haip-database 0.1.4

A generic database interface.

haip.config

haip.database is a minimalistic async database interface for Python 3

Features

• minimalistic: query and do
• db by name: reference your database (sessions) per config-name
• sql templates: SQL seperated from code. placeholders are escaped automatically.
• db pools: db connection pooling
• dict short notation: row.fieldname == row['fieldname']
• db supported: mysql, oracle, mssql, (ongoing)

Getting Started

Installing

pip install haip-database

or from source:

git clone https://github.com/haipdev/database.git

Usage

Config-files

databases:
    mysql_test_db:
        type: mysql
        host: host
        port: port
        database: database
        username: username
        password: password
        max_connections: 3
        max_idle_connections: 3

    oracle_test_db:
        type: oracle
        host: host
        port: port
        username: username
        password: password
        service_name: service_name

    mssql_test_db:
        type: mssql
        driver: driver
        host: host
        port: port
        database: TEST
        username: username
        password: password

The database is identified by the section-name (e.g. 'mysql_test_db').

Common options

• type: mysql | oracle
• host: hostname/IP of the database server - default='127.0.0.1'
• port: port the database server is listening at - default depends on the type (mysql=3306, oracle=1521)
• username: username for login
• password: password for login
• autocommit: true | false - default=true
• max_connections: max number of open connections for this database (otherwise you will get DatabasePoolExhaustedExcpetions)
• max_idle_connections: max number of open idle connections

Mysql options

• database: name of the database on the database-server

Prerequisite: python module "mysql_connector"

Oracle options

• service_name: service-name (higher priority then SID)
• sid: sid

Prerequisite: python module "cx_Oracle"

MSSQL options

• driver: driver definition string e.g. ""{SQL Server}"

Prerequisite: python module "pyodbc"

Example

/path-to-my-config-dir/databases.yml

databases:
    testdb:
        type: mysql
        username: testuser
        host: 127.0.0.2

(optionally you can place e.g. the passwords in seperate files):

/path-to-my-config-dir/dev/databases.yml

databases:
    testdb:
        password: testpassword

Functions

Query

async def query(db_name, query_template, *values, **args)

• db_name: the name of the database as defined in the configuration files (main section "databases" - in the example above e.g. "testdb")
• query_template: the filename of the template-file containing the SQL query. This file must have the suffix ".sql".
• *values: values for query placeholders
• **args: the template-vars for the query_template jinja template.

This function returns an array of arrays (array of rows).

Query example

/path-to-my-config-dir/queries/firstname.sql

SELECT firstname, lastname
    FROM users
    WHERE firstname = '{{ firstname }}

import haip.config as config
import haip.database as database

config.load('/path-to-my-config-dir', 'dev')
rows = await database.query('testdb', 'queries/firstname.sql', firstname='Reinhard')

for row in rows:
    firstname = row[0]
    lastname = row[1]

await database.shutdown()

shortcuts

async def query_assoc(db_name, query_template, **args)

Like "query" but returns an array of dicts. e.g. rows[0]['firstname'] or rows[0].firstname

async def query_first(db_name, query_template, **args)

Like "query" but returns only the first row as dict. e.g. row['firstname'] or row.firstname. If no rows found None will be returned.

Insert/Updates

async def do(db_name, query_template, **args)

Arguments like "query". This function returns the number of rows effected by this statement.

Do example

/path-to-my-config-dir/queries/update.sql

UPDATE users
    SET firstname='Test'
WHERE lastname = '{{ lastname }}

import haip.config as config
import haip.database as database

config.load('/path-to-my-config-dir', 'dev')
changes = await database.do('testdb', 'queries/update.sql', lastname='Hainz')

print(f'effected rows: {changes}')
await database.shutdown()

Procedures

async def call(db_name, procedure)

• db_name: as above
• procedure: the name of the procedure to be called

Pool shutdown

async def shutdown()

Close all open connections in the connection pool normally used before closing your application. If you do not close the open connections in the connection pool you will see some warnings on your db servers. So calling this function is not realy necessary but beautifies your code.

Running the tests

Tests are written using pytest and located in the "tests" directory.

pytest tests

Contributing

Feel free to use and enhance this project. Pull requests are welcome.

Authors

• Reinhard Hainz - Initial work - haipdev

License

This project is licensed under the MIT License - see the LICENSE file for details

Dependencies

• haip-config
• haip-template