Files
igny8/igny8-ai-seo-wp-plugin/extract_docx.ps1
2025-11-11 21:16:37 +05:00

25 lines
679 B
PowerShell

Add-Type -AssemblyName System.IO.Compression.FileSystem
$docxPath = "e:\GitHub\igny8-ai-seo\docs\Igny8 WP Plugin to Igny8 App Migration Plan.docx"
$zip = [System.IO.Compression.ZipFile]::OpenRead($docxPath)
$entry = $zip.Entries | Where-Object { $_.FullName -eq "word/document.xml" }
if ($entry) {
$stream = $entry.Open()
$reader = New-Object System.IO.StreamReader($stream)
$xml = $reader.ReadToEnd()
$reader.Close()
$stream.Close()
# Extract text from XML (simple approach)
$xml = $xml -replace '<[^>]+>', "`n"
$xml = $xml -replace '\s+', " "
$xml = $xml -replace '\n\s*\n', "`n"
Write-Output $xml.Trim()
}
$zip.Dispose()