graphviz-git-dvc.dot

/*
This Graphviz file represents a comparison between Git and DVC (Data Version Control) commands. 
Each node in the graph represents a command, and the edges represent the flow of data or control. 
The graph is divided into two clusters: one for Git and one for DVC. 

This graph is created with assistance from Claude Sonnet, an AI assistant by Anthropic.
*/

digraph DVC_vs_Git {
    node [shape=box, fontname="Arial"]
    rankdir=LR  // Arrange nodes from left to right

    // Title
    labelloc="t"
    label="Comparison of DVC and Git Commands\n(Created with assistance from Claude, an AI assistant by Anthropic)"

    // Key
    key [shape=none, label=<
        <table border="0" cellborder="1" cellspacing="0" cellpadding="4">
            <tr><td bgcolor="lightblue">Git</td><td bgcolor="lightgray">DVC</td></tr>
            <tr><td>git init: Initialize a new Git repository</td><td>dvc init: Initialize a new DVC repository</td></tr>
            <tr><td>git add: Stage changes for the next commit</td><td>dvc add: Track data files or models with DVC</td></tr>
            <tr><td>git commit: Create a new commit with staged changes</td><td>dvc commit: Create a new commit with changes to DVC files</td></tr>
            <tr><td>git push: Upload local commits to a remote repository</td><td>dvc push: Upload data files or models to a remote storage</td></tr>
            <tr><td>git pull: Download commits from a remote repository</td><td>dvc pull: Download data files or models from a remote storage</td></tr>
            <tr><td>git checkout: Switch branches or restore files</td><td>dvc checkout: Restore data files or models from a specific commit</td></tr>
            <tr><td></td><td>dvc repro: Reproduce the data pipeline from the DVC files</td></tr>
            <tr><td></td><td>dvc exp run: Run a command and capture its dependencies</td></tr>
        </table>
    >]

    // Center the key
    { rank=same; key; }

    subgraph cluster_git {
        label = "Git"
        labeljust = l
        bgcolor = "#ADD8E6" # lightblue
        style = filled

        git_init [label="git init"]
        git_add [label="git add"]
        git_commit [label="git commit"]
        git_push [label="git push"]
        git_pull [label="git pull"]
        git_checkout [label="git checkout"]

        // Arrange nodes horizontally
        { rank=same; git_init; git_add; git_commit; git_push; }
        { rank=same; git_pull; git_checkout; }

        git_init -> git_add -> git_commit -> git_push
        git_pull -> git_checkout
    }

    subgraph cluster_dvc {
        label = "DVC"
        labeljust = l
        bgcolor = "lightgray"
        style = filled

        dvc_init [label="dvc init"]
        dvc_add [label="dvc add"]
        dvc_commit [label="dvc commit"]
        dvc_push [label="dvc push"]
        dvc_pull [label="dvc pull"]
        dvc_checkout [label="dvc checkout"]
        dvc_repro [label="dvc repro"]
        dvc_exp_run [label="dvc exp run"]

        // Arrange nodes horizontally
        { rank=same; dvc_init; dvc_add; dvc_commit; dvc_push; dvc_repro; dvc_exp_run; }
        { rank=same; dvc_pull; dvc_checkout; }

        dvc_init -> dvc_add -> dvc_commit -> dvc_push
        dvc_pull -> dvc_checkout
        dvc_commit -> dvc_repro [style=dashed]
        dvc_exp_run -> dvc_add [style=dashed]
    }

    // Colored edges for integration points
    git_commit -> dvc_commit [style=dashed, color=red]
    git_add -> dvc_add [style=dashed, color=red]
    git_init -> dvc_init [style=dashed, color=red]
    git_push -> dvc_push [style=dashed, color=red]
}