On Mon, May 16, 2016 at 01:19:59PM +0200, Marc Haber wrote: > On Mon, May 16, 2016 at 12:57:09PM +0200, Julian Andres Klode wrote: > > On Mon, May 16, 2016 at 12:35:01PM +0200, Marc Haber wrote: > > It is the default since 0.5.10 in 2003: > > > > commit 906fbf8886926eeb302332d997c9bd861291e155 > > Author: Arch Librarian <arch@canonical.com> > > Date: Mon Sep 20 17:03:21 2004 +0000 > > > > * Make APT::Get::Show-Upgraded (aka apt-get -u) default... > > Author: mdz > > Date: 2003-08-22 02:46:09 GMT > > * Make APT::Get::Show-Upgraded (aka apt-get -u) default to true. > > > > > > (Unfortunately, we did not convert the Arch Librarian commits sanely > > when we moved APT from bzr to git, otherwise we'd have real author > > and date at the top....) (That probably has been gone wrong at the CVS to arch step already [with arch to bzr a 'free' upgrade], but yeah. There are other oddities in the very old history like the first commit not actually being the first commit timewise by a long shot…) > Is there a corresponding --no-show-upgraded option to apt-get? Does > still having the --show-upgraded option still make sense? All existing (boolean) options have a negated option as well, so yes, --show-upgraded exists just as well as --no-show-upgraded does (and a few other variants) as documented at the front of the OPTIONS section in the manpage of apt-get. What we probably should do is deprecate "-u" (as that doesn't deserve a short option, but not sure) and document the "--no-" version as we do with install-recommends, download and a few more. btw: As every commandline option is backed by a configuration option you have even more ways of changing the value… Best regards David Kalnischkies
Attachment:
signature.asc
Description: PGP signature