Skip to content

opendatacube/datacube-dataset-config

Repository files navigation

Datacube Dataset Config Examples and Scripts

Scripts for indexing data into ODC instances.

Documented Examples

GDAL3 Note

Please note - we are aware that using these scripts with GDAL 3.0+ can cause Latitudes and Longitudes to become transposed. When using these scripts, please use a python environment with an earlier version of GDAL.

odc-product-delete (In beta)

SQL scripts to delete ODC products and related records in ODC, Explorer and OWS DB. Each script takes product_name as an input parameter.

Usage:

psql -v product_name=<product-to-delete> -f <scriptname.sql> -h <database-hostname> <dbname>
  • delete_odc_product.sql

    • This script will delete a product and all datasets from an ODC DB.
    • It deletes records from tables agdc.dataset_source, agdc.dataset_location, agdc.dataset and agdc.dataset_type.
    • It also deletes indexes and view related to the ODC product.
  • delete_odc_product_explorer.sql

    • This script will delete records related to an ODC product from the Explorer Schema in the ODC DB.
    • This script needs to be run before delete_odc_product.sql as script makes reference to ODC DB.
    • It deletes records from these tables cubedash.dataset_spatial, cubedash.time_overview, cubedash.product.
    • Also, it refreshes materialised view cubedash.mv_dataset_spatial_quality.
  • delete_odc_product_ows.sql

    • This script will delete records from tables wms.product_ranges and wms.sub_product_ranges.
    • This script needs to be run before delete_odc_product.sql as OWS DB maintains a foreign key to ODC DB.
    • Table agdc.dataset_type has foreign key constraint wms.product_ranges_id_fkey on table wms.product_ranges

WARNING

  • AS ODC doesn't have product deletion feature with an intention to keep an immutable history of datasets, these scripts are created to manually delete an entire ODC Product.
  • These scripts deletes data, so review the scripts thoroughly and use it very carefully!
  • For more detail, read here