Re: ORDER BY TABLENAME, possible bug

From: Thomas Kellerer <spam_eater(at)gmx(dot)net>
To: pgsql-bugs(at)postgresql(dot)org
Subject: Re: ORDER BY TABLENAME, possible bug
Date: 2016-11-18 23:37:58
Message-ID: o0o3ch$k0v$1@blaine.gmane.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

David G. Johnston schrieb am 18.11.2016 um 23:48:
>
> Is this somewhere in the documentation? The only place I could find
> where there is a hint of this use, is the Note in Row Constructors in
> https://www.postgresql.org/docs/current/static/sql-expressions.html
> <https://www.postgresql.org/docs/current/static/sql-expressions.html>
> that uses a table alias without the .* in an expression: ROW(t, 42)
>
> ​https://www.postgresql.org/docs/9.6/static/rowtypes.html
>
> ​"Whenever you create a table, a composite type is also automatically created, with the same name as the table, to represent the table's row type."
>
> So, its documented and in technically correct location. I'm not sure
> if introducing this material in a "tutorial" would be a gain or just
> confuse the student. It seems to be something one picks up somehow
> (trial and error, mailing list, stumbling upon it in the docs or
> elsewhere on the Internet) as one increases their knowledge of SQL to
> an intermediate level.

I think the chapter about "Select Lists"[1] would be the approriate place to explain this.

I would also love to see an explanation there on why "select (a,b,c)" is something different then "select a,b,c"

I see far too many people putting the columns of the select list between parentheses and then being confused about the output.

I don't know if the Postgres behaviour is mandated by the SQL standard.

If it's not, that should be documented, maybe in the chapter I mentioned above or in the compatibility section of SELECT.
If that _is_ mandated by the standard, then I think some small note/warning might be useful (especially to newcomers)

Thomas

[1] https://www.postgresql.org/docs/current/static/queries-select-lists.html

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Richard W.M. Jones 2016-11-19 23:03:12 PostgreSQL cannot be compiled on RISC-V
Previous Message Tom Lane 2016-11-18 23:33:30 Re: ORDER BY TABLENAME, possible bug