Since RHEL 6.1 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.

Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.

Hi,

I have noticed that reported problem occurs for any non-English language. Also verified translation files (Po files), they all look clean.

Digging through the problem, I have tried to compare results between English(US) and Other languages like Japanese(ja_JP), found that there's issue with following lines of code in libvirt/tools/virsh.c file

  11393         if (desc[0]) {
  11394             /* Print the description only if it's not empty.  */
  11395             fputs(_("\n  DESCRIPTION\n"), stdout);
  11396             fprintf(stdout, "    %s\n", desc);
  11397         }

Please check these lines.

Thanks!
Ankit

(In reply to comment #3)

> Digging through the problem, I have tried to compare results between
> English(US) and Other languages like Japanese(ja_JP), found that there's issue
> with following lines of code in libvirt/tools/virsh.c file
> 
>   11393         if (desc[0]) {
>   11394             /* Print the description only if it's not empty.  */
>   11395             fputs(_("\n  DESCRIPTION\n"), stdout);
>   11396             fprintf(stdout, "    %s\n", desc);

No direct problem in those lines.  The first line should be translated in-place; the second (the fprintf of desc) should have been translated back where desc is declared several lines earlier:

        const char *desc = _(vshCmddefGetInfo(def, "desc"));

and that should happen correctly in the po file if all of the description strings are marked with N_() (I didn't spot any that were missing).

However, it DOES point out a flaw - the use of N_("") is a bug.  Gettext reserves the translation of the empty string for its own use, and therefore, it is a bug if we ever mark or translate an empty string.  And since we do have several instances of code such as:

static const vshCmdInfo info_pool_name[] = {
    {"help", N_("convert a pool UUID to pool name")},
    {"desc", ""},

that means that our initialization of desc must account for an unmarked empty string rather than blindly using _() on the description text.

I'll propose a patch upstream.

Will be picked up by the next rebase.

commit 491858bf3cd7b91029744287f6173f795b3108f9
Author: Eric Blake <eblake>
Date:   Mon Jun 20 14:25:08 2011 -0600

    virsh: avoid bogus description
    
    https://bugzilla.redhat.com/show_bug.cgi?id=682121
    
    Gettext reserves the empty string for internal use, and it must
    not be passed through _().  We were violating this for commands
    that (for whatever reason) used "" for their description.
    
    * tools/virsh.c (vshCmddefHelp): Don't translate empty string.
    Reported by Tatsuo Kawasaki.

Reproduced this issue on libvirt-0.9.2-1.el6.x86_64
#LANG=ja_JP virsh help iface-name

  名前
    iface-name - convert an interface MAC address to interface name

  形式
    iface-name <interface>

  詳細
    Project-Id-Version: ja
Report-Msgid-Bugs-To: libvir-list
POT-Creation-Date: 2010-04-30 18:31+0200
PO-Revision-Date: 2009-09-24 10:12+0900
Last-Translator: Kiyoto Hashida <khashida>
Language-Team: Japanese <jp>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
X-Generator: KBabel 1.11.4
Plural-Forms: Plural-Forms: nplurals=1; plural=0;





  オプション
    [--interface] <string>  interface mac

Tested this issue with:

libvirt-0.9.3-1.el6
qemu-kvm-0.12.1.2-2.167.el6
kernel-2.6.32-164.el6

  名前
    iface-name - convert an interface MAC address to interface name

  形式
    iface-name <interface>

  オプション
    [--interface] <string>  interface mac

So change the status to VERIFIED.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2011-1513.html