From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andreas Enge Subject: Re: Rollback problems Date: Thu, 24 Jan 2013 00:17:54 +0100 Message-ID: <201301240017.55013.andreas@enge.fr> References: <201301232148.46744.andreas@enge.fr> <87mwvzsdzt.fsf@gnu.org> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="Boundary-01=_i+GAR+VRqvuAwVs" Content-Transfer-Encoding: 7bit Return-path: Received: from eggs.gnu.org ([208.118.235.92]:35806) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Ty9aD-0001R9-Ux for bug-guix@gnu.org; Wed, 23 Jan 2013 18:18:23 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Ty9a4-00060R-OZ for bug-guix@gnu.org; Wed, 23 Jan 2013 18:18:17 -0500 In-Reply-To: <87mwvzsdzt.fsf@gnu.org> List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org Sender: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org To: Ludovic =?utf-8?q?Court=C3=A8s?= Cc: bug-guix@gnu.org --Boundary-01=_i+GAR+VRqvuAwVs Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Am Mittwoch, 23. Januar 2013 schrieb Ludovic Court=C3=A8s: > And what if you roll back once you=E2=80=99re at the empty profile? Then nothing should happen. > It seems more intuitive for me to error out like this, because there was > really nothing but nothingness before =E2=80=9Chello=E2=80=9D was install= ed. :-) > WDYT? No, I disagree; when I have nothing, install hello and roll back, I should= =20 be back to nothing. Some other opinions would be useful on this matter. > This is expected (same behavior as nix-env.) Profile generations are > not deleted unless you explicitly do so; this is what guarantees that > one can roll back anywhere they want. >=20 > But I wonder if this is really worth the trouble. In my experience, a > scenario like the one above rarely happens, if ever. I find the behaviour of roll back currently very confusing, and the=20 situation looks reasonable to me: I install hello, it works, so I keep it. I install freetype, it does not work, so I drop it again. I install file, it does not work, so I drop it again. Now I expect to have only hello, but I have hello and freetype. Of course, instead of rolling back, I can also uninstall. But the same=20 situation occurs when one replaces the package names by different versions,= =20 I suppose. I see installing packages and rolling back as steps forward and backward.=20 Going into direction B and back, then going into direction C and back=20 should not leave me with one step forward in direction B. Note that I did not use nix before; so I am just arguing from what would=20 intuitively be the correct behaviour of --roll-back for me: Being in=20 situation A, doing something to bring me into situation B and rolling back= =20 should put me into situation A again, not something that resembles=20 situation A, but with memories of B. Again, some other opinions would be useful. Andreas --Boundary-01=_i+GAR+VRqvuAwVs Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable

Am Mittwoch= , 23. Januar 2013 schrieb Ludovic Court=C3=A8s:

> And wh= at if you roll back once you=E2=80=99re at the empty profile?

&nb= sp;

Then nothin= g should happen.

&nb= sp;

> It see= ms more intuitive for me to error out like this, because there was

> really= nothing but nothingness before =E2=80=9Chello=E2=80=9D was installed. :-)=

> WDYT?<= /p>

&nb= sp;

No, I disag= ree; when I have nothing, install hello and roll back, I should be back to = nothing. Some other opinions would be useful on this matter.

&nb= sp;

> This i= s expected (same behavior as nix-env.) Profile generations are

> not de= leted unless you explicitly do so; this is what guarantees that

> one ca= n roll back anywhere they want.

>

> But I = wonder if this is really worth the trouble. In my experience, a

> scenar= io like the one above rarely happens, if ever.

&nb= sp;

I find the = behaviour of roll back currently very confusing, and the situation looks re= asonable to me:

I install h= ello, it works, so I keep it.

I install f= reetype, it does not work, so I drop it again.

I install f= ile, it does not work, so I drop it again.

Now I expec= t to have only hello, but I have hello and freetype.

&nb= sp;

Of course, = instead of rolling back, I can also uninstall. But the same situation occur= s when one replaces the package names by different versions, I suppose.

&nb= sp;

I see insta= lling packages and rolling back as steps forward and backward. Going into d= irection B and back, then going into direction C and back should not leave = me with one step forward in direction B.

&nb= sp;

Note that I= did not use nix before; so I am just arguing from what would intuitively b= e the correct behaviour of --roll-back for me: Being in situation A, doing = something to bring me into situation B and rolling back should put me into = situation A again, not something that resembles situation A, but with memor= ies of B.

&nb= sp;

Again, some= other opinions would be useful.

&nb= sp;

Andreas

&nb= sp;

--Boundary-01=_i+GAR+VRqvuAwVs--