leonoyd/codeviz

Create a visual code dependency graph for C/C++ projects

CodeViz

Visualize code dependencies between files in your C/C++ projects.

CodeViz is a simple, cross-platform, python3 script that creates a code dependency graph using source and header files.

It works by first creating a dot file. The dot file is then then passed into graphviz, which is a supporting program that will automagically create a map of your code.

Installation

You will need to have a python3 interpreter installed to run the script.

The script itself requires graphviz installed. Follow the install instructions from the graphviz website for your platform:

• Arch Linux (AUR)
• Fedora
• FreeBSD
• Mac
• Ubuntu/Debian
• Windows
• Solaris

Once installed, make sure that you have these tools in your environment path.

To install codeviz into your system:

$ git clone https://github.com/jmarkowski/codeviz.git
$ cd codeviz
$ python3 setup.py install

Example

The image below shows the result of generating the visual code map for the original git source at hash e83c516.

Usage

Run codeviz inside of your source directory. By default, it will only use the source/header files in the working directory (system header files such as stdio.h are ignored).

$ codeviz

You may also search for source/header files recursively as so

$ codeviz -r

If you want a black and white version of the generated dependency graph

$ codeviz -n

What if you want a black and white version of the generated dependency graph, AND you would like to ignore files that are not including any other header in your source files?

$ codeviz -m

You can specify the name of an output file. The extension will be passed into graphviz to generate the appropriate file.

$ codeviz -o jpeg-file.jpg
$ codeviz -o postscript-file.ps
$ codeviz -o png-file.png

See here for a complete list of supported output formats

Finally, can you exclude certain files and directories from being used.

$ codeviz -r --exclude=unit-tests/* --exclude=test *.{c,cpp,h}   # exclude tests