Hi all,

I'm wondering if you could give me input on the suggested workflow when
using Git on the client, a Git server repo and Subversion.

I've been using Git for my work projects for a couple of weeks now and I
really like the advantages it offers. We use SVN as our primary source
control system at work, so I rely heavily on git svn rebase and git svn
dcommit.

I actually have two workplaces, one at the office and one at home. Sometimes
while I work on some Git topic branch I want to take those changes home and
finish my work there. To make this happen I created a repo at Github to
store my work in progress that's not yet ripe for SVN. Every time I switch
workplaces I do a git push (to Github). After arriving at the new location,
I run git pull to get my WIP changes. Every now and then it happens that
there are some SVN commits made by my colleagues that I already pushed to
Github. But if I arrive at the new location and I do a git svn rebase those
SVN commits are applied again.

The make things a bit better to understand:
1. At work
Starting at SVN commit A
work...
git commit
Colleage commits to SVN in the meantime: SVN commit B
git svn rebase
git push
(I leave)
Colleague commits to SVN: SVN commit C

2. At home:
git pull origin master
git svn rebase (to apply the colleague's changes in SVN commit C)

Not only do my colleague's latest changes get applied (SVN commit C), but
also those changes from SVN commit B will be rebased, often resulting in
merge conflicts.

Is there a way to omit SVN commit B during the SVN rebase? As matter of fact
it's already in my Git repo (and the according git-svn-id is in the log).

Thank you for any hints,

Alex
--
Alexander Groß
http://therightstuff.de/

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to