Re: Doc: Rework contrib appendix -- informative titles, tweaked sentences - Mailing list pgsql-hackers

From Karl O. Pinc
Subject Re: Doc: Rework contrib appendix -- informative titles, tweaked sentences
Date
Msg-id 20230330012705.7295be34@slate.karlpinc.com
Whole thread Raw
In response to Re: Doc: Rework contrib appendix -- informative titles, tweaked sentences  (Noah Misch <noah@leadboat.com>)
Responses Re: Doc: Rework contrib appendix -- informative titles, tweaked sentences
List pgsql-hackers
On Wed, 29 Mar 2023 21:32:05 -0700
Noah Misch <noah@leadboat.com> wrote:

> On Sun, Jan 22, 2023 at 02:42:46PM -0600, Karl O. Pinc wrote:
> > v10-0001-List-trusted-and-obsolete-extensions.patch  
> 
> > + <para id="contrib-obsolete">
> > +   These modules and extensions are obsolete:
> > +
> > +   <simplelist type="inline">
> > +     <member><xref linkend="intagg"/></member>
> > +     <member><xref linkend="xml2"/></member>
> > +   </simplelist>
> > + </para>  
> 
> Commit a013738 incorporated this change.  Since xml2 is the only
> in-tree way to use XSLT from SQL, I think xml2 is not obsolete.  Some
> individual functions, e.g. xml_valid(), are obsolete.  (There are
> years-old threats to render the module obsolete, but this has never
> happened.)

Your point seems valid but this is above my station.
I have no idea as to how to best resolve this, or even how to make the
resolution happen now that the change has been committed.
Someone who knows more than me about the situation is needed
to change the phrasing, or re-categorize, or rework the xml2
module docs, or come up with new categories of obsolescence-like 
states, or provide access to libxslt from PG, or something.

I am invested in the patch and appreciate being cc-ed.

Regards,

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



pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: pg_basebackup: Correct type of WalSegSz
Next
From: "houzj.fnst@fujitsu.com"
Date:
Subject: RE: Support logical replication of DDLs