@VincenzoP1948 - actually - if you still want to look at the possibility of rendering a PDF to an Image; there is an ('pdfBOX') API to do this;  but as it is done on per-page basis - you'll need to work out some details - about how to 'stitch' together multiple pages together into single image.

On PRPC722: the following Java Step will render the first page (page zero) to a PNG file:

byte[] byteArray=(byte[])tools.getParameterPage().getParameterValue("PDFDocument");
String imageName=tools.getParamValue("ImageName");
java.io.ByteArrayInputStream bis=null;  
bis = new java.io.ByteArrayInputStream(byteArray);

try {  
    org.apache.pdfbox.pdmodel.PDDocument doc=org.apache.pdfbox.pdmodel.PDDocument.load( bis );
    org.apache.pdfbox.rendering.PDFRenderer pdfRenderer = new org.apache.pdfbox.rendering.PDFRenderer(doc);
    java.awt.image.BufferedImage bim = pdfRenderer.renderImageWithDPI(0, 300 , org.apache.pdfbox.rendering.ImageType.RGB); //page 0.
 
    java.io.ByteArrayOutputStream baos = new java.io.ByteArrayOutputStream();
    javax.imageio.ImageIO.write( bim, "png", baos );
    baos.flush();
    String result=tools.sendFile(baos.toByteArray(), imageName, false, null, true);
}  
catch(Exception e) { throw new PRRuntimeException(e);}  
finally {  
  if (bis!=null) {  
    try { bis.close(); }  
    catch(Exception e) { throw new PRRuntimeException(e); }  
}  
}

This assume two parameters: 'ImageName' (just a String like 'test.png' for instance) and a Java Object - this has to be an ByteArray holding the PDF content.

You test can with running a HTMLTOPDF operation first: then using the 'Pass Current Parameter Page' option when you can the 'wrapper' Activity above.

BTW: If you are not using PRPC722 - you will need to check which version of 'PDBox' is shipped with PRPC.

It appears the library was refactored slightly (it used to be 're-packaged' under a pega package; now it is using the same package name as the original Apache Library).

You can check your system using the following SQL:

select distinct(pzjar) from pr_engineclasses where upper(pzjar) like '%PDF%'; 

To get the version of the JAR; then you can use something like this to get the package name used:

select distinct(pzpackage) from pr_engineclasses where pzjar='pdfbox-2.0.1.jar'
order by pzpackage;

BTW: the actual implementation of the 'pdfBox' library has also changed between versions; see this StackOverflow post for the difference (I'm using code based on the second version 2.x example).

http://stackoverflow.com/questions/23326562/apache-pdfbox-convert-pdf-to-images