Code Metrics – Cyclomatic Complexity

Tips Search

Code Metrics – Cyclomatic Complexity

  • Comments 8

SKU: Premium, Ultimate

Versions: 2008, 2010

Code: vstipTool0131

 

When working with code metrics, one of the least understood items seems to be cyclomatic complexity. Essentially, with cyclomatic complexity, higher numbers are “bad” and lower numbers are “good”. We use cyclomatic complexity to get a sense of how hard any given code may be to test, maintain, or troubleshoot as well as an indication of how likely the code will be to produce errors. At a high level, we determine the value of cyclomatic complexity by counting the number of decisions made in our source code. In this tip we will start out with a simple example of cyclomatic complexity to understand the concept quickly then give some additional information on actual usage and suggested limits. Finally, I’ve added a section of citations that can be used to dig deeper into this subject.

 

Cyclomatic Complexity: A Simple Example

Cyclomatic complexity is defined as measuring “the amount of decision logic in a source code function” [NIST235]. Simply put the more decisions that have to be made in code, the more complex it is.

Let’s see it in action. Create a new console application and immediately calculate your code metrics by going to Analyze | Calculate Code Metrics for Solution:

image

 

Notice the cyclomatic complexity is at 2 (the lowest value possible). If I add non-decision code, notice the complexity doesn’t change:

image

 

 

If I add a decision, the cyclomatic complexity value goes up by 1:

image

 

When I change the if statement to a switch statement with 4 decisions to be made then it goes from the original 2 to 6:

image

 

Let’s take a look at a larger code base. Here are the metrics from the Tailspin Toys sample code (http://msdn.microsoft.com/en-us/library/aa645517(VS.71).aspx):

image

 

Notice that most of the items, as we drill down into the Products_Related class, have a value of 1 but a couple of them have a complexity of 5. By itself this may not be a big deal, but given that most other members have a 1 in the same class I would definitely look closer at those two items and see what is in them. You can do this by right-clicking the item and choosing Go To Source Code from the context menu. Let’s take a closer look at Product.set(Product):

image

 

Given all the if statements, we can see why the cyclomatic complexity is at a 5. At this point we may decide that this is an acceptable level of complexity or we might refactor to reduce the complexity.

 

 

The Magic Number

As with many metrics in our industry, there is no exact cyclomatic complexity limit that fits all organizations. However, [NIST235] does indicate that a limit of 10 is a good starting point:

“The precise number to use as a limit, however, remains somewhat controversial. The original limit of 10 as proposed by McCabe has significant supporting evidence, but limits as high as 15 have been used successfully as well. Limits over 10 should be reserved for projects that have several operational advantages over typical projects, for example experienced staff, formal design, a modern programming language, structured programming, code walkthroughs, and a comprehensive test plan. In other words, an organization can pick a complexity limit greater than 10, but only if it is sure it knows what it is doing and is willing to devote the additional testing effort required by more complex modules.” [NIST235]

 

 

Cyclomatic Complexity and Line Numbers

As I mentioned in vstipTool130 (“Code Metrics – Lines of Code”), just looking at the number of lines of code by itself is, at best, a very broad predictor of code quality. There is some basic truth to the idea that the more lines of code in a function, the more likely it is to have errors. However, when we combine cyclomatic complexity with lines of code then we have a much clearer picture of the potential for errors.

As described by the Software Assurance Technology Center (SATC) at NASA:

“The SATC has found the most effective evaluation is a combination of size and [Cyclomatic] complexity. The modules with both a high complexity and a large size tend to have the lowest reliability. Modules with low size and high complexity are also a reliability risk because they tend to be very terse code, which is difficult to change or modify.” [SATC]

 

 

Code Analysis

When using code analysis, the Design Guideline rule sets contain maintainability areas:

image

 

Inside the maintainability area is a rule for complexity:

image

 

This rule issues a warning when the cyclomatic complexity reaches 25 so it is a good thing to have to keep you from having excessive complexity. You can learn more about the rule here: http://msdn.microsoft.com/en-us/library/ms182212.aspx

 

 

Special Note on Visual Studio 2010

There may be differences when calculating code metrics using Visual Studio 2010 that don’t apply to Visual Studio 2008. The online documentation (http://msdn.microsoft.com/en-us/library/ee703787.aspx) gives the following reasons:

  • The function contains one or more catch blocks. In previous versions of Visual Studio, catch blocks were not included in the calculation. In Visual Studio 2010, the complexity of each catch block is added to the complexity of the function.
  • The function contains a switch (Select Case in VB) statement. Compiler differences between Visual Studio 2010 and earlier versions can generate different MSIL code for some switch statements that contain fall-through cases.

 

 

Putting It All Together

The bottom line is that a high complexity number means greater probability of errors with increased time to maintain and troubleshoot. Take a closer look at any functions that have a high complexity and decide if they should be refactored to make them less complex.

 

 

Citations

[MCCABE5]

McCabe, T. and A. Watson (1994), Software Complexity (CrossTalk: The Journal of Defense Software Engineering).

 

[NIST235]

Watson, A. H., & McCabe, T. J. (1996). Structured Testing: A Testing Methodology Using the Cyclomatic Complexity Metric (NIST Special Publication 500-235). Retrieved May 14, 2011, from McCabe Software web site: http://www.mccabe.com/pdf/mccabe-nist235r.pdf

 

[SATC]

Rosenberg, L., Hammer, T., Shaw, J. (1998). Software Metrics and Reliability (Proceedings of IEEE International Symposium on Software Reliability Engineering). Retrieved May 14, 2011, from Penn State University web site: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.104.4041&rep=rep1&type=pdf

  • Very helpful article to explain the CC.

  • Helpful.. Can you give a rough outline on how to implement the cylomatic complexity?

  • Hey Swapna :)

    Not sure what you are asking. This metric is used on written code to determine likelihood of errors. You should, of course, take this into account while you are writing your code but this is just coding practice. I would suggest reading Uncle Bob's book on Clean Code to help write good, clean methods:

    www.amazon.com/.../0132350882

    Z

  • Right. In case I want to build my own metric to calculate the same in visual studio, how should i proceed!?

  • Can you pls tell me the logic behind how to calculate cyclomatic complexity ?

  • Here is the actual way to do calculations:

    en.wikipedia.org/.../Cyclomatic_complexity

    Slightly easier to understand logic can be found here:

    www.ehow.com/how_7870069_calculate-cyclomatic-complexity.html

    The values may not match exactly between the two methods but they are close enough.

    Z

  • hey can you please tell me alternatives of to calculate the code complexity factor..??

    like c&k ...

  • Hello Amit :)

    I show how to do cyclomatic complexity above so not sure if you are asking for that or an overall number. I'm not a big fan of one number to represent the overall health of the code simply because I think it obscures essential pointers to clues for fixing the code. I would suggest you focus on the metrics that mean the most to your organization and use them. Of course, you should never, EVER use "lines of code" as a metric for anything other than showing you how many lines of code you have written. It is a completely useless metric otherwise.

    Z

Page 1 of 1 (8 items)
Leave a Comment
  • Please add 2 and 7 and type the answer here:
  • Post