Andrey Mashenkov created IGNITE-24763:
-----------------------------------------
Summary: Sql. Unification of column names in disaster recovery
system views
Key: IGNITE-24763
URL: https://issues.apache.org/jira/browse/IGNITE-24763
Project: Ignite
Issue Type: Improvement
Components: sql
Reporter: Andrey Mashenkov
Assignee: Andrey Mashenkov
Fix For: 3.1
We have a bunch of system views with inconsistent column naming.
The following proposal is to always use the entity name prefix, even if the
field is directly related to the current view.
As an example:
In TABLES there will be a TABLE _NAME and TABLE_ID field, not just NAME and ID.
This will always allow us to read the query fields unambiguously and easily do
a natural join.
However, AI3 has already been introduced and shouldn't break compatibility.
Let's add extra columns for each of the system views where the column name
doesn't correspond to the aforementioned principle and mark somehow old columns
as deprecated (we still don't have a common approach to how it can be marked).
So as a result implementing of the ticket, we will have duplicate columns for
part of the views, but they will be removed in the future.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
