0 Replies Latest reply on May 10, 2011 5:27 AM by developer_randi

    Assembly of pdfs - "insufficent data for an image" and blank page

    developer_randi

      Hello.

       

      We are using Adobe Livecycle Assembler 8.2 to assemble a set of pdfs uploaded by end-users together with a pdf that our application create. We have been using the assembler and pdf generator for å few years.

       

      Lately we have had a couple of situations that our custoumer is not very happy with.

       

      Situation 1:

      Assembly of 5 pdf is sucessful, but when we open the assembled pdf, one page (representing one attachemnt upload by an end-user) is displayed as an blank page. The attachment is correctly displayed in the table of content with a bookmark to the blank page. The first time scrolling over the blank page in Adobe Reader X, a popup with the message "Insufficent data for an image" is shown. Our applications has no knowledge that the assembly is not sucessful, so its difficult to identify the problem. It seems like the pdf is a scanned image (Producer: AXIS 70U Network Document Server). Have also viewed the assembled pdf in another pdf-reader, but still the same blank page and popup.

       

      Situation 2:

      Assembly of a few pdfs is sucessful, but when we open the assembled pdf, one page (representing one attachemnt upload by an end-user) is displayed as an blank page. So fare same as situation 1. However, the attachment is not displayed in the table of content and no popup is given when scrolling over the blank page representing the attachment, just a blank page.

       

      I am looking at a way of detecting pdf that would cause these problems (if I can't fix the pdfs). Maybe there is a way (programtically with iText?) to check if the pdf contains fonts and if not, regard it as a scanned image and mark it as problematic (most scanned pdfs are assembled just fine). Earlier on we had another problem when assembly pdfs, we then got an exception "PDFUnsupportedFeatureException: FlateDecode Filter". Our solution was then to use PDF Generator to convert all uploaded pdfs(attachments) to ps, then back to pdf and reassemble. This approch also works in situation 1 and 2. However we have no failing applications with exceptions now... Doing the pdf2ps2pdf for all attachments is very time consuming and not a good solution for our customer.

       

      Do anybody have a clue what is wrong with our pdfs? Is there a way of solving this when doing the assembly? Any DDX-tags that might do anything nice?

       

      The data in the pdfs are sensitive information, so I can't attach them to this post. Can send the metadata if requested.

       

      Any response would be appreciated.