new commit is overly verbose

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Troup wrote:
> Public bug reported:
>
> With bzr 0.92, I see this:
>
> # bzr commit -m "<wibble>"
> Committing revision 37 to "/etc/".
> modified postgresql/8.2/main/pg_hba.conf
> modified postgresql/8.2/main/pg_ident.conf
> Committed revision 37.
>
> But it use to look like this:
>
> # bzr commit -m "<wibble>"
> modified postgresql/8.2/main/pg_hba.conf
> modified postgresql/8.2/main/pg_ident.conf
> Committed revision 37.
> #
>
> How come we get told the revision number twice now?

Because people wanted to be told the location they'd committed to in
case it was a mistake, and I reasoned that if it was a mistake, they
should be able to detect it while committing and abort it.

> (I know it seems petty, but I do a lot of small commits, so this is
> going to bug me enough that I feel I at least ought to report the bug.)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHTakR0F+nu1YWqI0RAht/AJ98sawYbH8vDIdKX4SZRiTVPgI3WgCfcQOa
CDTqDn0GRl42i5b9nu1DGH4=
=SYvd
-----END PGP SIGNATURE-----

This behaviour was modified in response to bug 52479. The ML discussion is at https://lists.ubuntu.com/archives/bazaar/2007q3/030503.html

We could:

a) Not show the revision number in the first line, so at least the data is different:
Committing to "/etc/".
modified postgresql/8.2/main/pg_hba.conf
modified postgresql/8.2/main/pg_ident.conf
Committed revision 37.

b) Not show the final line
Committing revision 37 to "/etc/".
modified postgresql/8.2/main/pg_hba.conf
modified postgresql/8.2/main/pg_ident.conf

c) Change the final line to not show the number:
Committing revision 37 to "/etc/".
modified postgresql/8.2/main/pg_hba.conf
modified postgresql/8.2/main/pg_ident.conf
Successful/Finished/Committed

d) Show the revision number at the beginning, and the revision id at the end
Committing revision 37 to "/etc/".
modified postgresql/8.2/main/pg_hba.conf
modified postgresql/8.2/main/pg_ident.conf
Committed revision: <email address hidden>

I don't see a huge problem with the repeated number, but I do understand that it seems a little redundant.
We've generally chosen to avoid printing revision ids unless specifically requested for them, so I think (d) is out.
I don't really like (b) because commits usually take a little while. Though I suppose having your prompt return might be enough of an indication that it has finished.
(a) is okay, it feels a bit like we know what will happen, go ahead and let the user know about it, but we don't have to.

I'd personally prefer (a) with a small modification:

Committing to: /etc/
modified postgresql/8.2/main/pg_hba.conf
modified postgresql/8.2/main/pg_ident.conf
Committed revision 37.

to make is more consistant with other commands, e.g. pull:

Using saved location: /etc/
 M postgresql/8.2/main/pg_hba.conf
 M postgresql/8.2/main/pg_ident.conf
All changes applied successfully.
Now on revision 37.

Lukáš Lalinský пишет:
> I'd personally prefer (a) with a small modification:
>
> Committing to: /etc/
> modified postgresql/8.2/main/pg_hba.conf
> modified postgresql/8.2/main/pg_ident.conf
> Committed revision 37.
>
> to make is more consistant with other commands, e.g. pull:
>
> Using saved location: /etc/
> M postgresql/8.2/main/pg_hba.conf
> M postgresql/8.2/main/pg_ident.conf
> All changes applied successfully.
> Now on revision 37.
>
+1

Lukáš's form is rather trivial to implement if someone wants to. (It should amount to approximately 1 line of code changed).

But would this satisfy James? Since he is the one bothered by the verbosity.

John A Meinel <email address hidden> writes:

> Luk����'s form is rather trivial to implement if someone wants to. (It
> should amount to approximately 1 line of code changed).
>
> But would this satisfy James? Since he is the one bothered by the
> verbosity.

It'd be fine by me.

--
James