如何基于文件时长删除重复文件并保留最早原文件（支持指定路径与目录）

Great start using file hashes to detect duplicates—this is the most accurate method since it checks actual file content, not just filenames or sizes. Let's tweak your script to support a target path, traverse subdirectories recursively, and ensure only the earliest original file is kept for each set of duplicates.

Here's the revised, fully functional script:

param(
    [Parameter(Mandatory=$false)]
    [string]$TargetPath = (Get-Location).Path
)

# Get all files in target path and subdirectories, calculate their SHA256 hash
$files = Get-ChildItem -Path $TargetPath -File -Recurse | 
         Select-Object FullName, CreationTime, 
             @{Name='FileHash'; Expression={(Get-FileHash -Path $_.FullName -Algorithm SHA256).Hash}}

# Group files by their hash (duplicates share the same hash)
$duplicateGroups = $files | Group-Object -Property FileHash | Where-Object {$_.Count -gt 1}

# Process each group: keep the earliest file, delete the rest
foreach ($group in $duplicateGroups) {
    # Sort files by creation time (oldest first)
    $sortedFiles = $group.Group | Sort-Object -Property CreationTime

    # Keep the first (earliest) file, mark the rest for deletion
    $keepFile = $sortedFiles[0]
    $deleteFiles = $sortedFiles | Select-Object -Skip 1

    Write-Host "`nKeeping original file (created $($keepFile.CreationTime)): $($keepFile.FullName)"
    Write-Host "Marking $($deleteFiles.Count) duplicate(s) for deletion:"
    
    foreach ($file in $deleteFiles) {
        Write-Host "- $($file.FullName) (created $($file.CreationTime))"
        # Uncomment the line below to actually delete files (test first!)
        # Remove-Item -Path $file.FullName -Force -Confirm:$false
    }
}

Write-Host "`nDuplicate cleanup process completed."

Key Improvements & Explanations: #

• Specify a Target Path: The param block lets you pass a path like E:\media\ when running the script. If no path is provided, it defaults to your current working directory.
• Recursive Directory Traversal: The -Recurse flag in Get-ChildItem ensures we scan every subdirectory under the target path.
• Reliable Duplicate Detection: Using Get-FileHash with SHA256 guarantees that only files with identical content are flagged as duplicates (avoids false positives from same-size different-content files).
• Preserve the Earliest File: Each group of duplicates is sorted by CreationTime, so the oldest file is kept, and all newer copies are marked for deletion.

How to Use: #

1. Save the script as Remove-Duplicates.ps1
2. Open PowerShell, navigate to the script's location
3. Run it with your target path:

   .\Remove-Duplicates.ps1 -TargetPath "E:\media\"
   

4. Important: Test first! The deletion line is commented out—run the script once to verify which files will be removed, then uncomment the Remove-Item line when you're confident.

Safety Tips: #

• Always back up critical files before running duplicate deletion scripts.
• If you need to scan system-protected directories, run PowerShell as Administrator.
• You can modify the hash algorithm (e.g., MD5) if needed, but SHA256 is more secure and less prone to collisions.

内容的提问来源于stack exchange，提问作者Juan Lopez