Re: Rationalizing EXPLAIN VERBOSE output

From: Peter Eisentraut <peter_e(at)gmx(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, <pgsql-hackers(at)postgreSQL(dot)org>
Subject: Re: Rationalizing EXPLAIN VERBOSE output
Date: 2002-03-11 02:25:44
Message-ID: Pine.LNX.4.30.0203102121460.684-100000@peter.localdomain
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Tom Lane writes:

> What I would suggest is that EXPLAIN VERBOSE ought to emit either
> non-pretty-print or pretty-print dump format, not both (probably control
> this with debug_pretty_print or another newly-invented GUC parameter;
> IMHO the factory default should be pretty-printing).

Sounds good. I think we can reuse the parameter.

> A disadvantage of elog(INFO) output for pretty-printed plans is that
> AFAIK psql doesn't currently have any way of capturing NOTICE output
> into a file. I find it much better to look at pretty-printed dumps
> in Emacs than on a terminal window, mainly because Emac's M-C-f and
> M-C-b commands understand the nesting structure so it's easy to move
> around in the dump with them. How hard would it be to get psql to
> send notice output into a \g file?

\g (and \o) send only the query results to a file. The idea is that you
want to save the results, but if there's a warning or error, you want to
see it. We could add alternative commands (\G and \O?) that save the
notices and errors as well. Not sure if this is useful beyond this
application. In non-interactive situations you'd usually use shell
redirections to save all output.

--
Peter Eisentraut peter_e(at)gmx(dot)net

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2002-03-11 02:28:49 Re: Rationalizing EXPLAIN VERBOSE output
Previous Message Peter Eisentraut 2002-03-11 01:53:02 Re: Do we still have locking problems with concurrent