0

Possible Duplicate:
Creating GitHub repository with only a subset of a local repository's history

I am preparing an internal project for an open-source release on GitHub, and I am trying to find the best way to clean up the repository. The issue is that we have a long, messy commit history that is riddled with internal passwords, references, and hacks, and would like to start the release from a single, current commit. I understand that I could just squash the commits, or recreate the repository from the working copy files to achieve this, but I would also like to maintain the original history locally. Ideally, I would like to create a branch with a single starting commit that still can be merged to from the master branch so that all future commits can be merged into this staging branch and then periodically pushed to GitHub.

Is this reasonable? If so, what's the best way to achieve this. If not, do you have any suggestions for other ways to address this?

Community
  • 1
  • 1
jmans
  • 5,648
  • 4
  • 27
  • 32

1 Answers1

1

You can start a new root commit of everything that you need including any clean up of passwords etc.

From then on, I would look at

  1. Smudge/clean scripts to manage env variances wrt passwords, config, etc.
  2. A good branching strategy like git flow or https://plus.google.com/109096274754593704906/posts/R4qkeyRadLR
  3. Take a look at git filter-branch to help you clean up your old history from storing sensitive information
user229044
  • 232,980
  • 40
  • 330
  • 338
Adam Dymitruk
  • 124,556
  • 26
  • 146
  • 141