Migrating from WordPress to some other site

parse-wordpress-xml.ps1

# Export your WordPress content before running this script

# Set the output path
$sitePath = "PATH_TO_SITE_OUTPUT"
$siteUrl = "YOUR_SITE_URL"

# Parse the XML
[xml]$xml = Get-Content .\YOUR_WORDPRESS_EXPORT.xml
# $xml.rss.channel.item | Get-Member
foreach ($item in $xml.rss.channel.item){
    # Only migrating posts and pages
    # Also not migrating the posts with query strings - need to look at those manually
    if (($item.post_type.InnerText -eq "page" -or $item.post_type.InnerText -eq "post") -and $item.link.IndexOf("?") -eq -1){
        $path = $item.link.Replace($siteUrl,"").Replace("/","\")
        if ($item.post_type.InnerText -eq "page"){
            $path = $sitePath + $path
        } else {
            $path = $sitePath + "\blog" + $path
        }
        # This will create the path with its subfolders
        if (!(Test-Path $path)){
            New-Item -ItemType Directory $path | Out-Null
        }
        $objectToConvert = New-Object -TypeName PSObject -Property @{
            PubDate = $item.post_date.InnerText 
            Link = $item.link
            Title = $item.title.InnerText
            Author = "sadukie"
            SlugName = $item.post_name.InnerText
            PostType = $item.post_type.InnerText
            Status = $item.status.InnerText
            Category = $item.category.InnerText
            Content = $item.encoded."#cdata-section"
        }
        
        # Check for Images
        $contentString = $objectToConvert.Content | Out-String
        $imagesToFetch = (Select-String -InputObject $contentString -Pattern "img src=""(https://[\w\d./-]*)""" -AllMatches).Matches
        $imgCount = $imagesToFetch.Count
        if ($imgCount -gt 0){
            Write-Host "Post has $imgCount images: " $objectToConvert.Title
            foreach ($imageToFetch in $imagesToFetch){
                # Get the captured value - just the URL
                $urlToFetch = New-Object System.Uri($imageToFetch.Groups[1].Value)
                # Fetch the image
                Write-Host "Download from: " $urlToFetch.AbsoluteUri
                # FileName
                $imgFileName = [System.IO.Path]::GetFileName($urlToFetch.LocalPath)
                $outputFilePath = $path + $imgFileName
                # Save it in this folder
                Write-Host "Saving to: " $outputFilePath
                # Uncommenting the line below will make the call to the web server to get the images
                # Invoke-WebRequest -Uri $urlToFetch -OutFile $outputFilePath
                # Update the content
                $relativeFileName = "./" + $imgFileName
                $contentString = $contentString.Replace($urlToFetch.AbsoluteUri,$relativeFileName)
            }
        }

        # Now let's create the file
        $filePath = $path + "\index.astro"
        if (!(Test-Path $filePath)){
            New-Item -ItemType File $filePath | Out-Null
            # Write the frontmatter
            $frontMatter = "===
pubDate: $($objectToConvert.PubDate)
title: $($objectToConvert.Title)
author: $($objectToConvert.Author)
category: $($objectToConvert.Category)
===
            "
            $frontMatter | Out-File -Append $filePath
            # Write the content - This is in HTML
            $contentString | Out-File -Append $filePath
        }
        
    }
}