rapyd_db

An opinionated lightweight wrapper around various database backend drivers. The wrapper unifies connection lifecycle, audit-trail logging, and streaming across SQL and document stores while staying out of the way of the underlying driver's query semantics.

Supported backends:

Install

pip install "rapyd_db[mysql,mssql,mongo]"

Quickstart

from rapyd_db.backends.mysql import MySQL

db = MySQL(host="localhost", user="root", password="...")

# fire-and-fetch (default): returns (rows_affected, lastrowid, results)
affected, last_id, rows = db.execute(
    "INSERT INTO users(email) VALUES (%s)",
    ("alice@example.com",),
)

# streaming for large SELECTs (server-side cursor → no MemoryError)
for row in db.execute("SELECT * FROM events", stream=True):
    process(row)

Logging

Every execute() mints a UUID and prepends it to log lines so you can correlate all log records belonging to a single query:

INFO:rapyd_db.backends:f2e47d8... - Connecting to DB
INFO:rapyd_db.backends.mysql:f2e47d8... - Starting executing query at 2026-05-09 09:00:00
INFO:rapyd_db.backends.mysql:f2e47d8... - Query: SELECT * FROM events
INFO:rapyd_db.backends.mysql:f2e47d8... - 2,844,047 row(s) affected in 10 second(s)
INFO:rapyd_db.backends:f2e47d8... - Closed connection to DB

The package logger ships a NullHandler. Configure logging in your application:

import logging
logging.basicConfig(level=logging.INFO)

Mongo example

from rapyd_db.backends.mongo import Mongo

db = Mongo(host="localhost", username="me", password="...")
db.execute(
    "insert_many",
    [{"emp_no": 1, "salary": 60_000}],
    database="hr",
    collection="salaries",
)

The first positional after the operation name is forwarded to the underlying PyMongo collection method. Pass database= and collection= as keyword arguments. Operations like server_info operate on the client itself and do not require those.

MSSQL example

from rapyd_db.backends.mssql import MSSQL

db = MSSQL(host="localhost", user="sa", password="...")
affected, _, rows = db.execute("SELECT @@VERSION AS version")
print(rows[0]["version"])

Supported Python versions

Python 3.9, 3.10, 3.11, 3.12, 3.13.

Development

See CONTRIBUTING.md.

Security

See SECURITY.md for vulnerability disclosure.

License

Apache 2.0 — see LICENSE.