Обсуждение: Doc patch, put pg_temp into the documentation's index

Hi,

2 patches:

pg_temp-toindex.patch
Puts pg_temp into the index of the docs.
(Line lengths are ugly so the change can
be easily reviewed.)

pg_temp-reformat.patch
Reformats the doc source after the
above patch.  (Fixes line length.)

Regards,

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."
                 -- Robert A. Heinlein

On Fri, 2012-09-28 at 11:10 -0500, Karl O. Pinc wrote:
> pg_temp-toindex.patch
> Puts pg_temp into the index of the docs.

But there is no object called pg_temp.  It always pg_temp_NNNN
something.  How should that be indexed?

Peter Eisentraut <peter_e@gmx.net> writes:
> On Fri, 2012-09-28 at 11:10 -0500, Karl O. Pinc wrote:
>> pg_temp-toindex.patch
>> Puts pg_temp into the index of the docs.

> But there is no object called pg_temp.  It always pg_temp_NNNN
> something.  How should that be indexed?

We do <replaceable>NNNN</> in a lot of places, and that seems
serviceable enough, at least in output formats where the NNNN can be
rendered differently from plain text.  I don't remember though whether
the sgml index infrastructure allows markup in an index item.
        regards, tom lane

On 11/17/2012 12:19:02 AM, Peter Eisentraut wrote:
> On Fri, 2012-09-28 at 11:10 -0500, Karl O. Pinc wrote:
> > pg_temp-toindex.patch
> > Puts pg_temp into the index of the docs.
>
> But there is no object called pg_temp.  It always pg_temp_NNNN
> something.  How should that be indexed?

My thought is not to index the db object; it isn't
particularly interesting to a user.  Instead what's
indexed is the token pg_temp, used when
setting search_path.   The utility of the token is
explained in several places in the docs.

Regards,

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."                -- Robert A. Heinlein

On Sat, 2012-11-17 at 11:33 -0600, Karl O. Pinc wrote:
> On 11/17/2012 12:19:02 AM, Peter Eisentraut wrote:
> > On Fri, 2012-09-28 at 11:10 -0500, Karl O. Pinc wrote:
> > > pg_temp-toindex.patch
> > > Puts pg_temp into the index of the docs.
> > 
> > But there is no object called pg_temp.  It always pg_temp_NNNN
> > something.  How should that be indexed?
> 
> My thought is not to index the db object; it isn't
> particularly interesting to a user.  Instead what's
> indexed is the token pg_temp, used when
> setting search_path.   The utility of the token is
> explained in several places in the docs.

Actually, since this is the pg_temp alias for the search path, it is
appropriate.  So committed as is.

On 11/17/2012 05:10:12 PM, Peter Eisentraut wrote:
> On Sat, 2012-11-17 at 11:33 -0600, Karl O. Pinc wrote:

> >
what's
> > indexed is the token pg_temp, used when
> > setting search_path.

> Actually, since this is the pg_temp alias for the search path, it is
> appropriate.  So committed as is.

Thanks for the work on this and the other patches you've helped
me out with, and for the larger work on PG of course.
I would feel like I was cluttering the channel if I
sent a thanks each time but I do want to acknowledge
both your help and the work the other Postgres people
do.

Regards,

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."                -- Robert A. Heinlein