eruffaldi · October 22, 2025 11:36
diff --git a/docx2pdf_ex.ps1 b/docx2pdf_ex.ps1
 # DOCX to PDF after Markdown generation
 # Copyright Emanuele Ruffaldi MMI 2024
 # License: MIT
 #
 # Usage:
 #      script INPUT.DOCX
 #
 # Output: INPUT.PDF
 #
 # Requirements
 # - remove existing PDF
 # - FAIL if DOCX missing
 # - FAIL if PDF cannot be generated
 # - FAIL if there are missing references
 # - launches Word unattended, hidden and timeout of 60s
 # - regenrates fields and toc
 # - export PDF without comments and changes
 # - steps: open, update fields and toc, check references, save, export pdf

 param (
    [string]$in
 )

 # Ensure the input file path is absolute
 $inputFilePath=$in
 $inputFilePath = [System.IO.Path]::GetFullPath($inputFilePath)

 if (-not (Test-Path $inputFilePath)) {
    Write-Error "The specified input file does not exist: $inputFilePath"
    exit 1
 }

 # Generate the output file path by changing the extension to .pdf
 $outputFilePath = [System.IO.Path]::ChangeExtension($inputFilePath, ".pdf")

 # Remove if exists
 if ((Test-Path $outputFilePath)) {
    Remove-Item $outputFilePath
 }

 # Timeout
 $timeoutSeconds = 60

 $ProcessWordDocument = {
    param ($inputFilePath, $outputFilePath)

    Write-Output "ProcessWordDocument"

    # Create a new Word application object
    $word = New-Object -ComObject Word.Application
    $word.Visible = $false  # Keep Word application hidden
    $word.DisplayAlerts = [Microsoft.Office.Interop.Word.WdAlertLevel]::wdAlertsNone  # Disable alerts

    try {
        # Open the document
        $document = $word.Documents.Open($inputFilePath)

        $document.Fields.Update()

        # Update table of contents specifically
        foreach ($toc in $document.TablesOfContents) {
            $toc.Update()
        }

        # Save the updated Word document
        $document.Save() 

        # Check for any errors in fields (missing references)
        $missingReferences = $false
        foreach ($field in $document.Fields) {
            if ($field.Result.Text -match 'Error! Reference source not found') {
                $missingReferences = $true
                break
            }
        }

        if ($missingReferences) {
            throw "The document contains missing references."
        }

        # Hide comments
        $word.ActiveWindow.View.ShowRevisionsAndComments = $false
        $word.ActiveWindow.View.RevisionsFilter.Markup = [Microsoft.Office.Interop.Word.WdRevisionsMarkup]::wdRevisionsMarkupNone
        $word.ActiveWindow.View.RevisionsFilter.View = [Microsoft.Office.Interop.Word.WdRevisionsView]::wdRevisionsViewFinal

        # Export the document to PDF (17 is the WdSaveFormat enumeration for PDF)
        $document.SaveAs([ref] $outputFilePath, 17)
        
        # Close the document
        $document.Close($false)
    } catch {
        Write-Error "An error occurred: $_"
    } finally {
        # Quit the Word application
        $word.Quit()

        # Release the COM objects
        [System.Runtime.Interopservices.Marshal]::ReleaseComObject($document) | Out-Null
        [System.Runtime.Interopservices.Marshal]::ReleaseComObject($word) | Out-Null

        # Clean up
        Remove-Variable -Name document
        Remove-Variable -Name word
    }
 }

 # Enable this for debugging without launching the Job   
 $testMode = $false

 if($testMode)
 {
    & $ProcessWordDocument $inputFilePath $outputFilePath
 }
 else
 {
    # Start the Word processing job
    $job = Start-Job -ScriptBlock $ProcessWordDocument -ArgumentList $inputFilePath, $outputFilePath

    # Monitor the job with timeout
    $job | Wait-Job -Timeout $timeoutSeconds
    Receive-Job $job -ErrorAction Stop

    if ($job.State -eq 'Completed') {
    }
    elseif ($job.State -eq 'Failed') {
        Write-Output "Document processing Failed: " + $job.ChildJobs[0].Error + ($job.ChildJobs[0].JobStateInfo.Reason.Message)
    } else {
        Write-Output "Document processing timed out. Terminating job: " + $job.State
        Stop-Job -Job $job
    }
 }

