MysqldbPythonWrapper

mysqldb_wrapper is a python module that will help you easily communicate with your database and provides encryption with Fernet using the cryptography module. This module is inspired by SQLAlchemy but is even simpler to use and has less limitations (but is also less complete).

IMPORTANT: This module is only made to use with simple mysql queries. You can still use it with more complex queries, but no interface is provided for those.

Installation

This module is available via pip:

pip install mysqldb_wrapper

Usage

The table

First, you need to create a table. Here's an example:

from mysqldb_wrapper import Base, Id

class Test(Base):
    __tablename__ = "test"

    id = Id()
    another_table_id = Id()
    hashed = bytes()
    number = int(1)
    string = str("string")
    boolean = bool(True)

To explain in details:

• id = Id() is, as its name implies, the database object id. Always put one Id.
• another_table_id = Id() is the id of another table in the database.
• hashed = bytes() will be a hashed string, always declare it empty.
• number = int(1) will be a int with a default value 1.
• string = str("string") will be a string with a default value string.
• boolean = bool(True) will be a boolean with a default value boolean.

Easy, right ?

Note: This implementation does not support complex types, so keep your table simple ;)

The database

You need to create your database by hand Although this module will not create the database for you, it will create the tables automatically !

Note: keep in mind that if your table change (remove fields, change field types, ...), you will need to remove/update them by hand.

To initialize a session with the database, use:

from mysqldb_wrapper import Session

session = Session(DB_USERNAME, DB_PASSWORD, DB_TEST, FERNET_KEY)

So here we have:

• DB_USERNAME is the username of the mysql user
• DB_PASSWORD is the password of the mysql user
• DB_TEST is the name of the database
• FERNET_KEY is the encryption key. You can create one with Fernet.generate_key(), but keep in mind to always use the same one. (So generate one by hand then keep it somewhere safe)

And now we have all we need to use the mysql database.

Operations

Adding an object

Remember your table ? Just create a new object from it and use the add method of the session with it. Also, the table objects supports **kwargs.

obj = Test(hashed="abcd", string="word")
session.add(obj)

And obj now automatically contains the id generated by the database. Also, this method returns the object in case you want to chain methods.

Updating an object

Using the obj from before:

obj.number = 2553166
obj.string = "a sentence"
obj.boolean = True
session.update(obj)

This method also returns the object.

Deleting an object

And again:

session.delete(obj)

obj now is not in the database anymore but you can still continue to use it, just remember it doesn't have an id (set to 0). Like the others, this method returns the object.

Querying objects

We have more to talk about in this section.

obj = session.query(Test).first()

This one is easy. You put the table you want in query() and you want to get the first result.

obj_list = session.query(Test).all()

Same than before, except this time you retrieve all the results. It is guaranteed to return a list, even if empty.

Now the tricky part:

obj = session.query(Test).where(Test.id == 2).first()

This time, we are still querying in the Test table, but we want a specific result, the one where the id is equal to 2. You don't have to worry about doing anything more, everything is handled by the Base class your Test table inherited from.

Remember we had a hash in our table ? Don't worry, it is handled as well. How does it work ? Like the others field. =)

obj = session.query(Test).where(Test.hashed == "abcd").first()

You can also chain the where methods, use variables and query all objects:

a_string = "something"
list_obj = session.query(Test).where(Test.another_table_id == 2).where(Test.hashed == a_string).all()

IMPORTANT: The where method only works with equality as of now. Also, you can only query by Id and hashes.