On Tuesday, February 1, 2022, Дмитрий Иванов <
firstdismay@gmail.com> wrote:
Why are COLUMN GENERATED ALWAYS AS called "methods" and singled out as a separate field type?
This sounds like you are describing something written. Can you provide a link to where that is?
It is possible to give a link, but it would require authorization:
Here are the contents of the topic:
DEFAULT VALUE <> GENERATED ALWAYS AS STORED
the DEFAULT value is not recalculated when the reference column are changed. is not an identical construction GENERATED ALWAYS AS STORED
reply:
They are in a separate list “Methods”.
This approach was used for other databases long before generated columns were implemented in PostgreSQL.
====================
It has to do with the syntax of generated and that you can generate data in different ways. Calling those ways “methods” seems reasonable.
Which community? There are many ways in which a "data schema [can be] presented as data". In PostgreSQL there is only a single source of truth for what the data schema is - the "System Catalogs" [1]. Those tables are made available to the user in the pg_catalog schema.
I came across a long acronym defining the name of the approach, then I decided that to come up with something fundamentally new is difficult enough. That everything already has a formal name. But I can't find it anymore. I need it to position my solution.
Shema - shema
Table Entity
id | propery1| property2| property3
Shema-data
Table Entity
id| name
Table Property Entity
id | id_entity | name| val
The Shema-data thing is called the Boogyman pattern because teachers use it to scare students and illustrate what not to do when designing a data model. You can also find it abbreviated “EAV anti-pattern” where EAV stands for entity-attribute-value.
David J.