Redcopy

A simple Python library for easily making a copy of a Redshift Database on the same or different instance.

Development

Create virtual environment

python3 -m venv venv
source ./venv/bin/activate

Install

pip install redcopy

Basic usage

Redcopy uses redshift-connector for interacting with the Redshift cluster

To start using, create a connection following the redshift-connector convention:

from redcopy import core
src = core.connect(host='your-connection-info',
                   ...)

Then pass this connection to fetch table DDL. This will return a dict in the form of: {'schema.tablename': 'CREATE TABLE ... '}

src_ddl = core.get_src_ddl(connection=src)

Then you can execute that DDL:

core.execute_ddl(connection=dest, ddl=src_ddl)

To export and import actual data:

core.unload_source(connection=src,
                   iam_role_arn='your-iam-role',
                   s3_path='s3://your-s3-path/')
core.load_target(connection=dest,
                 iam_role_arn='your-iam-role',
                 s3_path='s3://your-s3-path/')

Future work:

• Add automated tests
• Copy data using cross-database queries on ra3 instances
• Add schema DDL to extract
• Add user grants to extract
• Add view DDL to extract
• Add external table DDL to extract
• Add options to customize which tables/schemas are exported
• Format logging a bit better
• Export the generated DDL as a file

Feel free to make pull requests or issues!

Building a Release:

python3 -m build --wheel .
twine upload -r pypi dist/*