You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@subversion.apache.org by "Karan, Cem (Civ, ARL/CISD)" <CK...@arl.army.mil> on 2005/03/25 15:50:41 UTC

What is the algorithm for 'svn merge'?

Hi all, I'm hoping someone can explain this to me. What is the algorithm
behind svn merge? I ask because I see that svn can automatically merge
changes. I can see that this is easy to do if the changes between two
files are actually identical, but how does it handle things when there
are differences? How does it decide which changes go into the files and
which get left out? I just want to make sure that my C code doesn't
suddenly start acting funky due to a merge...
 
Thanks,
Cem Karan

Re: What is the algorithm for 'svn merge'?

Posted by Ben Collins-Sussman <su...@collab.net>.
On Mar 25, 2005, at 9:50 AM, Karan, Cem ((Civ, ARL/CISD)) wrote:

> Hi all, I'm hoping someone can explain this to me. What is the 
> algorithm behind svn merge? I ask because I see that svn can 
> automatically merge changes. I can see that this is easy to do if the 
> changes between two files are actually identical, but how does it 
> handle things when there are differences? How does it decide which 
> changes go into the files and which get left out? I just want to make 
> sure that my C code doesn't suddenly start acting funky due to a 
> merge...

It's a basic diff3 algorithm.  'man diff3' to learn about it and play 
with GNU's implementation of diff3.


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org