Visualize dependencies between Python modules

Summary

Pydeps is a tool for visualizing dependencies between Python modules, aiding in the navigation and understanding of complex codebases.

Abstract

Pydeps is a Python package that simplifies the process of understanding the intricate relationships between modules in large Python projects. By generating dependency graphs, it helps developers navigate through distributed and modular code, such as those found in repositories like Detectron2 and TensorFlow Object Detection API. The tool is easily installed via pip and requires Graphviz for graph generation. Pydeps offers a range of command-line arguments to customize the visualization of dependencies, including the ability to show import cycles and limit the depth of dependency visualization. This can be particularly useful for managing complex repositories where understanding the interdependencies is crucial for efficient code maintenance and development.

Opinions

The author finds Pydeps to be an essential tool for dealing with the complexity of understanding dependencies in large Python projects.
The difficulty in navigating through files to find function definitions in modular codebases is acknowledged as a common challenge.
Pydeps is praised for its ability to create clear visualizations of module dependencies, which significantly improves the user's ability to comprehend the structure of a codebase.
The tool's utility is highlighted by its ability to filter out distant dependencies with the --max-bacon argument, enhancing the clarity of the dependency graphs for complex repositories.
The author suggests that Pydeps is not only useful for its command-line interface but can also be integrated into Python as a library for programmatic use.
The article concludes with a recommendation for Pydeps, emphasizing its cost-effectiveness compared to other AI services.

Better understand your code modules using Pydeps

Have you ever gone through repositories like Detectron or TF Object Detection API? How easy it is for you to navigate or understand the code? Isn’t it better if you have a tool that helps better understand the code?

The major hurdle we face in understanding these repositories is knowing dependencies or linkages. Each file in these repos is linked to many other files through import. The code is so much distributed and modular that it is confusing sometimes to navigate to the actual definition of a function.

Recently I have faced a similar issue and had a hard time navigating through files for getting function definitions.

So I was looking for a way to get a graph of all dependencies between python scripts. Then I found this tool called “Pydeps”. After I generate a graph of all dependencies, I was able to better navigate and understand the code. If you are having a hard time understanding any complex python code, use Pydeps for visualizing the module dependencies.

Usage

Usage is very simple. Just run pydeps <filename>. It will create a graph of all dependencies for this file. They also have many optional arguments in the command line.

usage: pydeps [-h] [--debug] [--config FILE] [--no-config] [--version] [-L LOG] [-v] [-o file] [-T FORMAT] [--display PROGRAM] [--noshow] [--show-deps] [--show-raw-deps] [--show-dot] [--nodot] [--no-output] [--show-cycles] [--debug-mf INT] [--noise-level INT] [--max-bacon INT] [--pylib] [--pylib-all] [--include-missing] [-x PATTERN [PATTERN ...]] [-xx MODULE [MODULE ...]] [--only MODULE_PATH [MODULE_PATH ...]] [--externals] [--reverse] [--cluster] [--min-cluster-size INT] [--max-cluster-size INT] [--keep-target-cluster] [--rmprefix PREFIX [PREFIX ...]] fname

The description of all these arguments is given in their Github. Feel free to look at it for understanding all use cases. We will see examples of the most useful arguments here.

--max-bacon:

--max-bacon is used to specify the level to which the dependencies must be shown. It lets you filter out modules that are more than a given number of hops away. the default value for this is 2. Below is the example output for — max-bacon as 2 and -max-bacon as 0— which means infinite.

If you have a complex repository, — max-bacon will be useful to cut down the dependencies for better visualization.

--show-cycles:

Another important argument that is useful is “ — show-cycles”. If a folder has inter-linked imports i.e., two python scripts importing from one another, It forms a cycle. This argument displays those cycles. See the below example.

relimp:
    - __init__.py
    - a.py: |
        from . import b
    - b.py: |
        from . import a

These are some of the useful arguments in Pydeps. You can of course also import pydeps from Python and use it as a library. This tool is very helpful for visualizing the dependencies of your python code. Find out more about other arguments and usage on their official Github page. Thank you!

Programming

Visualize dependencies between Python Modules

Better understand your code modules using Pydeps

Installation

Usage

Examples

Without any optional arguments:

--include-missing:

--max-bacon:

--show-cycles:

References