Comments on: Magic Numbers http://thediscoblog.com/2006/02/24/magic-numbers/ Can you dig it man? Thu, 09 Sep 2010 10:42:36 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: The Disco Blog » Blog Archive » Code coverage coterie confab http://thediscoblog.com/2006/02/24/magic-numbers/comment-page-1/#comment-100905 The Disco Blog » Blog Archive » Code coverage coterie confab Mon, 01 Dec 2008 16:49:50 +0000 http://thediscoblog.com/?p=15#comment-100905 [...] therein lies the most telling metric that a code coverage report can covey — that which isn’t [...] [...] therein lies the most telling metric that a code coverage report can covey — that which isn’t [...]

]]> By: The Disco Blog » Improving code complexity http://thediscoblog.com/2006/02/24/magic-numbers/comment-page-1/#comment-27 The Disco Blog » Improving code complexity Sat, 11 Mar 2006 16:57:56 +0000 http://thediscoblog.com/?p=15#comment-27 [...] In fact, a number of studies in the early days of computing (before the Golden Age of Disco) did show a correlation between the number of paths through code and defects. One such metric that arose from these studies was Cyclomatic Complexity. This integer based metric precisely measures complexity by counting the distinct paths through a method; moreover, various studies over the years have determined that methods having a cyclomatic complexity (CC or sometimes referred to as CCN) value greater than 10 have a higher risk of defects. [...] [...] In fact, a number of studies in the early days of computing (before the Golden Age of Disco) did show a correlation between the number of paths through code and defects. One such metric that arose from these studies was Cyclomatic Complexity. This integer based metric precisely measures complexity by counting the distinct paths through a method; moreover, various studies over the years have determined that methods having a cyclomatic complexity (CC or sometimes referred to as CCN) value greater than 10 have a higher risk of defects. [...]

]]> By: The Disco Blog » Don’t focus on the elephant http://thediscoblog.com/2006/02/24/magic-numbers/comment-page-1/#comment-26 The Disco Blog » Don’t focus on the elephant Thu, 09 Mar 2006 21:14:20 +0000 http://thediscoblog.com/?p=15#comment-26 [...] Automating code inspections with analysis tools handles 80% of big picture and allows humans to intervene in the 20% that matters. For instance, Java’s PMD will run 180+ rules against a file every time it changes. If a particularly concerning rule is violated, such as a high cyclomatic complexity value, someone can take a look. Can you imagine trying to accomplish this process manually? Why would someone want to? That’s so establishment! [...] [...] Automating code inspections with analysis tools handles 80% of big picture and allows humans to intervene in the 20% that matters. For instance, Java’s PMD will run 180+ rules against a file every time it changes. If a particularly concerning rule is violated, such as a high cyclomatic complexity value, someone can take a look. Can you imagine trying to accomplish this process manually? Why would someone want to? That’s so establishment! [...]

]]> By: john miller http://thediscoblog.com/2006/02/24/magic-numbers/comment-page-1/#comment-23 john miller Fri, 24 Feb 2006 13:29:23 +0000 http://thediscoblog.com/?p=15#comment-23 I dig it! Some people use branch coverage over line coverage, but cyclomatic path coverage is even better than that because it tests the different decisions in the method independently of one another, which is what branch coverage misses. Imagine a method composed of two separate IF methods. You can achieve 100% branch coverage by running two tests: one that makes both decisions evaluate TRUE and one that makes them both evaluate FALSE. But you could be skipping over a bug that exists on the TRUE-FALSE path. I dig it! Some people use branch coverage over line coverage, but cyclomatic path coverage is even better than that because it tests the different decisions in the method independently of one another, which is what branch coverage misses. Imagine a method composed of two separate IF methods. You can achieve 100% branch coverage by running two tests: one that makes both decisions evaluate TRUE and one that makes them both evaluate FALSE. But you could be skipping over a bug that exists on the TRUE-FALSE path.

]]> By: Joe Ponczak http://thediscoblog.com/2006/02/24/magic-numbers/comment-page-1/#comment-22 Joe Ponczak Fri, 24 Feb 2006 13:06:17 +0000 http://thediscoblog.com/?p=15#comment-22 The interesting issue to me is that Cobetura reported 100% branch coverage based on 1 JUnit test, even though the 'if' statement clearly has two branches. Also, as the number of decisions increase (more than 3), the value of branch coverage decreases. 100% branch coverage, in many cases, provides a false sense of security. Basis-path coverage, derived by the number of executed cyclomatic paths, is a much better measurement because it looks at the relationships between decisions. The interesting issue to me is that Cobetura reported 100% branch coverage based on 1 JUnit test, even though the ‘if’ statement clearly has two branches.

Also, as the number of decisions increase (more than 3), the value of branch coverage decreases. 100% branch coverage, in many cases, provides a false sense of security. Basis-path coverage, derived by the number of executed cyclomatic paths, is a much better measurement because it looks at the relationships between decisions.

]]>