Metabase Export/Import

This python library allows to export and import a community version instance of Metabase

Example Scripts

Two scripts are provided to import and export fields, cards and dashboards of a specific database configuration of metabase:

# Export each component separately
python3 metabase_export.py [cards|fields|metrics|dashboards]

# Export everything together
python3 metabase_export.py all

# Export everything together and store the raw JSON without parsing it
python3 metabase_export.py --raw all

The script produces 3 files for each exported elements (the name of the database is user as prefix) : my_database_fields_exported.csv, my_database_cards_exported.json and my_database_dashboard_exported.json

# Import each component separately
python3 metabase_import.py [cards|fields|metrics|dashboards]

# Export everything together
python3 metabase_import.py all 

The script imports from 3 files, one for each elements : my_database_fields_forimport.csv, my_database_cards_forimport.json and my_database_dashboard_forimport.json

Configuration

Available flags for the scripts:

flag	description
--verbose	increase output verbosity of each command
--dry-run	run the script without making POST/PUT requests
--raw	store the raw JSON without parsing it

It is recommended to predefine all the following environment variables in a .env file:

env var	description
MB_DATA_DIR	Directory to use for the export/import operations
MB_EXPORT_HOST	Source Metabase instance (https://<url>/api/)
MB_EXPORT_USERNAME	Admin username
MB_EXPORT_PASSWORD	Admin username password
MB_EXPORT_DB	The database name to export
MB_IMPORT_HOST	Destination Metabase instance (https://<url>/api/)
MB_IMPORT_USERNAME	Admin username
MB_IMPORT_PASSWORD	Admin username password
MB_IMPORT_DB	The name of the same database in the destination instance

Library calls

database creation/deletion

import metabase

# connect to metabase
ametabase = metabase.MetabaseApi("http://localhost:3000/api/", "metabase_username", "metabase_password")

# add a sqlite database located at /path/to/database.sqlite. The metabase associated name is my_database
ametabase.create_database("my_database", 'sqlite', {"db":"/path/to/database.sqlite"})

#ametabase.delete_database('my_database')

users and permisssions

ametabase.create_user("[email protected]", "the_password", {'first_name': 'John', 'last_name': 'Doe'})

# Add a group and associate it with our new user
ametabase.membership_add('[email protected]', 'a_group')

# allow read data and create interraction with my_database for users members of our new group (a_group)
ametabase.permission_set_database('a_group', 'my_database', True, True)

collections and permissions

# create a collection and its sub collection
ametabase.create_collection('sub_collection', 'main_collection')

# allow write right on the new collections to the membres of a_group
ametabase.permission_set_collection('main_collection', 'a_group', 'write')
ametabase.permission_set_collection('sub_collection', 'a_group', 'write')

schema

# export and import the schema of fields
ametabase.export_fields_to_csv('my_database', 'my_database_fields.csv')
ametabase.import_fields_from_csv('my_database', 'my_database_fields.csv')

cards and dashboards

ametabase.export_cards_to_json('my_database', 'my_database_cards.json')
ametabase.export_dashboards_to_json('my_database', 'my_database_dashboard.json')

ametabase.import_cards_from_json('my_database', 'my_database_cards.json')
ametabase.import_dashboards_from_json('my_database', 'my_database_dashboard.json')

Development

Development of this repository is done with Poetry.

Install dependencies

make install

Test

The tests are running with pytest

make test

# Run a tests matching an expression TEST_FUNC=<expression>
TEST_FUNC="invalid" make test