 # Clean up any lingering Word processes
 # Get-Process winword -ErrorAction SilentlyContinue | ForEach-Object { $_.Kill() }

 if ((Test-Path $outputFilePath)) {
    exit 0
 }
 else
 {
    exit 1
 }
	# DOCX to PDF after Markdown generation
	# Copyright Emanuele Ruffaldi MMI 2024
	# License: MIT
	#
	# Usage:
	# script INPUT.DOCX
	#
	# Output: INPUT.PDF
	#
	# Requirements
	# - remove existing PDF
	# - FAIL if DOCX missing
	# - FAIL if PDF cannot be generated
	# - FAIL if there are missing references
	# - launches Word unattended, hidden and timeout of 60s
	# - regenrates fields and toc
	# - export PDF without comments and changes
	# - steps: open, update fields and toc, check references, save, export pdf

	param (
	[string]$in
	)

	# Ensure the input file path is absolute
	$inputFilePath=$in
	$inputFilePath = [System.IO.Path]::GetFullPath($inputFilePath)

	if (-not (Test-Path $inputFilePath)) {
	Write-Error "The specified input file does not exist: $inputFilePath"
	exit 1
	}

	# Generate the output file path by changing the extension to .pdf
	$outputFilePath = [System.IO.Path]::ChangeExtension($inputFilePath, ".pdf")

	# Remove if exists
	if ((Test-Path $outputFilePath)) {
	Remove-Item $outputFilePath
	}

	# Timeout
	$timeoutSeconds = 60

	$ProcessWordDocument = {
	param ($inputFilePath, $outputFilePath)

	Write-Output "ProcessWordDocument"

	# Create a new Word application object
	$word = New-Object -ComObject Word.Application
	$word.Visible = $false # Keep Word application hidden
	$word.DisplayAlerts = [Microsoft.Office.Interop.Word.WdAlertLevel]::wdAlertsNone # Disable alerts

	try {
	# Open the document
	$document = $word.Documents.Open($inputFilePath)

	$document.Fields.Update()

	# Update table of contents specifically
	foreach ($toc in $document.TablesOfContents) {
	$toc.Update()
	}

	# Save the updated Word document
	$document.Save()

	# Check for any errors in fields (missing references)
	$missingReferences = $false
	foreach ($field in $document.Fields) {
	if ($field.Result.Text -match 'Error! Reference source not found') {
	$missingReferences = $true
	break
	}
	}

	if ($missingReferences) {
	throw "The document contains missing references."
	}

	# Hide comments
	$word.ActiveWindow.View.ShowRevisionsAndComments = $false
	$word.ActiveWindow.View.RevisionsFilter.Markup = [Microsoft.Office.Interop.Word.WdRevisionsMarkup]::wdRevisionsMarkupNone
	$word.ActiveWindow.View.RevisionsFilter.View = [Microsoft.Office.Interop.Word.WdRevisionsView]::wdRevisionsViewFinal

	# Export the document to PDF (17 is the WdSaveFormat enumeration for PDF)
	$document.SaveAs([ref] $outputFilePath, 17)

	# Close the document
	$document.Close($false)
	} catch {
	Write-Error "An error occurred: $_"
	} finally {
	# Quit the Word application
	$word.Quit()

	# Release the COM objects
	[System.Runtime.Interopservices.Marshal]::ReleaseComObject($document) \| Out-Null
	[System.Runtime.Interopservices.Marshal]::ReleaseComObject($word) \| Out-Null

	# Clean up
	Remove-Variable -Name document
	Remove-Variable -Name word
	}
	}

	# Enable this for debugging without launching the Job
	$testMode = $false

	if($testMode)
	{
	& $ProcessWordDocument $inputFilePath $outputFilePath
	}
	else
	{
	# Start the Word processing job
	$job = Start-Job -ScriptBlock $ProcessWordDocument -ArgumentList $inputFilePath, $outputFilePath

	# Monitor the job with timeout
	$job \| Wait-Job -Timeout $timeoutSeconds
	Receive-Job $job -ErrorAction Stop

	if ($job.State -eq 'Completed') {
	}
	elseif ($job.State -eq 'Failed') {
	Write-Output "Document processing Failed: " + $job.ChildJobs[0].Error + ($job.ChildJobs[0].JobStateInfo.Reason.Message)
	} else {
	Write-Output "Document processing timed out. Terminating job: " + $job.State
	Stop-Job -Job $job
	}
	}

	# Clean up any lingering Word processes
	# Get-Process winword -ErrorAction SilentlyContinue \| ForEach-Object { $_.Kill() }

	if ((Test-Path $outputFilePath)) {
	exit 0
	}
	else
	{
	exit 1
	}