NetBSD Problem Report #52710

From www@NetBSD.org  Thu Nov  9 17:15:37 2017
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 60F3F7A1EA
	for <gnats-bugs@gnats.NetBSD.org>; Thu,  9 Nov 2017 17:15:37 +0000 (UTC)
Message-Id: <20171109171536.801717A237@mollari.NetBSD.org>
Date: Thu,  9 Nov 2017 17:15:36 +0000 (UTC)
From: jamie@dyslexicfish.net
Reply-To: jamie@dyslexicfish.net
To: gnats-bugs@NetBSD.org
Subject: Error in vis(1) manpage
X-Send-Pr-Version: www-1.0

>Number:         52710
>Category:       bin
>Synopsis:       Error in vis(1) manpage
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    bin-bug-people
>State:          open
>Class:          doc-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Nov 09 17:20:00 +0000 2017
>Last-Modified:  Wed Nov 15 11:45:00 +0000 2017
>Originator:     Jamie Landeg Jones
>Release:        head
>Organization:
>Environment:
n/a (not running netbsd)
>Description:
The -F and -f options are incorrectly specified, as detailed in the attached fix patch to the manual pages source file.

For convenience, the diff from the resultant text is:

30,38c30,37
<      -F foldwidth
<              Causes vis to fold output lines to foldwidth columns (default
<              80), like fold(1), except that a hidden newline sequence is used,
<              (which is removed when inverting the file back to its original
<              form with unvis(1)).  If the last character in the encoded file
<              does not end in a newline, a hidden newline sequence is appended
<              to the output.  This makes the output usable with various editors
<              and other utilities which typically don't work with partial
<              lines.
---
>      -f      Causes vis to fold output lines at N columns (default 80, over&#8208;
>              ridden by -F.), like fold(1), except that a hidden newline
>              sequence is used, (which is removed when inverting the file back
>              to its original form with unvis(1)).  If the last character in
>              the encoded file does not end in a newline, a hidden newline
>              sequence is appended to the output.  This makes the output usable
>              with various editors and other utilities which typically don't
>              work with partial lines.

40c39,40
<      -f      Same as -F.
---
>      -F foldwidth
>              Set the column width to foldwidth when using -f.

51,52c51,52
<              still doubled and hidden newline sequences inserted if -f or -F
<              is selected.  When combined with the -f flag, vis becomes like an
---
>              still doubled and hidden newline sequences inserted if -f is
>              selected.  When combined with the -f flag, vis becomes like an

>How-To-Repeat:
man vis
>Fix:
# Patch errors in "vis(1)" manpage:
--- contrib/vis/vis.1.orig      2016-12-02 01:00:16.614591000 +0000
+++ contrib/vis/vis.1   2016-12-23 04:14:50.771339000 +0000
@@ -75,10 +75,12 @@
 .Ar extra ,
 per
 .Xr svis 3 .
-.It Fl F Ar foldwidth
+.It Fl f
 Causes
 .Nm
-to fold output lines to foldwidth columns (default 80), like
+to fold output lines at N columns (default 80, overridden by
+.Fl F . ) ,
+like
 .Xr fold 1 ,
 except
 that a hidden newline sequence is used, (which is removed
@@ -89,9 +91,11 @@
 This makes
 the output usable with various editors and other utilities which
 typically don't work with partial lines.
-.It Fl f
-Same as
-.Fl F .
+.It Fl F Ar foldwidth
+Set the column width to
+.Ar foldwidth
+when using
+.Fl f .
 .It Fl h
 Encode using the URI encoding from RFC 1808.
 .Pq Dv VIS_HTTPSTYLE
@@ -106,8 +110,6 @@
 Turns off any encoding, except for the fact that backslashes are
 still doubled and hidden newline sequences inserted if
 .Fl f
-or
-.Fl F
 is selected.
 When combined with the
 .Fl f

>Audit-Trail:
From: Thomas Klausner <tk@giga.or.at>
To: gnats-bugs@NetBSD.org
Cc: Christos Zoulas <christos@zoulas.com>
Subject: Re: bin/52710: Error in vis(1) manpage
Date: Mon, 13 Nov 2017 15:28:15 +0100

 On Thu, Nov 09, 2017 at 05:20:00PM +0000, jamie@dyslexicfish.net wrote:
 > >Number:         52710
 > >Category:       bin
 > >Synopsis:       Error in vis(1) manpage

 I wonder if this is a bug in vis.c instead.

                 case 'F':
                         if ((foldwidth = atoi(optarg)) < 5) {
                                 errx(1, "can't fold lines to less than 5 cols");
                                 /* NOTREACHED */
                         }
                         markeol++;
                         break;
                 case 'f':
                         fold++;         /* fold output lines to 80 cols */
                         break;          /* using hidden newline */

 Perhaps this should be "fold++" instead of "markeol++".
  Thomas

From: Jamie Landeg-Jones <jamie@dyslexicfish.net>
To: christos@zoulas.com, tk@giga.or.at, netbsd-bugs@NetBSD.org,
        jamie@dyslexicfish.net, gnats-bugs@NetBSD.org, gnats-admin@NetBSD.org
Cc: 
Subject: Re: bin/52710: Error in vis(1) manpage
Date: Wed, 15 Nov 2017 11:44:37 +0000

 Thomas Klausner <tk@giga.or.at> wrote:

 >  I wonder if this is a bug in vis.c instead.
 >  
 >                  case 'F':
 >                          if ((foldwidth = atoi(optarg)) < 5) {
 >                                  errx(1, "can't fold lines to less than 5 cols");
 >                                  /* NOTREACHED */
 >                          }
 >                          markeol++;
 >                          break;
 >                  case 'f':
 >                          fold++;         /* fold output lines to 80 cols */
 >                          break;          /* using hidden newline */
 >  
 >  Perhaps this should be "fold++" instead of "markeol++".
 >   Thomas

 I see what you're saying: Rather than '-F' simply setting the columns for use by '-f', the
 '-F' would basically imply the '-f', and be used instead of it.

 Hmm, make sense.. The manpage would still be wrong, but not quite as much!

 Cheers, Jamie

Home	PR Database Search