Your message dated Mon, 21 Mar 2016 18:46:10 +0100
with message-id <[email protected]>
and subject line Re: Bug#809563: reportbug: how to know if a package needs 
arch-qualif
has caused the Debian Bug report #809563,
regarding [reportbug] how to know if a package needs arch-qualif
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.)


-- 
809563: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=809563
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: reportbug
Version: 6.4.3
Severity: normal

_Steps to reproduce_

Invoke reportbug with a binary package name which fulfills the following
conditions:

 - it is not a source package name at the same time
 - you have this package installed for multiple architectures

Two packages which match these criteria on my system are libgl1-mesa-dri and
gcc-4.7-base.

_Expected behaviour_

Reportbug detects that all different architectures of this package actually
are the same bug reporting target and the user does not have to select one of
these packages.

_Actual behaviour_

Reportbug gives the following question:

----
Which of the following installed packages is the bug in?

1 gcc-4.7-base  GCC, the GNU Compiler Collection (base package)

2 gcc-4.7-base  GCC, the GNU Compiler Collection (base package)

3 gcc-4.7-base  Uninstalled/non-existent package

Select one of these packages:
----

-- Package-specific info:
** Environment settings:
EDITOR="vim"
INTERFACE="text"

** /home/adrian/.reportbugrc:
reportbug_version "4.8"
mode advanced
ui text
email "[email protected]"
no-cc
header "X-Debbugs-CC: [email protected]"
smtphost reportbug.debian.org

-- System Information:
Debian Release: wheezy/sid
  APT prefers unstable
  APT policy: (990, 'unstable'), (101, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.5-trunk-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages reportbug depends on:
ii  apt               0.9.7.5
ii  python            2.7.3-2
ii  python-reportbug  6.4.3

reportbug recommends no packages.

Versions of packages reportbug suggests:
pn  claws-mail                               <none>
pn  debconf-utils                            <none>
pn  debsums                                  <none>
pn  dlocate                                  <none>
pn  emacs22-bin-common | emacs23-bin-common  <none>
ii  file                                     5.11-2
ii  gnupg                                    1.4.12-6
pn  postfix | exim4 | mail-transport-agent   <none>
ii  python-gtk2                              2.24.0-3
pn  python-gtkspell                          <none>
pn  python-urwid                             <none>
ii  python-vte                               1:0.28.2-5
ii  xdg-utils                                1.1.0~rc1+git20111210-6

Versions of packages python-reportbug depends on:
ii  apt               0.9.7.5
ii  python            2.7.3-2
ii  python-debian     0.1.21+nmu2
ii  python-debianbts  1.11
ii  python-support    1.0.15

python-reportbug suggests no packages.

-- no debconf information

--- End Message ---
--- Begin Message ---
Hi!

On Sun, 2016-02-21 at 11:24:56 +0100, Guillem Jover wrote:
> I'm not entirely sure what you expect out of this bug report. Is this
> a support request? If so a mail to debian-dpkg might have been better.
> Also CCing the package address helps seeing the issue more quickly.
> 
> On Fri, 2016-01-01 at 14:04:05 +0000, Sandro Tosi wrote:
> > Dear dpkg maintainer, how can reportbug understand if a package needs
> > arch-qualification or not? for example
> > 
> > $ dpkg-query -W gcc-5-base
> > gcc-5-base:amd64        5.2.1-24
> > gcc-5-base:i386 5.2.1-24
> > 
> > but
> > 
> > $ dpkg-query -W dpkg
> > dpkg    1.18.3
> 
> «dpkg-query -W» will arch-qualify packages when needed. The current
> rules for arch-qualifying can be found in dpkg-query(1), in the
> description of the binary:Package virtual field.
> 
> > we are currently using the 'status' file (we are parsing it matching
> > on '^Package:'), where:
> > 
> > $ dpkg --status gcc-5-base:amd64
> > Package: gcc-5-base
> > Status: install ok installed
> > Priority: required
> > Section: libs
> > Installed-Size: 197
> > Maintainer: Debian GCC Maintainers <[email protected]>
> > Architecture: amd64
> > [8<]
> > 
> > so the Package line doesnt have the arch qualification so our
> > file-parsing doesnt return it, but from the command line we need to
> > specify it else:
> > 
> > $ dpkg --status gcc-5-base
> > dpkg-query: error: --status needs a valid package name but
> > 'gcc-5-base' is not: ambiguous package name 'gcc-5-base' with more
> > than one installed instance
> 
> Any command that requires a specific package instance, needs to be
> unambiguous, the commands accepting patterns do not need to, and
> something like «dpkg-query -W gcc-5-base» is equivalent to
> «dpkg-query -W gcc-5-base:*».
> 
> > how can we improve that in reportbug? is there a better way to
> > retrieve such information? ideally if someone wants to report a bug
> > against 'gcc-5-base' we should be able to return the 2 entries as of:
> > 
> > $ dpkg-query -W gcc-5-base
> > gcc-5-base:amd64        5.2.1-24
> > gcc-5-base:i386 5.2.1-24
> > 
> > (we also provide the description, that's why we use the status db).
> 
> I'm not exactly sure what you are asking for here. You need to either
> parse the status file, and match the arch-qualifying rules documented
> in dpkg-query. Or you could use something lile:
> 
>   «dpkg-query -f '${binary:Package} ${Version}\n ${Description}\n' pkg»
> 
> or any other formatting you might want to parse.

I'm closing this request now. If there's anything else to be done, please
feel free to reopen.

Thanks,
Guillem

--- End Message ---

Reply via email to