[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#681660: apt: ds



On Sun, Jul 15, 2012 at 02:50:28PM +0530, Faheem Mitha wrote:
> 
> 
> On Sun, 15 Jul 2012, Julian Andres Klode wrote:
> 
> >On Sun, Jul 15, 2012 at 02:23:00PM +0530, Faheem Mitha wrote:
> 
> >>It seems to me that in this situation, apt should exit with an error,
> >>since the squeeze-backports version requested is not available, but
> >>instead the stable version is installed.
> >
> >-t sets the target release. You want dh-lisp/squeeze-backports, which
> >then should result in:
> >
> >   $ apt-get install dh-lisp/squeeze-backports
> >   [...]
> >   E: Release 'squeeze-backports' for 'dh-lisp' was not found
> 
> Yes, it sets the target release. Since the target does not exist,
> should apt not abort?

It sets a global target release. If you don't have squeeze-backports
in your sources.list, it will error out with something like:

    E: The value 'squeeze-backports' is invalid for APT::Default-Release as such a release is not available in the sources

(in unstable, not in stable). Whether the package exists in the
target release or not is irrelevant:

  apt-get install -t squeeze-backports package

    means: give packages from squeeze-backports priority 990
    and install the version of 'package' with the highest
    priorirty

  apt-get install package/squeeze-backports

    means: install 'package' from squeeze-backports


> 
> >And you really messed up your bug report by embedding the
> >bug email with its headers inside your real email.
> 
> Not really my fault. The machine which the report is for cannot send
> email, so I saved it and tried to send it from my home machine using
> reportbug's --body-file option, but it sent it without giving me a
> chance to edit. The documentation is less than clear about this in
> the squeeze reportbug man page. See the bug report
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=484245

Just remember it for the next time.

> I can close this one and resubmit if anyone cares.

The bug is closed. That's what my last email did (although
I did not mention this explicitly). There's no reason to
reopen the bug report in my opinion.

Warning a user that a package does not exist in the release
configured by APT::Default-Release (which is the same as -t)
really seems a bit to extreme.

-- 
Julian Andres Klode  - Debian Developer, Ubuntu Member

See http://wiki.debian.org/JulianAndresKlode and http://jak-linux.org/.


Reply to: