Clarification regarding the model_name column
Hello PA Team!
While exploring the Pyramid repository database, I came across something I'd like to clarify.
I noticed that in the public.server_log_transaction table, the model_name column appears to store the model's unique ID (GUID) rather than its descriptive name. This was a bit confusing at first, especially since the adjacent database_name column in the same table does store the actual descriptive name of the database.
After some investigation, I've concluded that this is likely by design to ensure data integrity, since model names (as content items) can be changed by users, while their IDs remain static?
Could the Pyramid team or someone from the community confirm if this is the correct and intended approach?
Also, as a point of feedback, has there been any consideration for renaming this column to model_id in a future release to make the schema more intuitive? (if it's possible)
Thanks!
1 reply
-
Hi
As a general rule, Pyramid does not support nor encourage investigation of the structure of the underlying repository schema. Apart from the published "usage" model, any additional analysis of the repository contents is not supported. The underlying schema (with related detailed tables and columns) can and often does change from release to release to support changes in functionality of the overall product.
Generally, internally Pyramid uses GUIDs to refer to individual analytic objects (models, Discovers, Formulates etc.) in order to provide a unique reference to that specific item.
Hope that helps.
Ian