mysql-wrap

mysql_wrap

A wrapper for Python Mysqldb with pandas functionality.

• Emiliano Lupo @HENN, June 2024
• Documentation https://pypi.org/project/mysqlwrap/
• License: GPL v2

Built on top of the SimpleMysql package available at https://github.com/knadh/simplemysql

Installation

with pip:

pip install mysql_wrap

from source:

python -m pip install .

Usage

For normal connection

from mysql_wrap import MysqlWrap, ConnectionOptions

options = ConnectionOptions(
	host="127.0.0.1",
	db="mydatabase",
	user="username",
	passwd="password",
	keep_alive=True # try and reconnect timedout mysql connections?
)

db = MysqlWrap(**options)

For SSL Connection

from mysql_wrap import MysqlWrap

db = MysqlWrap(
    host="127.0.0.1",
    db="mydatabase",
    user="username",
    passwd="password",
    ssl = {'cert': 'client-cert.pem', 'key': 'client-key.pem'},
    keep_alive=True # try and reconnect timedout mysql connections?
)

# insert a record to the <em>books</em> table
db.insert("books", {"type": "paperback", "name": "Time Machine", "price": 5.55, year: "1997"})

book = db.getOne("books", ["name"], ["year = 1997"])

print "The book's name is " + book.name

Utility methods

getDataTypefromDType(), setMySqlFieldName()

Pandas methods

getTable(), createTable(), SyncColumns(), insertFromDataFrame(), InsertOrUpdateFromDataFrame(), CreateInsertTable(), CreateUpdateTable()

regular Query methods

insert(), update(), insertOrUpdate(), describe(), delete(), getOne(), getAll(), lastId(), query(), tableExist()

insert(table, record{})

Inserts a single record into a table.

db.insert("food", {"type": "fruit", "name": "Apple", "color": "red"})
db.insert("books", {"type": "paperback", "name": "Time Machine", "price": 5.55})

update(table, row{}, condition[])

Update one more or rows based on a condition (or no condition).

# update all rows
db.update("books", {"discount": 0})

# update rows based on a simple hardcoded condition
db.update("books",
	{"discount": 10},
	["id=1"]
)

# update rows based on a parametrized condition
db.update("books",
	{"discount": 10},
	("id=%s AND year=%s", [id, year])
)

insertBatch(table, rows{})

Insert Multiple values into table.

# insert multiple values in table
db.insertBatch("books", [{"discount": 0},{"discount":1},{"discount":3}])

insertOrUpdate(table, row{}, key)

Insert a new row, or update if there is a primary key conflict.

# insert a book with id 123. if it already exists, update values
db.insertOrUpdate("books",
		{"id": 123, type": "paperback", "name": "Time Machine", "price": 5.55},
		"id"
)

getOne(table, fields[], where[], order[], limit[])

getAll(table, fields[], where[], order[], limit[])

Get a single record or multiple records from a table given a condition (or no condition). The resultant rows are returned as namedtuples. getOne() returns a single namedtuple, and getAll() returns a list of namedtuples.

book = db.getOne("books", ["id", "name"])

# get a row based on a simple hardcoded condition
book = db.getOne("books", ["name", "year"], ("id=1"))

# get multiple rows based on a parametrized condition
books = db.getAll("books",
	["id", "name"],
	("year > %s and price < %s", [year, 12.99])
)

# get multiple rows based on a parametrized condition with an order and limit specified
books = db.getAll("books",
	["id", "name", "year"],
	("year > %s and price < %s", [year, 12.99]),
	["year", "DESC"],	# ORDER BY year DESC
	[0, 10]			# LIMIT 0, 10
)

lastId()

Get the last insert id

# get the last insert ID
db.lastId()

lastQuery()

Get the last query executed

# get the SQL of the last executed query
db.lastQuery()

delete(table, fields[], condition[], order[], limit[])

Delete one or more records based on a condition (or no condition)

# delete all rows
db.delete("books")

# delete rows based on a condition
db.delete("books", ("price > %s AND year < %s", [25, 1999]))

query(table)

Run a raw SQL query. The MySQLdb cursor is returned.

# run a raw SQL query
db.query("DELETE FROM books WHERE year > 2005")

commit()

Insert, update, and delete operations on transactional databases such as innoDB need to be committed

# Commit all pending transaction queries
db.commit()

To run tests:

• add your test file to the tests/ folder

• import the modules you want to test using src.folder.module path

• run tests from terminal from the project root folder: python3 -m unittest tests.{test file}