Mailinglist Archive: yast-devel (144 mails)

< Previous Next >
Re: [yast-devel] RFC: Drop yast2 sw_single CLI interface
On Fri, 30 Aug 2013 10:14:49 +0200 (CEST)
Johannes Meixner <jsmeix@xxxxxxx> wrote:


On Aug 29 19:44 Rajko wrote (excerpt):
On Thu, 29 Aug 2013 15:36:52 +0200
Ladislav Slezak <lslezak@xxxxxxx> wrote:

Starting zypper automatically looks too "smart" for me, I'd leave
the decision to use zypper (or something else) to the user.

Think of it in a different way.

Majority of users are looking for performed operations, not tools
that are used.

If you plan to provide text ready for copy paste, you can issue
"yast -i <package>" is depreciated, using "zypper install <package>"
and then install package.

This way 99% of users that follow old way, possibly following some
out of date instructions, have nothing to do, and 1% that rely on
some "yast -i" specific behavior have warning if result is not what
they expect.

I think to be really backward compatible, there should be a popup
that informs the user as shown above and waits a specific time for
user accept or deny and if the timeout passed without user response
the default action should be to run "zypper install <package>".

In old YCP one would have used something like
Popup::TimedAnyQuestion() to implement such a behaviour.

Popup in CLI is real pain and I think we should not do it this
way. Complain to stderr sounds better for me.

This way "yast -i <package>" could still work even unattended.
Perhaps some users call "yast -i <package>" in scripts?
This might be of importance for those users that are not yet included
in the 99% and 1% above ;-)

Well, goal of yast -i <package> is install package. Thats what
documentation say, so if user depend on undocumented behavior he must
be prepared that it can change without big noise.


Kind Regards
Johannes Meixner

To unsubscribe, e-mail: yast-devel+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: yast-devel+owner@xxxxxxxxxxxx

< Previous Next >