Can't push upstream - duplicate commits

0

Entering edit mode

spollack • 0

@spollack-12387

Last seen 6.1 years ago

Biostatistics and Computational Biology…

Dear Bioconductor,

I am trying to fix the error that our package bumphunter is producing. The error is due to a data file that is used in the tests which was moved and has a new URL. But when I try to push to upstream, I get an error about duplicate commits. I did this: (i) followed the instructions in the page "force Bioconductor master to GitHub master" (ii) cherry-picked the commit from master_deprecated which fixes the bug and (iii) tried to push upstream. But I'm still getting the same error.

I can't figure out what to do. Do I have to rebase somewhere?

I have the git log --one-line --graph output I can send if that would be informative.

Any advice you can offer would be extremely helpful.

thanks,

- Sam

Duplicate commits:

commit b403a19b826e6f1247c3db382b39eef6e4939314
Author: hpages@fhcrc.org <hpages@fhcrc.org@bc3139a8-67e5-0310-9ffc-ced21a209358>
Date:   Mon Apr 24 19:50:57 2017 +0000
    bump x.y.z versions to odd y after creation of 3_5 branch
    git-svn-id: https://hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/bumphunter@129129 bc3139a8-67e5-0310-9ffc-ced21a209358

commit 455927c2a83a7358ec071c4417e75eedfa124955
Author: Herve Pages <hpages@fhcrc.org>
Date:   Mon Apr 24 19:50:57 2017 +0000
    bump x.y.z versions to odd y after creation of 3_5 branch
    git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/bumphunter@129129 bc3139a8-67e5-0310-9ffc-ced21a209358

git • 1.6k views

ADD COMMENT • link updated 6.1 years ago by Mike Smith ★ 6.6k • written 6.1 years ago by spollack • 0

0

Entering edit mode

Hi Sam,

I took a look at bumphunter. There are a couple of things I wanted to say.

First, it seems like this commit was the one creating all the duplicate commits for the package.

718d3d8 Merge with Bioconductor devel brach

$$ git show 718d3d8

commit 718d3d87af114f8dc404c3889c673fac861bb84c
Merge: b403a19 031c450
Author: Samuela Pollack <sampoll@users.noreply.github.com>
Date:   Fri May 5 16:21:39 2017 -0400

    Merge with Bioconductor devel brach

diff --cc DESCRIPTION
index b70f47d,0ff0369..f1719e1
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@@ -1,5 -1,5 +1,5 @@@
  Package: bumphunter
- Version: 1.17.0
 -Version: 1.11.6
++Version: 1.17.1
  Title: Bump Hunter
  Description: Tools for finding bumps in genomic data
  Authors@R: c(person(c("Rafael", "A."), "Irizarry", role = c("cre","aut"), email

You can see that there is a "chain" of duplicate commits all the way to the very beginning of the commit history.

(You can try, git log --oneline to view the duplicates)

Second, the way to fix duplicate commits is given at this location on the bioconductor website. http://bioconductor.org/developers/how-to/git/resolve-duplicate-commits/.

bumphunter was notified during the last release of the duplicate commit issue https://stat.ethz.ch/pipermail/bioc-devel/2018-April/013469.html. I believe it will be easier going forward if there is effort made to clean up the commit history.

If you have any questions, please feel free to let me know. <bioc-devel> mailing list is a better venue for this thread.

ADD REPLY • link 6.1 years ago nitesh.turaga ▴ 140

0

Entering edit mode

OK, thanks, I'll look at it.

ADD REPLY • link 6.1 years ago spollack • 0

3

Entering edit mode

Mike Smith ★ 6.6k

@mike-smith

Last seen 11 hours ago

EMBL Heidelberg

This support forum is principally intended for user questions.

For package development & maintainance help please use the Developer Mailing list at https://stat.ethz.ch/mailman/listinfo/bioc-devel

You'll get a more rapid response to git related issues there.

ADD COMMENT • link 6.1 years ago Mike Smith ★ 6.6k

0

Entering edit mode

If it really bugs you, you could try not reading it.

ADD REPLY • link 6.1 years ago spollack • 0

1

Entering edit mode

Questions about package development belong on the bioc-devel mailing list, where the experts on package development will see the post, and where others are more likely to see solutions to problems that they themselves experience during package development; Mike was primarily trying to be helpful. I asked Nitesh to respond here, so that your concern gets addressed.

ADD REPLY • link 6.1 years ago Martin Morgan 25k

Login before adding your answer.