0

So for some reason (miscommunication for what the actual subversion URL is and only 1 person working on the code previously) we have two different subversion repositories with what we want to be the same code. Basically the timeline went like this:

  1. Created X repository
  2. Put code in
  3. Created Y repository
  4. Put code from X in
  5. Made modifications to Y
  6. SVN repo Y got corrupted (so he just stopped committing.. not bothering to say something before the backups replaced themselves)
  7. I made changes to X

And, so now I need to somehow merge the changes from Y into X. What would be the best method for doing this?

Earlz
  • 62,085
  • 98
  • 303
  • 499
  • OT: just out of interest: how can you corrupt a repository? What is your error message? Which revisions are broken? All? I am very interested, as I am searching for an example of a real life corruption of a svn repository – Peter Parker Sep 27 '11 at 18:21
  • @Peter `corrupt node-revision 'm2g-5.0.r5/[long number]' missing id field in node-rev` or `'REPORT of '/svn/MyProject/!svn/vcc/default': could not read chunk size` or a few other error messages – Earlz Sep 27 '11 at 19:11

1 Answers1

1
  • create a branch from x at the revision where y was created
  • put the code from y in that branch
  • commit
  • merge the 'branch-y' to your trunk in x
jeroenh
  • 26,362
  • 10
  • 73
  • 104