octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: extra head


From: Daniel J Sebald
Subject: Re: extra head
Date: Fri, 16 Aug 2013 12:22:17 -0500
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.24) Gecko/20111108 Fedora/3.1.16-1.fc14 Thunderbird/3.1.16

On 08/16/2013 09:12 AM, Ben Abbott wrote:
On Aug 16, 2013, at 9:11 AM, Jordi Gutiérrez Hermoso wrote:

On 16 August 2013 01:15, Rik<address@hidden>  wrote:
On 08/15/2013 04:49 PM, Ben Abbott wrote:
Rik / Jordi,

After pushing, I noticed there is an extra head with no tag or bookmark.  I 
recall Jordi mentioned this is easily fixed, but I don't recall how.

$ hg heads
changeset:   17260:92effb035b97
bookmark:    @
tag:         tip
parent:      17254:7fb4461997aa
user:        Ben Abbott<address@hidden>
date:        Thu Aug 15 19:39:55 2013 -0400
summary:     Update axes position/outerpostion before and after tightinset.

changeset:   17259:a654c1eccd29
user:        Stefan Mahr<address@hidden>
date:        Thu Aug 15 14:38:01 2013 +0200
summary:     doc: Modify chapter 'continuation lines' to fit actual behaviour

changeset:   17112:eff3d78367f0
branch:      stable
parent:      17108:2c76db746701
user:        Jordi Gutiérrez Hermoso<address@hidden>
date:        Tue Jul 30 13:20:58 2013 -0400
summary:     * Makefile.am (doxyhtml): Restore doxyhtml argument in recursive 
make call

changeset:   16896:498b2dd1bd56
branch:      classdef
parent:      16698:13b3b92ea99c
parent:      16895:b8c37a855074
user:        John W. Eaton<address@hidden>
date:        Thu Jul 04 10:09:58 2013 -0400
summary:     periodic merge of default to classdef


Jordi might have a better solution but I just merged your head onto the
development branch to get rid of it.


This is the correct solution.

However, Ben, how did you manage to push an extra head? Typically "hg
push" will not allow you to do this unless you do --force.

Jordi, it might be the -C clean option that allows the push to continue. Here's the rule from the "man hg" documentation:

1. If  neither  -c/--check  nor  -C/--clean  is  specified,  and if the
   requested changeset is an ancestor  or  descendant  of  the  working
   directory's  parent,  the  uncommitted  changes  are merged into the
   requested changeset and the merged result is  left  uncommitted.  If
   the  requested  changeset is not an ancestor or descendant (that is,
   it is on another branch), the update is aborted and the  uncommitted
   changes are preserved.

Does the -C also apply to the second sentence?

In addition, I find it odd that Michael would get a merge complaint:

[pbdsl3:octave] hg -v pull
real URL is http://hg.savannah.gnu.org/hgweb/octave
pulling from http://www.octave.org/hg/octave
searching for changes
no changes found
divergent bookmark @ stored as @default
[pbdsl3:octave] hg -v update
abort: crosses branches (merge branches or update --check to force update)

From the third party perspective, shouldn't the extra head just seem like another branch? (This complaint is what I've found whenever I've forgotten to do a "rollback"... I just clone a new copy when I've done that.)

Ben, is this the first push you've done since starting to use the bookmarks? I think the clue lies in this statement:

divergent bookmark @ stored as @default

Could it be that hg doesn't know what bookmark you are referring to? So maybe it assumes the bookmark should be associated with the default branch? That is, there is this bookmark that somehow became "divergent". I wonder how this could become become divergent? Might it be that the patch you imported was created prior to using the bookmark feature?

hg pull
hg update -C @
hg import changeset.patch
hg outgoing
hg push ssh://address@hidden/octave
hg rollback
hg revert --all
hg pull
hg update -C @

It might be good to pause here a moment with further pushes. I see in the log

http://hg.savannah.gnu.org/hgweb/octave/

that John D's latest push now has an @ symbol attached. I don't recall seeing "default tip @" in the past. Did Rik's merge perhaps bookmark the tip? Bookmarking the tip doesn't make sense, does it?

Dan


reply via email to

[Prev in Thread] Current Thread [Next in Thread]