On 2014-08-12 01:10:28 Andrew Crabtree wrote:
> I'm seeing some oddity in one of my repositories where the root commit is 
> being output in 'rev-list' even
> when I pass in a reference that should exclude it from being output.
...
> Seeing the issue with versions of git from 1.7 to 2.1.

I have a small test repo on GitHub that demonstrates the problem.  I expect it 
is caused by the commit dates being out of sequence:

1. Clone the "revlist" repo:

# git clone https://github.com/pcharlan/revlist.git                             
Cloning into 'revlist'...

2. master has six commits (with out-of-order dates):

# git log --pretty='%h %cd' master
0f9e594 Sun May 20 01:01:01 2007 -0700
1ad887b Sun May 20 01:01:01 2007 -0700
cde7188 Sun May 20 01:01:01 2007 -0700
51af1c0 Sun May 20 01:01:01 2007 -0700   <-- date seven years ago
0e080e8 Tue Aug 12 11:51:19 2014 -0700  <-- present day
77ce973 Fri May 20 01:01:01 2011 -0700  <-- three years ago

3. Pruning master with origin/other (a branch with the same root and also 
out-of-order dates) has no effect, counter to expectation:

# git log --pretty='%h %cd' master ^origin/other
0f9e594 Sun May 20 01:01:01 2007 -0700
1ad887b Sun May 20 01:01:01 2007 -0700
cde7188 Sun May 20 01:01:01 2007 -0700
51af1c0 Sun May 20 01:01:01 2007 -0700
0e080e8 Tue Aug 12 11:51:19 2014 -0700
77ce973 Fri May 20 01:01:01 2011 -0700     <--- Reachable from origin/other

4. Pruning master with the parent of origin/other removes the root commit:

# git log --pretty='%h %cd' master ^origin/other~
0f9e594 Sun May 20 01:01:01 2007 -0700
1ad887b Sun May 20 01:01:01 2007 -0700
cde7188 Sun May 20 01:01:01 2007 -0700
51af1c0 Sun May 20 01:01:01 2007 -0700
0e080e8 Tue Aug 12 11:51:19 2014 -0700

Q: Is this expected behavior?  If so, and assuming that it's the dates that are 
throwing Git off, is there a way to tell Git to not care about the dates, or to 
care less than it does?

Thanks,

--Pete

Peter Harlan
HP Networking
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to