standards/59647: document sort(1) peculiarities

[uwe%stderr.spb.ru@localhost]

>Number:         59647
>Category:       standards
>Synopsis:       document sort(1) peculiarities
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    standards-manager
>State:          open
>Class:          doc-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Sep 10 13:45:00 +0000 2025
>Originator:     Valery Ushakov
>Release:        NetBSD 10.1
>Organization:
>Environment:
>Description:
Cf. bin/51726 or src/bin/sh/mkoptions.sh revision 1.8:

  date: 2024-04-17 02:30:19 +0300;  author: christos;  state: Exp;
  lines: +3 -3;  commitid: 2c7j1Tx6qooqqq6F;

  Be more explicit with sort fields to produce consistent results with
  gnu sort (Jan-Benedict Glaw)

Our sort behaves differently w.r.t. ordering options (-b, -d, ...),
that can be specified either "globbaly" or as part of "keydef" (what
our manual page calls kstart, kend).

I am still not sure from the description, if it's only -b that is
special in our sort, or all of the ordering options.  The man page
seems to imply the former, but a quick RTFS seems to imply the latter
at a (very) quick glance.

A manual page should be a refernce for the proverbial desperate
sysadmin at 3am Saturday, not an exegetical exercise.

>How-To-Repeat:
"Explain lie I'm five" src/bin/sh/mkoptions.sh revision 1.8:

  date: 2024-04-17 02:30:19 +0300;  author: christos;  state: Exp;
  lines: +3 -3;  commitid: 2c7j1Tx6qooqqq6F;

>Fix:
Write some prose for the sort(1) man page that explains how ordering
options are actually applied and how that behavour is different from
the posix/gnu treatment of those same options.  Including CAVEATS
section.

Not _directly_ related, but an EXAMPLES section that demonstrates
relevant material with specific cases and not abstract prose that
sounds like it's from a medieval Zenzizenzizenzic treatise badly
translated from Latin could be handy...