Structure of Projects in Version Control

Question

I know there are at least 10 different ways to structure project in version control. I'm curious what some methods being used are and which ones work for you. I've worked with SVN, TFS, and currently/unfortunately VSS. I've seen version control implemented very poorly and just OK, but never great.

Just to get the ball rolling, here is a review of things I've seen.

This example is SVN-based, but applies to most VCS's (not so much to distributed version control).

1. branch the individual projects that are part of site /division/web/projectName/vb/src/[trunk|branches|tags]

2. branch the whole site, in the case I've seen, the whole site except for core components was branched. /division/[trunk|branches|tags]/web/projectName/vb/src/

3. Use main-line a default, only branch when necessary for huge changes.

Answer 1

We practice highly componentised development using Java, we have about 250 modules in trunk that have independent life cycles. Dependencies are managed through Maven (that's a best practice right there), every iteration (bi-weekly) actively developed modules get tagged with a new version. 3 digit version numbers with strict semantics (major.minor.build - major changes means backwards incompatible, minor changes mean backwards compatible and build number changes mean backwards and forwards compatible). Our ultimate software product is an assembly that pulls in dozens of individual modules, again as Maven dependencies.

We branch modules/assemblies when we need to make a bug fix or enhancement for a released version and we can not deliver the HEAD version. Having tagged all versions makes this easy to do but branches still incur a significant administrative overhead (specifically keeping branches in sync with certain HEAD changesets) that are partly caused by our tools, Subversion is sub-optimal for managing branches.

We find that a fairly flat and above all predictable tree structure in the repository is crucial. It has allowed us to build release tools that take away a lot of the pain and danger from a manual release process (updated release notes, project compiles, unit tests run through, tag is made, no SNAPSHOT dependencies, etc). Avoid putting too much categorization or other logic in your tree structure.

We roughly do something like the following:

svnrepo/
  trunk/
    modules/
      m1/ --> will result in jar file
      m2/
      ...
    assemblies/
      a1/
      ...
  tags/
    modules/
      m1/
        1.0.0/
        1.0.1/
        1.1.0/
       m2/
      ...
    assemblies/
      a1/
        iteration-55/
        ...
  branches/
    m1/
      1.0/
      ...

For external dependencies, I can not overemphasize something like Maven: manage your dependencies as references to versioned, uniquely identified binary artifacts in a repository.

For intenal module/project structure: stick to a standard. Uniformity is key. Again, Maven can help here since it dictates a structure. Many structures are fine, as long as you stick to them.

Answer 2

Example for SVN:

trunk/

branch/

tags/

The trunk should be kept at a point where you can always push a release from it. There should be no huge gaping bugs that you know about(of course there will be eventually but that is what you should strive for).

Every time you need to make a new feature, do a design change, whatever, branch. Tag that branch at the start. Then when you are finished with the branch tag it at the end. This helps out with merging back into trunk.

Every time you need to push a release, tag. This way if something goes horribly wrong you can rollback to the previous release.

This setup keeps trunk as clean as possible and allows you to make quick bug fixes and push them out while keeping the majority of your development in branches.

Edit: For 3rd party stuff it depends. If I can avoid it I do not have it under source control. I keep it in a directory outside source control and include it from there. For things like jquery, I do leave it under source control. The reason is it simplifies my script for pushing. I can simply have it do an svn export and rsync.