On 8/25/05, Jürgen Hötzel <[EMAIL PROTECTED]> wrote:
> On Thu, Aug 25, 2005 at 10:10:24AM -0500, Aaron Griffin wrote:
> > On 8/25/05, Jürgen Hötzel <[EMAIL PROTECTED]> wrote:
> > > On Tue, Aug 23, 2005 at 05:02:30PM -0500, Aaron Griffin wrote:
> > > > On 8/23/05, eliott <[EMAIL PROTECTED]> wrote:
> > > > > Yum has the following feature:
> > > > >
> > > > > "check-update
> > > > > Implemented  so  you could know if your machine had any updates that 
> > > > > needed
> > > > > to be applied  without  running  it  interactively.
> > > > > Returns  exit  value of 100 if there are packages available for an 
> > > > > update.
> > > > > Also returns a list of the pkgs  to  be  updated  in list  format.  
> > > > > Returns
> > > > > 0  when  no  packages  are available for update."
> > > > >
> > > > > I would like to see something similar for pacman. It seems kind of a 
> > > > > kludge
> > > > > to pipe an N to pacman's output, then try interpreting the results..or
> > > > > worse..trying to use expect to supply an N to an update.
> > > > >
> > > > > Something like 'pacman -Sy --check-only' would be very nice.
> > > > >
> > > > > Are there other solutions than the ones listed? Does this seem like a
> > > > > reasonable feature request to anyone
> > > > > else?
> > > >
> > > >
> > > > Much to dibble's regret... (heh), here ya go:
> > > >
> > > > yes n | pacman -Syu | grep "Total Package Size" | cut -d: -f2 | tr -d \
> > > >
> > > Nice hack. But also a good example of UNIX philosophy's "avoid captive 
> > > user
> > > interface" tenet, which pacman doesn't adopt in this case.
> > >
> > > Jürgen
> >
> > captive user interface? as in requiring user interaction?
> >
> 
> Yes, a captive user interface moves user interaction outside the scope of
> the command interpreter. Captive user interfaces assume the the user is
> human. So you needed the "yes n" hack.
> 
> Jürgen

Yeah, I know... I can agree with that principal, but I don't mind
much... hell, the kernel PKGBUILD does (used to? does it still do
this?) "yes | menuconfig"

What bothers me though, is the "yes" program... especially when doing
"yes n"... it should be called "spam" or something... ln -s yes "spam
y", ln -s no "spam n" voila!

Heh

_______________________________________________
arch mailing list
[email protected]
http://www.archlinux.org/mailman/listinfo/arch

Reply via email to