[Bioc-devel] Your branch and 'bioc/master' have diverged

Fri Mar 11 17:52:48 CET 2016

----- Original Message -----
> From: "Dan Tenenbaum" <dtenenba at fredhutch.org>
> To: "Luca De Sano" <l.desano at campus.unimib.it>
> Cc: "bioc-devel" <bioc-devel at r-project.org>
> Sent: Friday, March 11, 2016 8:33:58 AM
> Subject: Re: [Bioc-devel] Your branch and 'bioc/master' have diverged

> ----- Original Message -----
>> From: "Luca De Sano" <l.desano at campus.unimib.it>
>> To: "Dan Tenenbaum" <dtenenba at fredhutch.org>
>> Cc: "bioc-devel" <bioc-devel at r-project.org>
>> Sent: Friday, March 11, 2016 4:20:59 AM
>> Subject: Re: [Bioc-devel] Your branch and 'bioc/master' have diverged
> 
>> But, after the what you suggest, git and svn be will synchronized?
>> 
> 
> You would basically need to cherry pick every time you want to get commits from
> git to svn and vice versa.
> 

A better solution might be to delete your git repository (after updating svn with the latest as described earlier), then re-create your github repository, and keep it in sync with svn using this technique:

https://github.com/Bioconductor/BiocGithubHelp/wiki/Managing-your-Bioc-code-on-hedgehog-and-github#appendix-i-a-simpler-way

It's also probably best to avoid any non-linear commit history in git.

