Comment 11 for bug 1066932

Don, your captured print job data is a PostScript file without the images, so the "Getting the data which would go to the printer" step is not needed any more as the images already get lost by the desktop application. In this case it is the Java engine. To print you have to click a button supplied by the Java application, not the usual "File"/"Print" of the browser, which means that the Java app generates the output stream (the stream would also be PDF if it comes from the browser). You should try to find a way to download or export the document and after that print the downloaded/exported file.

For me it looks like a problem of the Java engine. Are you sure that it did not change from 12.04 to 12.10?