Skip to content

Latest commit

 

History

History
144 lines (86 loc) · 5.02 KB

USAGE.rst

File metadata and controls

144 lines (86 loc) · 5.02 KB
Raw

Usage

The Brazil Data Cube Catalog system uses the Alembic Environment to upgrade/downgrade versions. If you would like to use it as experimental environments (dev only), you may consider to follow Development Mode.

For production environment, we strongly recommend you to adopt alembic migration way to be able to support improvements.

Development Mode

Note

This step is not recommended for production environments. It will not generate alembic migration tree.

For development purposes, we have prepared a command line bdc-db db create-schema to create all definitions for BDC-Catalog:

export SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb"
bdc-db db init
bdc-db db create-namespaces
bdc-db db create-extension-postgis
bdc-db db create-schema

Production Mode

Creating database definition

1. Create a PostgreSQL database and enable the required extensions (postgis, hstore):

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db db init

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db db create-namespaces

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db db create-extension-postgis

2. After that, run BDC-DB command to prepare the Brazil Data Cube data model:

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db alembic upgrade

3. The BDC-Catalog uses PostgreSQL Triggers along database definition. Use the following command line to create default triggers:

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db db create-triggers

Note

You can also check the available triggers with command:

bdc-db db show-triggers

Available triggers in "bdc_catalog.triggers"
        -> /home/raphael/devel/github/brazil-data-cube/bdc-catalog/bdc_catalog/triggers/band_metadata_expression.sql
        -> /home/raphael/devel/github/brazil-data-cube/bdc-catalog/bdc_catalog/triggers/timeline.sql
        -> /home/raphael/devel/github/brazil-data-cube/bdc-catalog/bdc_catalog/triggers/collection_statistics.sql
        -> /home/raphael/devel/github/brazil-data-cube/bdc-catalog/bdc_catalog/triggers/collection_tiles.sql

Updating an Existing Data Model

If you have followed the migration context and would like to upgrade to the latest module migration, use the following command:

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db alembic upgrade

Updating the Migration Scripts

Whenever you made change in a model, (or create a new model file and register with entry point bdc_db.models, you should generate a new revision and commit these changes into versioning control system (Git). Use the following command to generate migration changes:

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-db alembic revision "my revision" --branch=bdc_catalog

Note

Make sure to pass --branch=bdc_catalog in order to generate the alembic migration into the same BDC-Catalog branch tree.

Initial Data

Note

For this step, you will need to install few extra libraries:

pip3 install -e .[geo]

It will inject Shapely into virtual libraries.

We have prepared a minimal command line to insert collections into database. Please, refer to :doc:`cli` for further details.

You can load a initial data of sentinel-2 using the command line:

SQLALCHEMY_DATABASE_URI="postgresql://postgres:postgres@localhost:5432/bdcdb" \
bdc-catalog load-data --ifile examples/fixtures/sentinel-2.json

The following collection S2_L1C-1 will be loaded:

Collection S2_L1C-1 created
-> Creating Item S2A_MSIL1C_20151122T132122_N0204_R038_T23LMF_20151122T132134
- Total 1 items created.