zcov is wrapper around the basic facilities of gcov for generating pretty summaries for entire code bases. It is similar to lcov with an emphasis on nicer HTML output (including, for example, branch coverage), and a greatly simplified command line interface which tries to work around deficiencies in the metadata provided by gcov.

zcov uses '.zcov' files to store the information about a run, and provides several tools for working with these files. Use --help to see more information on the tools.

Recursively scan a directory structure looking for .gcda files and create a .zcov containing all the results. Usage: $ zcov scan output.zcov ~/public/coreutils-6.10.cov

Create an HTML report from a .zcov file. Usage: $ zcov genhtml output.zcov coverage-report/

zcov genhtml will make the output directory and write HTML files and an index.html for all the files in the .zcov. One useful option is '--root' which tells zcov to ignore any files outside the given root directory, and not to write that root directory in the output (for example, so your home directory path doesn't show up in the report).

Print a short textual summary of a .zcov file, for testing and simple command line uses. Usage: $ zcov summarize output.zcov

Merge multiple .zcov files into one. Usage: $ zcov merge output.zcov input1.zcov input2.zcov ... inputN.zcov

zcov is available under the BSD license. See LICENSE.txt.

zcov has the following dependencies:

• Python
• PIL
• enscript

zcov is written and maintained by Daniel Dunbar daniel@zuster.org

The JavaScript sorttable implementation as written by Stuart Langridge and is included under the X11 license.