This documentation is a preview of the pull request #2086

For the latest updates and improvements in production, open docs.codacy.com instead.

Which metrics does Codacy calculate?#

Codacy performs static code analysis and calculates code duplication, code complexity, and code coverage metrics for most supported programming languages.

The following sections describe how Codacy calculates each supported metric and where you can see each metric on the Codacy UI:

Grade
Issues
Complexity
Duplication
Code coverage

Note

Depending on certain characteristics of your repository, such as the number of source code files and their size, Codacy may apply limits to the code analysis that impact the calculation of the supported metrics.

Grade#

Codacy assigns an overall grade to your repository branches and to individual files to help you assess the code quality of your repository. Grades represent a weighted average of the available code quality metrics (issues, complexity, duplication, and coverage), and range from A to F:

Highest grade

Lowest grade

Codacy displays grades on the following places:

Place	Metric
Files page	Grade for each file in your repository
Repository Dashboard Codacy badge	Grade of each analyzed branch in your repository
Email notifications	Grade of your repository
Organization overview	Average grade of the repositories in your organization and grade of each repository
Repositories list	Grade of each repository in your organization

Issues#

Codacy calculates the number of issues in the following static code analysis categories:

Code style: Code formatting and syntax problems, such as variable names style and enforcing the use of brackets and quotation marks
Error prone: Code that may hide bugs and language keywords that should be used with caution, such as the operator == in JavaScript or Option.get in Scala
Code complexity: High complexity methods and classes that should be refactored
Performance: Code that can have performance problems
Compatibility: Mainly for frontend code, compatibility problems across different browser versions
Unused code: Unused variables and methods, code that can't be reached
Security: Potential security vulnerabilities, including hard-coded passwords and keys (secret scanning), vulnerable dependencies (software composition analysis or SCA), and insecure code patterns (static application security testing or SAST). For more information, see the complete list of security issue categories
Documentation: Methods and classes that don't have the correct comment annotations
Best practice: Code that doesn't follow the recommended coding standards and best practices
Comprehensibility: Code that can be difficult to understand and modify

Besides this, Codacy also allows you to compare issues across repositories with different sizes by calculating the issue cost relative to a baseline of 1 point per line of code, where the cost of each issue depends on its severity: Critical = 10 points, Medium = 5 points, Minor = 1 point. This means that if your repository has 50% issues, the amount and severity of the issues in your repository is half of the baseline.

Codacy displays issues on the following places:

Place	Metric
Commit detail page Pull request detail page Email notifications	Number of new and fixed issues introduced by the commit or pull request
Files page	Number of issues in each file
Issues page	List of all issues detected in each branch
Repository Dashboard	Issue percentage and how the metric is evolving over time
Organization overview	Average issue percentage of the repositories in your organization and issue percentage of each repository
Repositories list page	Issue percentage in each repository in your organization

Complexity#

Codacy uses cyclomatic complexity to identify files with complex methods in your repository. Cyclomatic complexity is the number of linearly independent paths through the source code of a method: the more control flow statements used in a method, the higher the value. Methods with a high cyclomatic complexity are more difficult to test and more likely to have defects. Learn more about code complexity on Codacy's blog.

Codacy calculates complexity as follows:

The complexity value for each file is the highest cyclomatic complexity of the methods in the file.
A file is considered complex if its cyclomatic complexity value is higher than the threshold File is complex when over.
The complexity value of a commit or pull request is the sum of the cyclomatic complexity of the files that were changed in the commit or pull request and that have a high complexity increase, defined as an increase of 4 or more.

Codacy displays complexity on the following places:

Place	Metric
Commit detail page Pull request detail page Email notifications	Variation of the complexity value introduced by the commit or pull request
Files page	Complexity value of each file
Repository Dashboard	Percentage of complex files in your repository and how the metric is evolving over time
Organization overview	Average percentage of complex files in the repositories in your organization and percentage of complex files in each repository
Repositories list page	Percentage of complex files in each repository in your organization

Duplication#

Codacy identifies clones or sequences of duplicate code that exist in at least two different places of the source code of your repository. Clones typically indicate deeper code quality issues and should be eliminated through abstraction when possible.

Codacy calculates duplication as follows:

The duplication value for each file is the number of clones in the file.
A file is considered duplicated if the number of clones in the file is higher than the threshold File is duplicated when over.
The duplication value of a commit or pull request is the number of clones introduced by the commit or pull request.

Note

You can customize the rules for identifying duplicated blocks of code when using PMD CPD to analyze the source code of your repository.

Codacy displays duplication on the following places:

Place	Metric
Commit detail page Pull request detail page Email notifications	Number of clones added or fixed by a commit or pull request
Files page	Duplication value of each file
Repository Dashboard	Percentage of duplicated files in your repository and how the metric is evolving over time
Organization overview	Average percentage of duplicated files in the repositories in your organization and percentage of complex files in each repository
Repositories list page	Percentage of duplicated files in each repository in your organization

Code coverage#

Code coverage describes the degree to which the source code of a program is tested. There are several types of coverage, but Codacy uses line coverage, which measures the percentage of coverable lines of code that are covered by automated tests. Learn more about code coverage on Codacy's blog.

You must set up your CI/CD pipeline to upload code coverage data to Codacy. Because of this, the tool that you use to generate the coverage reports is responsible for creating the data that Codacy then uses to calculate code coverage.

Codacy calculates code coverage as follows:

The coverage value for each file is the percentage of coverable lines that are covered by tests in the file. If a line is covered multiple times, Codacy counts it as a single covered line when calculating coverage.
A repository is considered to have acceptable coverage if the percentage of coverable lines that are covered by tests in the repository is higher than the threshold Coverage is under.

The coverage variation of a commit or pull request is the increase or drop in the percentage of coverable lines that are covered by tests in the repository because of the changes of the commit or pull request.
The diff coverage of a pull request is the percentage of coverable lines that the pull request added or modified that are covered by tests.

If a pull request doesn't add or modify any coverable lines, the diff coverage is ∅ (not applicable). This scenario happens when the only changes in a pull request are:
- Deleted lines
- Added or modified lines that aren't coverable

Note

If you encounter a situation where Codacy shows an unexpected drop in coverage, learn about the most common reasons causing those scenarios.

Once the coverage setup is complete, Codacy displays coverage data on the following places:

Place	Metric
Commit detail page Pull request detail page Email notifications	Variation in percentage points of the coverage value for all files in the commit or pull request
Pull request detail page	Diff coverage for the changes included in the pull request
Files page	Coverage percentage of each file
Repository Dashboard	Coverage of the most recent commit of the selected branch and its evolution over time
Codacy badge	Coverage of the most recent commit of the configured branch
Organization overview	Average coverage of the repositories in your organization and coverage of each repository
Repositories list page	Coverage of each repository in your organization

Which metrics does Codacy calculate?#

Grade#

Issues#

Complexity#

Duplication#

Code coverage#

See also#

Share your feedback 📢