Hi Sorin,
I don't know exactly what you mean. Please give an example.
In the Oxygen Developer's transformation scenario, I configure the FO processor to use the method "at", which will result in an Area Tree XML file, as described in the Apache FOP website as one of the intermediate formats:
I think I need to use / manipulate this area tree, because I want to do some index processing that can only be done after the page-flow has been arranged by FOP. That's because the XSL-FO 1.1 properties for indexing (
http://xmlgraphics.apache.org/fop/compl ... ng-section) are not supported by FOP. Let me give an example:
If I use normal <fo:page-number-citation> elements for every indexed term in the text flow, I might end up with an index entry like this:
pragmatic 29, 30, 30, 35, 36, 37, 49, 59, 59, 59, 59, 109
But normally we would expect an index entry to look something like this:
[quuote]pragmatic 29, 30, 35-37, 49, 59, 109[/quote]
To eliminate duplicate page numbers can only be done in the area tree of FOP, as far as I know, because pagination information isn't available in the FO file. So far things seemed clear to me. My problem began when I recognized that the page numbers differ from the PDF output when I use the "at" method in the Oxygen transformation scenario's FO configuration. Which is then making the whole area tree useless, because I cannot be sure that the page numbers in the area tree's index are identical with the page numbers in the PDF. And that's where the info from
http://xmlgraphics.apache.org/fop/1.1/i ... html#usage comes in: to ensure that the area trees makes the same page breaking decisions as the pdf renderer, the area tree renderer has to mimic the pdf renderer.
Hope I'm writing in an intelligible enough way to make the problem clear. If not, don't hesitate to ask further for specifics.
Best regards,