Re: Plans for partitioning of inheriting tables - Mailing list pgsql-general

From Adrian Klaver
Subject Re: Plans for partitioning of inheriting tables
Date
Msg-id 666c82fe-be27-4f32-99ad-8952a56f2282@aklaver.com
Whole thread Raw
In response to Re: Plans for partitioning of inheriting tables  (thiemo@gelassene-pferde.biz)
Responses Re: Plans for partitioning of inheriting tables
List pgsql-general
On 11/1/24 12:16, thiemo@gelassene-pferde.biz wrote:
> Thanks, I shall have a look into it. I was under the assumption the the 
> create table like would create no more than a structural copy.

Not sure what you mean by structural copy, but the table created by 
CREATE TABLE LIKE will not have any association with the table it was 
created from.

https://www.postgresql.org/docs/current/sql-createtable.html

"Unlike INHERITS, the new table and original table are completely 
decoupled after creation is complete. Changes to the original table will 
not be applied to the new table, and it is not possible to include data 
of the new table in scans of the original table."

> 
> Torsten Förtsch <tfoertsch123@gmail.com <mailto:tfoertsch123@gmail.com>> 
> escribió:
> 
>> Thiemo,
>> it looks to me like you are using inheritance just to make sure your 
>> SOURCES and TOPO_FILES tables have some common columns. If you are not 
>> actually querying the TEMPLATE_TECH table and expect to see all the 
>> rows from the other 2 tables in that one table combined, then you 
>> could use CREATE TABLE (LIKE ...) instead of inheritance. That way 
>> your "child" tables would become normal tables and you could use 
>> declarative partitioning on them.
>> Even if you are querying the TEMPLATE_TECH table, you could still do 
>> that by turning the TEMPLATE_TECH table into a view which performs a 
>> UNION ALL over the other tables.
> 
> 

-- 
Adrian Klaver
adrian.klaver@aklaver.com




pgsql-general by date:

Previous
From: thiemo@gelassene-pferde.biz
Date:
Subject: Re: Plans for partitioning of inheriting tables
Next
From: Thiemo Kellner
Date:
Subject: Re: Plans for partitioning of inheriting tables