Code Churn for Git Repositories

Code Churn for Git Repositories.ps1

param(
    [string]$RepoPath = ".",
    [string]$AuthorCsv = "churn_by_author.csv",
    [string]$FileCsv = "churn_by_file.csv",
    [string]$JsonOutput = "churn_dashboard.json",
    [string[]]$ExcludeFolders = @("bin/", "obj/", "node_modules/", "dist/", "packages/", "out/")
)

Set-Location $RepoPath

# Pull churn data from the entire repo history
$log = git log --numstat --pretty="%H|%an" --no-merges

$results = @()
$currentCommit = $null
$currentAuthor = $null

foreach ($line in $log) {

    # Commit + Author line
    if ($line -match "^([0-9a-f]{40})\|(.+)$") {
        $currentCommit = $matches[1]
        $currentAuthor = $matches[2].Trim()
        continue
    }

    # numstat line
    if ($line -match "^(\d+|-)\s+(\d+|-)\s+(.+)$") {
        $added = if ($matches[1] -eq "-") { 0 } else { [int]$matches[1] }
        $deleted = if ($matches[2] -eq "-") { 0 } else { [int]$matches[2] }
        $file = $matches[3]

        # Determine top-level folder
        $topFolder = if ($file -match "^(.*?)/") { $matches[1] + "/" } else { "<root>" }

        # Skip excluded folders
        if ($ExcludeFolders -contains $topFolder) {
            continue
        }

        $results += [pscustomobject]@{
            Commit  = $currentCommit
            Author  = $currentAuthor
            File    = $file
            Added   = $added
            Deleted = $deleted
        }
    }
}

# -------------------------------
# GROUP 1: Churn by Author
# -------------------------------
$byAuthor = $results |
    Group-Object Author |
    ForEach-Object {
        [pscustomobject]@{
            Author  = $_.Name
            Added   = ($_.Group | Measure-Object Added -Sum).Sum
            Deleted = ($_.Group | Measure-Object Deleted -Sum).Sum
            Net     = (($_.Group | Measure-Object Added -Sum).Sum - ($_.Group | Measure-Object Deleted -Sum).Sum)
        }
    }

$byAuthor | Export-Csv -Path $AuthorCsv -NoTypeInformation


# -------------------------------
# GROUP 2: Churn by File
# -------------------------------
$byFile = $results |
    Group-Object File |
    ForEach-Object {
        [pscustomobject]@{
            File    = $_.Name
            Added   = ($_.Group | Measure-Object Added -Sum).Sum
            Deleted = ($_.Group | Measure-Object Deleted -Sum).Sum
            Net     = (($_.Group | Measure-Object Added -Sum).Sum - ($_.Group | Measure-Object Deleted -Sum).Sum)
        }
    }

$byFile | Export-Csv -Path $FileCsv -NoTypeInformation


# -------------------------------
# JSON OUTPUT FOR DASHBOARDS
# -------------------------------
$dashboard = [pscustomobject]@{
    GeneratedOn = (Get-Date)
    ExcludedFolders = $ExcludeFolders
    ChurnByAuthor = $byAuthor
    ChurnByFile   = $byFile
}

$dashboard | ConvertTo-Json -Depth 10 | Out-File $JsonOutput -Encoding utf8

Write-Host "`nExports complete:"
Write-Host " - $AuthorCsv"
Write-Host " - $FileCsv"
Write-Host " - $JsonOutput"