Automatically add partitions to PostgreSQL tables

Author: Letu
Pseudonym: lottu
Personal profile: at present, I am engaged in the design and development of database architecture in a game company, and have led the company to go to “O” related workhttps://www.cnblogs.com/lottu

PostgreSQL introduces the feature of “partition” table, which liberates the tedious and inefficient implementation of partition table by using “table inheritance” + “trigger”. When adding partitions, SQL is executed manually.
Purpose of demonstration: use Python to automatically add partitions to PostgreSQL tables.
Python version: Python 3+
pip3 install psycopg2

1、 Configure data sources

database.ini Files: recording database connection parameters

[adsas]
host=192.168.1.201
database=adsas
user=adsas
password=adsas123
port=5432

[test]
host=192.168.1.202
database=adsas
user=adsas
password=adsas123
port=5432

2、 Config script

config.py File: the following config() function reads database.ini File and returns the connection parameters. The config() function is located in the config.py In the file

#!/usr/bin/python3
from configparser import ConfigParser
 
def config(section ,filename='database.ini'):
    # create a parser
    parser = ConfigParser()
    # read config file
    parser.read(filename)
 
    # get section, default to postgresql
    db = {}
    if parser.has_section(section):
        params = parser.items(section)
        for param in params:
            db[param[0]] = param[1]
    else:
        raise Exception('Section {0} not found in the {1} file'.format(section, filename))
 
    return db

3、 Create subtable script

pg_ add_ partition_ table.py File: where create_ The table function is to create a sub table SQL. Where parameters

Parameter name	meaning
db	Point to database
table	Main table
sub_table	The name of the child table you are about to create
start_date	Start value of range boundary
end_date	End value of range boundary

#!/usr/bin/python3

import psycopg2
from config import config

#example: create table tbl_game_android_step_log_2021_07 PARTITION OF tbl_game_android_step_log FOR VALUES FROM ('2021-07-01') TO ('2021-08-01');
def create_table(db, table, sub_table, start_date, end_date):
    """ create subtable in the PostgreSQL database"""
    command = "create table {0} PARTITION OF {1} FOR VALUES FROM ('{2[0]}') TO ('{2[1]}');".format(sub_table, table, (start_date, end_date)) 
    conn = None
    try:
        # read the connection parameters
        params = config(section = db)
        # connect to the PostgreSQL server
        conn = psycopg2.connect(**params)
        cur = conn.cursor()
        # create table one by one
        cur.execute(command)
        # close communication with the PostgreSQL database server
        cur.close()
        # commit the changes
        conn.commit()
    except (Exception, psycopg2.DatabaseError) as error:
        print(error)
    finally:
        if conn is not None:
            conn.close()

4、 Executive document main.py
main.py : main file; generate partition table by executing main.
Example:

#!/usr/bin/python3
import datetime
from datetime import date
from dateutil.relativedelta import *
from pg_add_partition_table import create_table

#Get the 1st day of the next month
def get_next_month_first_day(d):
    return date(d.year + (d.month == 12), d.month == 12 or d.month + 1 , 1)
    
def create_sub_table(db, table):
    # Get current date
    d1 = date.today()
    # Get next month's date
    d2 = d1 + relativedelta(months=+1)
    # Get the 1st day of the next month;As the starting value of the partitioned table
    start_date = get_next_month_first_day(d1)
    # Gets the 1st of the next two months as the end value of the partitioned table
    end_date = get_next_month_first_day(d2)
    # get sub table name
    getmonth =  datetime.datetime.strftime(d2, '%Y_%m')
    sub_table = table + '_' + getmonth
    create_table(db, table, sub_table, start_date, end_date)
    
if __name__ == '__main__':
    create_sub_table('test', 'tbl_game_android_step_log');

The above example is a separate TABLE tbl_ game_ android_ step_ Log; create partition; if more than one table; use for statement to process
Multi table operation

for table in ['tbl_game_android_step_log', 'tbl_game_android_game_log','tbl_game_android_pay_log']:
        create_sub_table('test', table);

Before the presentation:

adsas=> select * from  pg_partition_tree('tbl_game_android_step_log');
               relid               |        parentrelid        | isleaf | level 
-----------------------------------+---------------------------+--------+-------
 tbl_game_android_step_log         |                           | f      |     0
 tbl_game_android_step_log_2020_12 | tbl_game_android_step_log | t      |     1
(2 rows)

After the presentation:

adsas=> select * from  pg_partition_tree('tbl_game_android_step_log');
               relid               |        parentrelid        | isleaf | level 
-----------------------------------+---------------------------+--------+-------
 tbl_game_android_step_log         |                           | f      |     0
 tbl_game_android_step_log_2020_12 | tbl_game_android_step_log | t      |     1
 tbl_game_android_step_log_2021_01 | tbl_game_android_step_log | t      |     1

Partition key: RANGE (visit_time)
Partitions: tbl_game_android_step_log_2020_12 FOR VALUES FROM ('2020-12-01 00:00:00') TO ('2021-01-01 00:00:00'),
            tbl_game_android_step_log_2021_01 FOR VALUES FROM ('2021-01-01 00:00:00') TO ('2021-02-01 00:00:00')

5、 Join timing task
N/A

For more wonderful content, please pay attention to the following platforms and websites:

Official account of China PostgreSQL branch (technical articles and technical activities):
PostgreSQL branch of Open Source Software Alliance

China PostgreSQL technical Q & a community:
www.pgfans.cn

Official website of China PostgreSQL branch:
www.postgresqlchina.com