parent-key-improvements

Registered by Aaron Rosen

There are two current shortcomings of the parent_key implementation in
congress.
* It's unable to be accessed from a subtable. This is needed in the neutron
  refactoring working to handle how neutron structures the response for
  routers.

* It's column name is always named parent_key. It would be helpful if we
  could have the column in the schema actually reflect it's name.
  For example, router_id per say.

Change the datasource framework code so this can be done. I plan to only
make this change for hdict because we don't think we will need vdict in
the future. If it turns out that we need this for vdict then I'll
add support for it.

To handle specifiy the column name on a parent_key on should add:
'parent-col-name': <NAME> to the translator. If this is not present
then the old behavior will still occur of having the column name called
parent_key.

Blueprint information

Status:
Complete
Approver:
None
Priority:
Undefined
Drafter:
Aaron Rosen
Direction:
Needs approval
Assignee:
None
Definition:
New
Series goal:
None
Implementation:
Implemented
Milestone target:
None
Started by
Tim Hinrichs
Completed by
Tim Hinrichs

Related branches

Sprints

Whiteboard

Gerrit topic: https://review.openstack.org/#q,topic:bp/parent-key-improvements,n,z

Addressed by: https://review.openstack.org/139159
    Break out inline method _get_schema

Addressed by: https://review.openstack.org/139160
    Break out _get_schema to get_schema(hdict|vdict|list)

Addressed by: https://review.openstack.org/140876
    Allow parent_key column to specify its name

Gerrit topic: https://review.openstack.org/#q,topic:bp/refactor-datasource-driver-framework,n,z

Addressed by: https://review.openstack.org/140855
    Make parent_key accessable from subtable

Addressed by: https://review.openstack.org/138594
    Break up convert_obj

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.