skip to content.
view text-only.

SVN Repositories

Most of my projects are kept in individual SVN repositories to maintain a source history and make them easily available.

I've chosen to use one repository per project to avoid artificially inflating revision numbers. Many projects change very rarely; I didn't want to check out revision 4000 when the code hasn't changed since 200.

Repository Access

All the repositories are stored on svn://jamessocol.com/, and are read-only for unauthenticated users.

The individual repositories are named after their projects. For example, Better Search Widget is in svn://jamessocol.com/better-search-widget.

I have not, however, been completely consistent on using dashes or underscores.

Repository Structure

Most repositories have two main subdirectories: trunk and tags. Some may have branches.

The trunk is usually stable, since most of the current projects are very small and the changes are fast and easy. However, it is bleeding-edge and there's no guarantee trunk will compile, let alone run stably.

When I commit any finished change, I do my best to remember to augment the minor revision number (ie: 1.1.x) and tag a new version in the tags directory. I do, from time to time, forget to do one of these things.

Once I tag a version, it won't change, bugs and all. But they are tested and stable enough for me to use.

One exception: sometimes I forget to change the version number in the source code before I tag the new version. I do go back and update that. Mostly.

wordpress

I've written a couple of WordPress plugins.