
    news_destionation.writerow(output_headers)

    for current_contact in original_contacts:
        if store in current_contact['Full Name']:
            output_destination = store_destination
        elif corp in current_contact['Full Name']:
            output_destination = corp_destination
        elif news in current_contact['Full Name']:
            output_destination = news_destionation
        else:
            output_destination = None
        
        if output_destination is not None:
            output_destination.writerow(current_contact[column] for column in output_headers)

但是我们可以看到很多重复的东西，它们通常很臭。我们可以这样简化代码：

import os, os.path
import csv
import re

original_contacts_filename = 'contactslist.csv'

source_directory = r'C:\Users\username\documents\contacts'

corporate_units_expected = ('store', 'corp', 'news')

target_directory = r'C:\Users\username\documents\contacts'
target_files_info = {
    current_unit: (
        current_name := os.path.join(target_directory, current_unit, f'{current_unit}2.csv'),
        open(current_name, 'w', newline='')
    )
    for current_unit in corporate_units_expected
}

for current_target, _ in target_files_info.values():
    os.makedirs(os.path.dirname(current_target), exist_ok=True)
    
matcher = re.compile(r'OU=([ 0-9a-zA-Z]+)')

with (
    open(original_contacts_filename, newline='') as original_contacts_file,
    target_files_info['store'][1], target_files_info['corp'][1], target_files_info['news'][1]
):
    original_contacts = csv.DictReader(original_contacts_file)

    writers = { 
        current_unit: csv.writer(current_target[1])
        for (current_unit, current_target) in target_files_info.items()
    }

    output_headers = ('First Name', 'Last Name', 'Full Name', 'Short Name', 'Phone Number')

    for current_writer in writers.values():
        current_writer.writerow(output_headers)
    
    for current_contact in original_contacts:
        if match_found := matcher.search(current_contact['Full Name']):
            current_writer = writers[match_found[1].lower()]
            current_writer.writerow(current_contact[column] for column in output_headers)

我们也可以有一个例子，我们事先不知道将这些条目分类到多少个文件中，但它变得更加复杂，因为我们不能直接使用with 语句。