Code review automates the process of checking code for bugs and adherence to best practices. It is quick and easy to run, allowing developers and architects to find problems early when they are easy to fix.
For each issue it identifies, code review explains the problem...
...then shows an example of problem code...
...and then offers a recommendation. In some cases, code review can automatically fix the problem.
Several code reviews are supplied, each containing sets of rules. You can choose to run particular code reviews at any stage, or run all of them.
You also can create rules. New rules can be architectural controls or general rules.
There are three severity levels: Problem, Warning, and Recommendation.
Problem is the highest level, and represents problems that must be fixed.
Warning means a potential problem that should be addressed.
Recommendation is the lowest level, but these problems should not be overlooked. Addressing them likely will prevent larger problems in the future.
If a quick fix is available, it is indicated by one of these icons.
This demonstration shows a developer importing files into a new Java project, running a code review, and applying a quick fix.
This concludes the tour. To learn more, refer to the Tutorial Gallery for a tutorial on running a code review, or refer to the Information Center.