DragonFly- master sys/vfs/hammer hammer.h

Simon 'corecode' Schubert corecode at fs.ei.tum.de
Fri Jan 9 15:53:36 PST 2009

Matthew Dillon wrote:
:commit 08abcb65fdb57d8fc161c68b56017a3e7bc7f634
:Merge: 8c9c78e20eec172ac69dbb2e227c049edcd73a85 f7e3064f83c9a8e68c1f1782081e09631fb49a67
:Author: Matthew Dillon <dillon at apollo.backplane.com>
:Date:   Sat Jan 3 11:38:36 2009 -0800
:    Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly into devel
:Summary of changes:
    Say what?  What did this do?
Aahahahaha.  Says the one who always pushes merge commits :)

Take a look with gitk.  You'll see a ton of branches and merges.  This 
happens every time you say "git pull crater" after you committed something 
and somebody has pushed since your last pull.  At that point both repos 
(crater and yours) have diverged, and git merges these changes.

This is like what cvs does when you have changes in your working tree and 
you do a cvs up.  The difference with git is that you already did a local 
commit, so it can't modify this state.  Instead it adds another commit 
which has two parents.

Now, most of the time changes don't touch the same files.  My commit mail 
script (well, my modifications) detects this case (if it is a zero net 
diff commit) and doesn't send a mail.  The commit stays, of course.

However if there were concurrent changes to the same file, it seems that 
my script still decides to push this mail.  Maybe it's better this way, 
maybe confusing.

Before anybody starts screaming "oh my GOD, automatic merges AND 
commits!":  git will complain loudly if there were conflicts, so this 
isn't any worse than doing a cvs up before doing a cvs ci.

  <3 the future  +++  RENT this banner advert  +++   ASCII Ribbon   /"\
  rock the past  +++  space for low €€€ NOW!1  +++     Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \

More information about the Commits mailing list