sqlalchemy, read from mysql and write to memory

Question

I have a program communicating with a mysql database, which I should avoid to modify it until everything works fine.

So I already have a way to create a fake in-memory database (using sqlalchemy+sqlite), but the problem is that I have to create the schema and populate the database all the time.

As for example

Table(
    'builddefinitions', meta,
    Column('idBuildDefinitions', String(20)),
    Column('BuildType', String(20)),
)

Is it possible instead to

• create a mysql engine for loading the tables and schema
• create a sqlite in-memory engine and load all the schema and data in it
• use that database to do all the updates safely

Is there a way to do this is sqlalchemy?

Answer 1

You do not have an straight way to do it. Take a look at this dumper: http://www.tylerlesmann.com/2009/apr/27/copying-databases-across-platforms-sqlalchemy/

I don't know how to maintain both updated, if it is what you want.

Answer 2

I would clone the database. This is why one usually has a production environment and a separate development environment (and in some cases testing and pre-production).