On 08/12/2018 05:41 PM, Samuel Williams wrote: > I wish the documentation would include performance details, i.e. this > operation is O(N) or O(1) relative to the number of rows. > > I found renaming a table was okay. > > How about renaming a column? Is it O(1) or proportional to the amount of > data? > > Is there any documentation about this?
The RENAME forms change the name of a table (or an index, sequence, view, materialized view, or foreign table), the name of an individual column in a table, or the name of a constraint of the table. There is no effect on the stored data. "
Just wondering - what about the case when the column being renamed is also referenced in an index or check constraint? (I would guess you cannot rename a column used in a check constraint without first removing it, but for an index, would this result in the index being rebuilt (or do you have to take care of that manually or are such references abstracted such that the column name "text" is irrelevant tot he actual structure of the index?).
Tim, as far as I know, names are only an attribute tagged to an OID. Internal relations are though these OIDs, not names, so renaming a column is really one-shot. Names are mainly a more convenient way of referring to objects.
Olivier
thanks Olivier, that is what I suspected and your explanation fits with my mental model. I had assumed table/column names are convenience for humans and that the system would use OIDs etc for internal references.