git_find_big.sh の改良版 (PowerShell 7)

git_find_big.ps1

#!/usr/bin/env -S pwsh -nop

$blobs = Start-ThreadJob {
	gci "$(git rev-parse --git-common-dir)/objects/pack/pack-*.idx" | % -Parallel {
		git verify-pack -v $_ | % {
			if ($_ -match '^([\da-f]+) blob +(\d+) (\d+)') {
				$blob = @{
					hash = $Matches[1]
					file_size = [long]$Matches[2]
					pack_size = [long]$Matches[3]
				}

				# 100KB以上
				if ($blob.file_size -ge (100 * 1024)) {
					Write-Output $blob
				}
			}
		}
	} | sort -Property file_size, pack_size -Descending
}

$files = @{}
git rev-list --all --objects | % {
	if ($_ -match '^([\da-f]+) (.+)$') {
		$files.Add($Matches[1], $Matches[2])
	}
}

foreach ($blob in Receive-Job -Wait -AutoRemoveJob $blobs) {
	$name = $files[$blob.hash]
	if ($null -ne $name) {
		Write-Host ("blob: {0}, size: {1:0,,.00}[MB], pack: {2:0,}[KB], {3}" -f $blob.hash, $blob.file_size, $blob.pack_size, $name)
	}
}