Skip to main content

CSV

Reading CSV files

  • csv.reader(<file-object>, delimiter=':') : input is CSV file. the parameter delimiter is optional

csv_file.txt

Sabrina Green,802-867-5309,System Administrator
Eli Jones,684-3481127,IT specialist
Melody Daniels,846-687-7436,Programmer
Charlie Rivera,698-746-3357,Web Developer
import csv
 f = open("csv_file.txt")
 csv_f = csv.reader(f)
 for row in csv_f:
     name, phone, role = row
     print("Name: {}, Phone: {}, Role: {}".format(name, phone, role))
f.close()

Output:

Name: Sabrina Green, Phone: 802-867-5309, Role: System Administrator
Name: Eli Jones, Phone: 684-3481127, Role: IT specialist
Name: Melody Daniels, Phone: 846-687-7436, Role: Programmer
Name: Charlie Rivera, Phone: 698-746-3357, Role: Web Developer

Generating CSV

  • csv.writer() : input is a list with sublist, for example [[col1, col2, col3], [col1, col2, col3]] 
  • .writerow() : 一次寫一筆
  • .writerows() : 一次寫多筆
import csv

hosts = [["workstation.local", "192.168.25.46"],["webserver.cloud", "10.2.5.6"]]
with open('hosts.csv', 'w') as hosts_csv:
    writer = csv.writer(hosts_csv)
    writer.writerows(hosts)

With list

Reading a CSV with the list

user_emails.csv

Full Name, Email Address
Blossom Gill, blossom@xyz.edu
Hayes Delgado, nonummy@utnisia.com
Petra Jones, ac@xyz.edu
Oleg Noel, noel@liberomauris.ca
Ahmed Miller, ahmed.miller@nequenonquam.co.uk
Macaulay Douglas, mdouglas@xyz.edu
Aurora Grant, enim.non@xyz.edu
  • list(csv.reader(file)) : 用 list() 函式將 CSV 內容轉成 List 格式，不使用函式也行，預設格式就是 List
  • user_data_list[1:] : 不包含第一行標題的所有內容
  • data[1].strip() : CSV 第 2 欄資料且移除前後空白字元
user_email_list = []

with open(csv_file_location, 'r') as f:
    user_data_list = list(csv.reader(f))
    user_email_list = [data[1].strip() for data in user_data_list[1:]]

With dictionary

Reading a CSV with the dictionary

  • csv.DictReader() : input is a CSV file, 預設第一行為標題行
# software.csv
# name,version,status,users
# MailTree,5.34,production,324
# CalDoor,1.25.1,beta,22
# Chatty Chicken,0.34,alpha,4

with open('software.csv') as software:
    reader = csv.DictReader(software)
    for row in reader:
        print(("{} has {} users").format(row["name"], row["users"]))

# Output:
# MailTree has 324 users
# CalDoor has 22 users
# Chatty Chicken has 4 users

Writing a CSV with the dictionary

  • csv.DictWriter(<file-object>, fieldnames=<column-list>) : input is a dictionary
  • .writerheader() : 寫標題行
  • .writerows() : input is a list with multiple dictionaries
users = [ {"name": "Sol Mansi", "username": "solm", "department": "IT infrastructure"}, 
 {"name": "Lio Nelson", "username": "lion", "department": "User Experience Research"}, 
  {"name": "Charlie Grey", "username": "greyc", "department": "Development"}]
keys = ["name", "username", "department"]
with open('by_department.csv', 'w') as by_department:
    writer = csv.DictWriter(by_department, fieldnames=keys)
    writer.writeheader()
    writer.writerows(users)

# by_department.csv:
# Name,username,department
# Sol Mansi,solm, IT infrastructure
# Lio Nelson,lion,User Experience Researcher
# Charlie Grey,greyc,Development

Back to top