September 18, 2014
Hot Topics:
RSS RSS feed Download our iPhone app

Rules for Effective Source Code Control

  • March 25, 2004
  • By Mike Gunderloy
  • Send Email »
  • More Articles »

Rule #3: Don't Hog the Files

The best developers make frequent small changes to the repository, rather than a few huge changes. In a check-in/edit/check-out system (like Visual SourceSafe) this means only checking out a few files at a time. In an edit/merge/continue system (like CVS) this means committing changes after each task, rather than waiting until the end of a day or even longer.

Minimizing the size of changes has several good effects. First, if you are working with a system that locks files, you can avoid locking other people out for longer than necessary. Second, by keeping your commits small, you vastly lower the chance of needing to merge two incompatible versions of code. Finally, by working in small chunks, you can keep the comments in the source code control system targeted and informative.

And of course, you should strive to have an informative check-in comment for every check-in. If you can't think of a way to summarize your changes, why did you even make them?

Rule #4: Use Labels and Branches Wisely

Labels (sometimes called tags) give you a way to mark a versioned set of files with some friendly name. This is useful because humans are much better at remembering things like "Beta 2" than "Version 3019.4". You should apply a label to your source code control repository at any significant point in time. This includes:

  • Public releases of the software
  • The start and finish of major code changes
  • The incorporation of a new third-party component
  • Builds that pass basic QA testing

Effective use of labels makes it easy to jump back to a particular point in time, whether to undo a terrible mistake or to see what state the system was in at that time.

By contrast, branches give you a way to develop two sets of source code for the same project in parallel. You should create a branch whenever different developers in the same project are following different rules. For example, if one team is finishing up version 1, while another is starting on version 2, that's a good time for a branch. Presumably the first group of developers is checking in small, careful changes, while the other is roughing out the broad outlines of new features. Because the rules for these activities are different, they can't easily be performed in a single set of source code files.

Branching is useful for exploration as well. If your main line of code is using a known, stable technique, but you want to explore a faster and potentially unstable way of accomplishing the same objectives, you should create a branch to explore the consequences. If the branch works out well, merge its changes back to the main line. If not, just stop working in the branch.

Finally, don't create a branch just because you know you'll need it in the future. Wait to make the version 2.0 branch until someone is actually going to work on it, to avoid extra merging before it's really needed.

A Word to the Wise

Source code control is one of those basic practices that every developer should use on every project. With the availability of several excellent free systems, there's no excuse for working without the source code control safety net. After you get used to storing files in a source code control repository, take the time to learn what else your system can do for you. You might just be surprised at the difference that it makes to your development efforts.

Mike Gunderloy is the author of over 20 books and numerous articles on development topics, and the lead developer for Larkware. Check out his new Sybex book Coder to Developer, which includes more advice on source code control and a wealth of other topicsx. When he's not writing code, Mike putters in the garden on his farm in eastern Washington state.





Page 2 of 2



Comment and Contribute

 


(Maximum characters: 1200). You have characters left.

 

 


Sitemap | Contact Us

Rocket Fuel