Re: AW: [HACKERS] Another TEMP table trick - Mailing list pgsql-hackers

From Hannu Krosing
Subject Re: AW: [HACKERS] Another TEMP table trick
Date
Msg-id 36B2F091.7EC6EDB1@trust.ee
Whole thread Raw
In response to Re: AW: [HACKERS] Another TEMP table trick  (Bruce Momjian <maillist@candle.pha.pa.us>)
Responses Re: AW: [HACKERS] Another TEMP table trick
List pgsql-hackers
Bruce Momjian wrote:
>
> > There could be a problem with GUI tools that rely on these rows
> > to format their output (like pgaccess or ODBC --> M$ Access) though.
> 
> Oh, never thought of that.  A select of pg_class will return no rows for
> that table because it is a temp table.

One more reson to move \d from psql to backend maybe with syntax like 
Oracle's "DESC xxx" unless there is something in ANSI standard for that.

Or implement the ANSI system tables (I think there were some ;) and
views.

Then the front-end tools can be advised to use these (and TEMP TABLES
can 
add rows to other (possibly structure-permanent) TEMP tables that are
UNIONed 
withe real pg_class to give them real values.

Or we can even implement just temp _rows_ for tables that exist in a 
session only (maybe like in independant uncommitted transactions), 
and add the info for temp tables to pg_class (and friends) as temp rows.

----------------
Hannu


pgsql-hackers by date:

Previous
From: Vadim Mikheev
Date:
Subject: Re: [HACKERS] READ COMMITTED isolevel is implemented ...
Next
From: Hannu Krosing
Date:
Subject: Re: [HACKERS] READ COMMITTED isolevel is implemented ...