Closed WORMSS closed 10 years ago
I know windows a few versions behind v2.1.3 and I am not sure if the latest versions of git have this bug (I currently do not know how to make a build, neither do I have a *nix machine up and ready).
Please try the Vagrant method.
I looked in the code, (was quicker), 2.2.0 still assumes the first merge is a fastforward.
So it seems this isn't a Windows specific issue, then? If so, this discussion should be moved to the upstream Git mailing-list.
Yep, sorry, just looking into the mailing list now. git/git does not have a issue tracker.. and there seems to be two for git-scm (google groups and mailing list).. So I am just researching everything now.
Thank you for your help.
I looked in the code, (was quicker), 2.2.0 still assumes the first merge is a fastforward.
I doubt it is quicker because you will have a harder time to contribute efficiently now. If you had chosen not to ignore my advice, you would probably already have a patch ready.
I can only just about read this code, let alone contribute a patch for it. Syntax to me is very alien. So a lot of what is going on I am guessing. I cannot even work out where --no-ff is being treated. I presume on a level above git-merge-octopus.sh but working that out is a bit harder than working out where some hardcoded output is coming from.
git version 1.9.4.msysgit.2
The output of git merge --no-ff a b c still states that the merge has done a fast forward. Consulting the log I can clearly see it has not done a fast forward. So this is purely a output bug rather than implementation.
I know windows a few versions behind v2.1.3 and I am not sure if the latest versions of git have this bug (I currently do not know how to make a build, neither do I have a *nix machine up and ready).
I reset master.
As you can see, it still says Fast-forwarding to: a when it should not.