Prepare your major dependency updates with graph-guided refactoring analysis

ChangingDot uses static analysis and LLMs to create a graph of all changes that result from a intial seed change.

This initial change can be anything, but we mostly focus our technology on refactorings and more specifically major dependency upgrades.

Using the graph, we can :

• Gain crucial insight the impact of the upgrade
• Have an idea of the nature of the resulting changes
• Rapidly uncover hard to find couplings that drag the project on and on
• A initial upgrade attempt

This information can be used before any major change to :

• Estimate the effort and time required
• Figure out the best strategy to minimize the effort of the update (use codemodes, big bang, divide and conquer...)
• Prepare for the breaking changes that are induced
• Never be surprised by unexpected coupling

Test us out using our CLI

pip install changing-dot

Installation

At the moment we only support C#, python support is planned for the near future.

You need to have .NET8.0 runtime installed on your machine. You can install it here

Creating your first graph

To test out the tech, you can first download our simple test repo :

git clone git@github.com:ChangingDot/ChangingLevels.git

Our CLI takes as input a path to a configuration file. In this readme we use a very simple change with config file stored in a CDN. If you want to try out the technology on one of your own repo, check out the documentation.

Make sure you are in the ChangingLevels repo and run :

cdot create -c https://rawcdn.githack.com/ChangingDot/ChangingDot/8f88a91b7954613cef2e5b5fcc4e1a647c81cd28/examples/example.yaml

Visualizing your graph

You can then visualize the created graph by running this command :

cdot visualize -c https://rawcdn.githack.com/ChangingDot/ChangingDot/8f88a91b7954613cef2e5b5fcc4e1a647c81cd28/examples/example.yaml

Running analyses on the graph

This graph can be used as a base for different analyses and even code changes. It gives crucial inforamtion on the impact of a seed change. We don't currently support any analyses but plan to implement the following :

• Breaking change analysis
• Spread analysis
• Complexity/ type of changes

If you have any ideas for other analyses or uses for the graph, to not hesitate to open an issue, or send us a message.

Committing the changes

A longer term goal of ChangingDot is to automate actual refactorings and therefore actually change the code. You can actually test this out using the following command.

cdot commit -c https://rawcdn.githack.com/ChangingDot/ChangingDot/8f88a91b7954613cef2e5b5fcc4e1a647c81cd28/examples/example.yaml

If you do not want to automatically commit the changes, you can apply them to your local git repo to review and changes them later on

cdot apply-changes -c https://rawcdn.githack.com/ChangingDot/ChangingDot/8f88a91b7954613cef2e5b5fcc4e1a647c81cd28/examples/example.yaml