Your message dated Sat, 16 Nov 2024 16:31:00 -0700
with message-id <[email protected]>
and subject line Re: Bug#1087663: man-db: man --encoding=ASCII error message 
and error exit
has caused the Debian Bug report #1087663,
regarding man-db: man --encoding=ASCII error message and error exit
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
1087663: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1087663
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: man-db
Version: 2.13.0-1
Severity: normal

Previously this "worked" but I notice that now with at this the
current Sid version it produces the following noise upon exit.

    rwp@sid:~$ env -i PATH=$PATH HOME=$HOME USER=$USER TERM=$TERM MANWIDTH=80 
man --encoding=ASCII mv
    man: command exited with status 127: iconv -c -f UTF-8 -t ASCII//TRANSLIT | 
sed -e '/^[[:space:]]*$/{ N; /^[[:space:]]*\n[[:space:]]*$/D; }' | 
LESS=-ix8RmPm Manual page mv(1) ?ltline %lt?L/%L.:byte %bB?s/%s..?e (END):?pB 
%pB\%.. (press h for help or q to quit)$PM Manual page mv(1) ?ltline 
%lt?L/%L.:byte %bB?s/%s..?e (END):?pB %pB\%.. (press h for help or q to quit)$ 
MAN_PN=mv(1) pager

    rwp@sid:~$ echo $?
    3

I have been using --encoding=ASCII for a long time.  Personal
preference.  Then recently noticed the above upon 'q' to exit the
pager and error exit code.  But the below is slightly more concise to
document.

    rwp@sid:~$ env -i PATH=$PATH HOME=$HOME USER=$USER TERM=$TERM MANWIDTH=80 
man --encoding=ASCII mv >/dev/null
    man: command exited with status 127: iconv -c -f UTF-8 -t ASCII//TRANSLIT | 
col -b -p -x | sed -e '/^[[:space:]]*$/{ N; /^[[:space:]]*\n[[:space:]]*$/D; }'

    rwp@sid:~$ env -i PATH=$PATH HOME=$HOME USER=$USER TERM=$TERM MANWIDTH=80 
man --encoding=US-ASCII mv >/dev/null
    man: command exited with status 127: iconv -c -f UTF-8 -t 
US-ASCII//TRANSLIT | col -b -p -x | sed -e '/^[[:space:]]*$/{ N; 
/^[[:space:]]*\n[[:space:]]*$/D; }'

Because I hit this of course I went and read the man page in detail.

       -E encoding, --encoding=encoding
              Generate output for a character encoding other than the default.
              For backward compatibility, encoding may be an nroff device such
              as ascii, latin1, or utf8 as well as a true character encoding
              such as UTF-8.

And having read that I did try the other documented encodings and
found that these two make no error.

    rwp@sid:~$ env -i PATH=$PATH HOME=$HOME USER=$USER TERM=$TERM MANWIDTH=80 
man --encoding=UTF-8 mv >/dev/null ; echo $?
    0

    rwp@sid:~$ env -i PATH=$PATH HOME=$HOME USER=$USER TERM=$TERM MANWIDTH=80 
man --encoding=ascii mv >/dev/null ; echo $?
    0

Clearly I can change from ASCII to ascii and have the behavior I
desire without an error but I report the problem regardless because it
is a new bug appearing from somewhere.  iconv itself is working.

    rwp@sid:~$ iconv --list | grep -i ascii
    ASCII//
    CSASCII//
    US-ASCII//

    rwp@sid:~$ echo dæmon | iconv -f UTF-8 -t ASCII//TRANSLIT
    daemon

Thank you for maintaining man in Debian!

Bob


-- System Information:
Debian Release: trixie/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 6.11.7-amd64 (SMP w/2 CPU threads; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: sysvinit (via /sbin/init)

Versions of packages man-db depends on:
ii  bsdextrautils          2.40.2-11
ii  bsdmainutils           12.1.8
ii  debconf [debconf-2.0]  1.5.87
ii  groff-base             1.23.0-5
ii  libc6                  2.40-3
ii  libgdbm6t64            1.24-2
ii  libpipeline1           1.5.8-1
ii  libseccomp2            2.5.5-1+b3
ii  zlib1g                 1:1.3.dfsg+really1.3.1-1+b1

man-db recommends no packages.

Versions of packages man-db suggests:
pn  apparmor                   <none>
ii  chromium [www-browser]     130.0.6723.116-1
ii  dwb [www-browser]          20150419git-2+b1
ii  edbrowse [www-browser]     3.8.9-6
ii  elinks [www-browser]       0.17.0-2+b1
ii  firefox [www-browser]      132.0.2-1
ii  firefox-esr [www-browser]  128.4.0esr-1
ii  groff                      1.23.0-5
ii  less                       643-1
ii  links [www-browser]        2.29-1+b3
ii  links2 [www-browser]       2.29-1+b3
ii  lynx [www-browser]         2.9.2-1
ii  midori [www-browser]       7.0-2.1+b1
ii  w3m [www-browser]          0.5.3+git20230121-2+b3

-- debconf information:
  man-db/install-setuid: false
  man-db/auto-update: true

--- End Message ---
--- Begin Message ---
Colin Watson wrote:
> The change in behaviour is due to a change in iconv(1).  On bookworm:
>
>   $ echo dæmon | iconv -f UTF-8 -t ASCII//TRANSLIT; echo $?
>   daemon
>   0
>
> On trixie:
>
>   $ echo dæmon | iconv -f UTF-8 -t ASCII//TRANSLIT; echo $?
>   daemon
>   1
>
> I'm not currently sure whether this is a glibc bug, a case where man(1)
> was always invoking iconv(1) wrongly but it only just started noticing,
> or some difference in my test environments other than the OS version.

That does seem like a bug in iconv there.  It worked.  But yet it
exited non-zero?

Just for a little extra debug I created an "iconv" wrapper that would
exit 0 regardless.  If I do that then man behaves as it did before.

I'll file a bug against libc-bin and see what they say about it.  And
I will close this bug because as you say the problem is over there not
here.

Bob

--- End Message ---

Reply via email to