You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Drop root_id column
This column makes it easy to query all rows belonging to the same property hierarchy. The same can be achieved using an hierarchical query on parent_id though. Dropping this column would also remove the corresponding FK constraint and index both of which are expensive during imports.
Drop FK constraint on "datatype_id"
This constraint is expensive during imports. Removing it would allow storing values not defined in the "datatype" table. However, such invalid values can simply be ignored in exports and queries.
Drop FK constraint on "namespace_id"
This constraint is expensive during imports. Removing it would allow storing values not defined in the "namespace" table. However, such invalid values can simply be ignored in exports and queries. The index on "namespace_id" must be kept though.
The text was updated successfully, but these errors were encountered:
To be consistent, also drop the FK constraint on the "objectclass_id" column of the "feature" table. The index on "objectclass_id" must be kept though.
Drop
root_id
columnThis column makes it easy to query all rows belonging to the same property hierarchy. The same can be achieved using an hierarchical query on
parent_id
though. Dropping this column would also remove the corresponding FK constraint and index both of which are expensive during imports.Drop FK constraint on "datatype_id"
This constraint is expensive during imports. Removing it would allow storing values not defined in the "datatype" table. However, such invalid values can simply be ignored in exports and queries.
Drop FK constraint on "namespace_id"
This constraint is expensive during imports. Removing it would allow storing values not defined in the "namespace" table. However, such invalid values can simply be ignored in exports and queries. The index on "namespace_id" must be kept though.
The text was updated successfully, but these errors were encountered: