Python 3 進階 —— 使用 PyMySQL 操作 MySQL

PyMySQL 是一個純 Python 實現的 MySQL 客戶端操作庫，支持事務、存儲過程、批量執行等。

PyMySQL 遵循 Python 數據庫 API v2.0 規范，並包含了 pure-Python MySQL 客戶端庫。

安裝

pip install PyMySQL

創建數據庫連接

import pymysql

connection = pymysql.connect(host='localhost',
                             port=3306,
                             user='root',
                             password='root',
                             db='demo',
                             charset='utf8')

參數列表：

參數	描述
host	數據庫服務器地址，默認 localhost
user	用戶名，默認為當前程序運行用戶
password	登錄密碼，默認為空字符串
database	默認操作的數據庫
port	數據庫端口，默認為 3306
bind_address	當客戶端有多個網絡接口時，指定連接到主機的接口。參數可以是主機名或IP地址。
unix_socket	unix 套接字地址，區別於 host 連接
read_timeout	讀取數據超時時間，單位秒，默認無限制
write_timeout	寫入數據超時時間，單位秒，默認無限制
charset	數據庫編碼
sql_mode	指定默認的 SQL_MODE
read_default_file	Specifies my.cnf file to read these parameters from under the [client] section.
conv	Conversion dictionary to use instead of the default one. This is used to provide custom marshalling and unmarshaling of types.
use_unicode	Whether or not to default to unicode strings. This option defaults to true for Py3k.
client_flag	Custom flags to send to MySQL. Find potential values in constants.CLIENT.
cursorclass	設置默認的游標類型
init_command	當連接建立完成之后執行的初始化 SQL 語句
connect_timeout	連接超時時間，默認 10，最小 1，最大 31536000
ssl	A dict of arguments similar to mysql_ssl_set()'s parameters. For now the capath and cipher arguments are not supported.
read_default_group	Group to read from in the configuration file.
compress	Not supported
named_pipe	Not supported
autocommit	是否自動提交，默認不自動提交，參數值為 None 表示以服務器為准
local_infile	Boolean to enable the use of LOAD DATA LOCAL command. (default: False)
max_allowed_packet	發送給服務器的最大數據量，默認為 16MB
defer_connect	是否惰性連接，默認為立即連接
auth_plugin_map	A dict of plugin names to a class that processes that plugin. The class will take the Connection object as the argument to the constructor. The class needs an authenticate method taking an authentication packet as an argument. For the dialog plugin, a prompt(echo, prompt) method can be used (if no authenticate method) for returning a string from the user. (experimental)
server_public_key	SHA256 authenticaiton plugin public key value. (default: None)
db	參數 database 的別名
passwd	參數 password 的別名
binary_prefix	Add _binary prefix on bytes and bytearray. (default: False)

執行 SQL

• cursor.execute(sql, args) 執行單條 SQL

  # 獲取游標
  cursor = connection.cursor()
  
  # 創建數據表
  effect_row = cursor.execute('''
  CREATE TABLE `users` (
    `name` varchar(32) NOT NULL,
    `age` int(10) unsigned NOT NULL DEFAULT '0',
    PRIMARY KEY (`name`)
  ) ENGINE=InnoDB DEFAULT CHARSET=utf8
  ''')
  
  # 插入數據(元組或列表)
  effect_row = cursor.execute('INSERT INTO `users` (`name`, `age`) VALUES (%s, %s)', ('mary', 18))
  
  # 插入數據(字典)
  info = {'name': 'fake', 'age': 15}
  effect_row = cursor.execute('INSERT INTO `users` (`name`, `age`) VALUES (%(name)s, %(age)s)', info)
  
  connection.commit()
  

• executemany(sql, args) 批量執行 SQL

  # 獲取游標
  cursor = connection.cursor()
  
  # 批量插入
  effect_row = cursor.executemany(
      'INSERT INTO `users` (`name`, `age`) VALUES (%s, %s) ON DUPLICATE KEY UPDATE age=VALUES(age)', [
          ('hello', 13),
          ('fake', 28),
      ])
  
  connection.commit()
  

注意：INSERT、UPDATE、DELETE 等修改數據的語句需手動執行connection.commit()完成對數據修改的提交。

獲取自增 ID

cursor.lastrowid

查詢數據

# 執行查詢 SQL
cursor.execute('SELECT * FROM `users`')

# 獲取單條數據
cursor.fetchone()

# 獲取前N條數據
cursor.fetchmany(3)

# 獲取所有數據
cursor.fetchall()

游標控制

所有的數據查詢操作均基於游標，我們可以通過cursor.scroll(num, mode)控制游標的位置。

cursor.scroll(1, mode='relative') # 相對當前位置移動
cursor.scroll(2, mode='absolute') # 相對絕對位置移動

設置游標類型

查詢時，默認返回的數據類型為元組，可以自定義設置返回類型。支持5種游標類型：

• Cursor: 默認，元組類型
• DictCursor: 字典類型
• DictCursorMixin: 支持自定義的游標類型，需先自定義才可使用
• SSCursor: 無緩沖元組類型
• SSDictCursor: 無緩沖字典類型

無緩沖游標類型，適用於數據量很大，一次性返回太慢，或者服務端帶寬較小時。源碼注釋：

Unbuffered Cursor, mainly useful for queries that return a lot of data, or for connections to remote servers over a slow network.

Instead of copying every row of data into a buffer, this will fetch rows as needed. The upside of this is the client uses much less memory, and rows are returned much faster when traveling over a slow network
or if the result set is very big.

There are limitations, though. The MySQL protocol doesn't support returning the total number of rows, so the only way to tell how many rows there are is to iterate over every row returned. Also, it currently isn't possible to scroll backwards, as only the current row is held in memory.

創建連接時，通過 cursorclass 參數指定類型：

connection = pymysql.connect(host='localhost',
                             user='root',
                             password='root',
                             db='demo',
                             charset='utf8',
                             cursorclass=pymysql.cursors.DictCursor)

也可以在創建游標時指定類型：

cursor = connection.cursor(cursor=pymysql.cursors.DictCursor)

事務處理

• 開啟事務
  connection.begin()

• 提交修改
  connection.commit()

• 回滾事務
  connection.rollback()

防 SQL 注入

• 轉義特殊字符
  connection.escape_string(str)

• 參數化語句
  支持傳入參數進行自動轉義、格式化 SQL 語句，以避免 SQL 注入等安全問題。

# 插入數據(元組或列表)
effect_row = cursor.execute('INSERT INTO `users` (`name`, `age`) VALUES (%s, %s)', ('mary', 18))

# 插入數據(字典)
info = {'name': 'fake', 'age': 15}
effect_row = cursor.execute('INSERT INTO `users` (`name`, `age`) VALUES (%(name)s, %(age)s)', info)

# 批量插入
effect_row = cursor.executemany(
    'INSERT INTO `users` (`name`, `age`) VALUES (%s, %s) ON DUPLICATE KEY UPDATE age=VALUES(age)', [
        ('hello', 13),
        ('fake', 28),
    ])

參考資料

• Python中操作mysql的pymysql模塊詳解
• Python之pymysql的使用

個人博客同步地址：
https://shockerli.net/post/python3-pymysql/