Dan

> Dan
> 
> 
>> 
>> --
>> Luca De Sano
>> Research Associate
>> 
>> Department of Informatics, Systems and Communication
>> University of Milan Bicocca
>> 
>> Institute of Biomembranes and Bioenergetics of the Italian National
>> Research Council (IBBE-CNR)
>> 
>> email: l.desano at campus.unimib.it
>> PEC: luca.desano at pec.it
>> 
>> On 10 March 2016 at 19:47, Dan Tenenbaum <dtenenba at fredhutch.org> wrote:
>> 
>>>
>>>
>>> ----- Original Message -----
>>> > From: "Luca De Sano" <l.desano at campus.unimib.it>
>>> > To: "Dan Tenenbaum" <dtenenba at fredhutch.org>
>>> > Cc: "bioc-devel" <bioc-devel at r-project.org>
>>> > Sent: Thursday, March 10, 2016 10:21:38 AM
>>> > Subject: Re: [Bioc-devel] Your branch and 'bioc/master' have diverged
>>>
>>> > On 10 March 2016 at 18:17, Dan Tenenbaum <dtenenba at fredhutch.org> wrote:
>>> >
>>> >>
>>> >>
>>> >> ----- Original Message -----
>>> >> > From: "Luca De Sano" <l.desano at campus.unimib.it>
>>> >> > To: "bioc-devel" <bioc-devel at r-project.org>
>>> >> > Sent: Thursday, March 10, 2016 3:22:14 AM
>>> >> > Subject: [Bioc-devel] Your branch and 'bioc/master' have diverged
>>> >>
>>> >> > Hi all,
>>> >> > I'm trying to update the devel branch of my package following the
>>> >> > instruction found in
>>> >> https://bioconductor.org/developers/how-to/git-mirrors/
>>> >> > .
>>> >> >
>>> >> > The repo of my project is: https://github.com/BIMIB-DISCo/TRONCO
>>> >> >
>>> >> >>From devel I've first tried to do a git merge master, with no results.
>>> >> I've
>>> >> > then tried the cherry-pick method and all seemed to work well.
>>> However,
>>> >> > during the git-svn dcommit, I've obtained a:
>>> >> >
>>> >> > svn: Commit blocked by pre-commit hook (exit code 1) with output:
>>> >> > The y in the x.y.z version number should be even in release
>>> >> > in file branches/RELEASE_3_2/madman/Rpacks/VegaMC/DESCRIPTION.
>>> >> > See http://bioconductor.org/developers/how-to/version-numbering/
>>> >> >
>>> >>
>>> >>
>>> >> Did you really get this commit error message about trying to commit to
>>> the
>>> >> devel branch of TRONCO? The message refers to the _release_ branch of
>>> >> _VegaMC_. If you did get that error message in response to a commit of
>>> >> TRONCO, then something is very wrong with that pre-commit hook.
>>> >>
>>> >
>>> > No. Sorry was just a cut&paste error, I was searching the mailing list
>>> for
>>> > similar errors ... The message was like:
>>> >
>>> > svn: Commit blocked by pre-commit hook (exit code 1) with output:
>>> > The y in the x.y.z version number should be even in release
>>> > in file [some path which I don't remember at the
>>> > moment]/madman/Rpacks/TRONCO/DESCRIPTION.
>>> > See http://bioconductor.org/developers/how-to/version-numbering/
>>> >
>>> >
>>> >
>>> >>
>>> >>
>>> >>
>>> >> > There has been a problem with a commit (a wrong version update which
>>> was
>>> >> > rejected from svn).
>>> >> >
>>> >> > I've fixed this problem on my master removing the wrong commit as
>>> >> described
>>> >> > here http://www.clock.co.uk/blog/deleting-a-git-commit . I've then
>>> >> tried to
>>> >> > bring those changes on devel doing this:
>>> >> >
>>> >>
>>> >> Which commit was 'wrong'?
>>> >>
>>> >
>>> > The problem was a wrong version update in a commit made one month ago.
>>> This
>>> > commit triggered the pre-commit hook and, even If I've fixed it with a
>>> new
>>> > commit, the problem has remained. So, following the instructions
>>> mentioned
>>> > above, I've removed the commit from the history tree.
>>> >
>>> >
>>> >
>>> >>
>>> >>
>>> >> >   - git reset --hard on devel to the last right commit
>>> >> >   - git cherry-pick on the others commit from my master branch
>>> >> >
>>> >> > After the cherry-pick, on git, all worked well. The resul of git log
>>> is:
>>> >> >
>>> >> > dex at darthvader ~/project/TRONCO $ git log --graph --decorate
>>> >> > --pretty=oneline --abbrev-commit --all
>>> >> > *   2e27604 (HEAD, origin/master, origin/development, origin/HEAD,
>>> >> master,
>>> >> > development, devel) Merge pull request #47 from
>>> BIMIB-DISCo/development
>>> >> >|\
>>> >> >| * 340c809 date update
>>> >> >|/
>>> >> > * 4700191 dependencies update
>>> >> > * 31c2cab regex problems
>>> >> > [...]
>>> >> >
>>> >> > I've then tried to commit to svn, but the git svn dcommit
>>> >> --add-author-from
>>> >> > now isn't workig...
>>> >> >
>>> >> > dex at darthvader ~/project/TRONCO $ git svn dcommit --add-author-from
>>> >> > Unable to determine upstream SVN information from HEAD history.
>>> >> > Perhaps the repository is empty. at /usr/lib/git-core/git-svn line
>>> 856.
>>> >> >
>>> >> > And the result of a git status is:
>>> >> >
>>> >> > dex at darthvader ~/project/TRONCO $ git status
>>> >> > On branch devel
>>> >> > Your branch and 'bioc/master' have diverged,
>>> >> > and have 615 and 25 different commits each, respectively.
>>> >> >  (use "git pull" to merge the remote branch into yours)
>>> >> >
>>> >> > nothing to commit, working directory clean
>>> >> >
>>> >> >
>>> >> > The situation at the moment is:
>>> >> >
>>> >> >
>>> >> >   -
>>> >> https://hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/TRONCO/R/
>>> >> >   is a mix of old and new version... I don't know why. Some files are
>>> up
>>> >> to
>>> >> >   date with my master and some others no.
>>> >> >
>>> >> >   - My master ( https://github.com/BIMIB-DISCo/TRONCO/tree/master )
>>> is
>>> >> up
>>> >> >   to date and I want to bring this on devel
>>> >> >
>>> >> > Any idea how to fix this?
>>> >>
>>> >>
>>> >> Basically, cherry-pick any commit that is in master but not devel. Then
>>> >> try git-svn dcommit.
>>> >>
>>> >> Dan
>>> >>
>>> >
>>> > This is not possible because some commits has been applyied not entirely
>>> to
>>> > devel. In
>>> > https://hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/TRONCO/ some
>>> > files seem to be updated with master and some others no. For example open
>>> > as.functions.R and caprese.algorith.R: both have been modified with
>>> >
>>> https://github.com/BIMIB-DISCo/TRONCO/commit/44e3d7a3f361cbe38227d1c74029afb6e6449e3d
>>> > , but only as.functions.R seems to be up to date.
>>> >
>>> > Can you simply revert everything on svn and on the git mirror to this
>>> > commit you made?
>>> >
>>> https://github.com/BIMIB-DISCo/TRONCO/commit/af70ecaba498642e7e19697c51b856cbe3bba8be
>>> >
>>> >
>>>
>>> I can but I am not sure it would solve your problem.
>>>
>>> Another thing you could do is simply:
>>>
>>> - check out the package using svn
>>> - copy the contents of your git repository over the files
>>>   in the svn working copy
>>> - svn commit the svn working copy.
>>>
>>> You might still have issues in the future when trying to merge from svn or
>>> commit from git. But this would solve your immediate problem.
>>>
>>> Dan
>>>
>>>
>>>
>>> >
>>> >
>>> >
>>> >
>>> >
>>> >
>>> >
>>> >>
>>> >>
>>> >> > Many thanks!
>>> >> >
>>> >> >
>>> >> > --
>>> >> > Luca De Sano
>>> >> > Research Associate
>>> >> >
>>> >> > Department of Informatics, Systems and Communication
>>> >> > University of Milan Bicocca
>>> >> >
>>> >> > Institute of Biomembranes and Bioenergetics of the Italian National
>>> >> > Research Council (IBBE-CNR)
>>> >> >
>>> >> > email: l.desano at campus.unimib.it
>>> >> > PEC: luca.desano at pec.it
>>> >> >
>>> >> >       [[alternative HTML version deleted]]
>>> >> >
>>> >> > _______________________________________________
>>> >> > Bioc-devel at r-project.org mailing list
>>> >> > https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>> >>
>>> >
>>> >
>>> >
>>> > --
>>> > Luca De Sano
>>> > Research Associate
>>> >
>>> > Department of Informatics, Systems and Communication
>>> > University of Milan Bicocca
>>> >
>>> > Institute of Biomembranes and Bioenergetics of the Italian National
>>> > Research Council (IBBE-CNR)
>>> >
>>> > email: l.desano at campus.unimib.it
>>> > PEC: luca.desano at pec.it
> 
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel