lilypond-devel
[Top][All Lists]
Advanced

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

Re: Bad translation merge


From: David Kastrup
Subject: Re: Bad translation merge
Date: Wed, 07 Mar 2012 16:06:35 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.0.92 (gnu/linux)

David Kastrup <address@hidden> writes:

> Julien Rioux <address@hidden> writes:
>
>> David Kastrup <dak <at> gnu.org> writes:
>>> I can reproduce the problems when merging.  It would appear that the
>>> history of the translation branch got messed up at some point of time in
>>> a manner that git can't recognize how to merge properly anymore.
>>
>> I don't know if this can help you, but I noticed the following. Usually
>> master is merged into lilypond/translation and then lilypond/translation
>> is merged into master.
>
> It should be merged into _staging_ (which will eventually pass into
> master).  I don't remember any chaos with regard to staging->master, so
> tampering with master was not likely a problem.
>
>> If you look at the history about 11 days ago, you can see that master
>> was merged into lilypond/translation, so far so good, but then it
>> looks like lilypond/translation~0 was rebased on top of master,
>> instead of merged into master.
>
> I see rebasing commits like
>
> commit 24f5f986998c23a1cbac15024d58ca6497093cce
> Author: Julien Rioux <address@hidden>
> Commit: Francisco Vila <address@hidden>
>
>     Doc-de: Compilation fix for de/notation.
>
> commit 12cfe6bafb8589b0780df84fb36981994ee8793a
> Author: Till Paala <address@hidden>
> Commit: Francisco Vila <address@hidden>
>
>     Doc-de: update the noation manual, adding snippet translations
>
> which also appear in a version committed by their actual authors.
> That's definitely not good.  But I don't see how this would explain the
> _loss_ of changes.  Digging further.

I don't get it.  If I take a look at scm (where some changes happened)
with

git log --date-order --graph --source origin origin/lilypond/translation scm

then the graph outputs

*   commit 9d1520b21710bd22872010ae9aa4c4899014e9d4     origin
|\  Merge: e995ed4 10af932
| | Author: Francisco Vila <address@hidden>
| | Date:   Wed Mar 7 10:23:11 2012 +0100
| | 
| |     Merge branch 'lilypond/translation' into staging
| |     
| |     Conflicts:
| |             Documentation/changes.tely
| |             Documentation/translations.itexi
| |             Documentation/web/news-front.itexi
| |             VERSION
| |             input/regression/layout-from.ly
| |             lily/context-def.cc
| |   
* | commit e995ed461610c2bb9c9cd43eaa715905b8525b95     origin
| | Author: David Kastrup <address@hidden>
| | Date:   Sun Feb 26 19:16:00 2012 +0100
| | 
| |     Allow music with layout instructions in output definitions.
| |     
| |     This allows things like
| |     
| |       \layout { \accidentalStyle modern }
| |     
| |     or
| |     
| |       \midi { \tempo 4 = 80 }
| |     
| |     to work as intended.
| |   
* | commit 4d7cac2e94f7fb8c86c69c6d6fceff1ddfb545e4     origin
| | Author: Carl Sorensen <address@hidden>
| | Date:   Sat Feb 18 09:49:56 2012 -0700
| | 
| |     Set context properties to allow half-measure beaming in 3/4 (Issue 2246)
| |     
| |     Introduce beamWholeMeasure, which allows beaming of a whole measure of
| |     eighth notes in 3/4 time (but not beaming of a half measure, since it
| |     appears to be a 6/8 beaming).  By default, set to #t.
| |     
| |     Introduce beamHalfMeasure, which allows beaming of a half measure of
| |     eighth notes in 3/4 time.  By default, set to #f.
| |     
| |     A non-empty beamExceptions entry will override beamHalfMeasure and
| |     beamWholeMeasure, since user-specified rules should override defaults.
| |     
| |     Includes updated documentation in English docs, and a new regtest.  No
| |     changes to translated docs.
| |   
* | commit c37c6f39d18274ccac28ed42559681ea271cc496     origin
| | Author: David Kastrup <address@hidden>
| | Date:   Thu Feb 23 19:19:37 2012 +0100
| | 
| |     Issue 2343: Faulty file-naming when outputting multiple \books
| |   
* | commit 09814b549186893c265bcdf835edbe242f6354cf     origin
|/  Author: David Kastrup <address@hidden>
|   Date:   Sun Feb 26 11:23:39 2012 +0100
|   
|       Implement ly:book? and ly:context-def? predicates
|  
* commit 10af9324c6d94453857e10951c4f58043e9a9f11       origin/lilypond/translat
...

So starting from the (unfortunately common) ancestor, the only branch
that has actual changes in scm is master.  Why does the merge then
prefer the old versions?

That does not make all that much sense to me.

-- 
David Kastrup




reply via email to

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