Save as pdf deleted bookmarks when combining pdf-files

Question

Hello,
when I save several books each to a pdf-file with save as pdf (or printing to .ps and converting afterwards), the bookmarks are deleted when I combine the pdf-files with Acrobat. Only one bookmark for the 1st. file exists in the new pdf-file.
The reference of the bookmarks in the original generated file is e.g. "g3.619626".
These references are not anymore available in the combined files.
I've tried several settings before saving as pdf but nothing helped.
Ther are no problems of combining ordinary pdf-files with bookmarks if they are generated by other programs then Framemaker.
I need to combine files as well within a book, as "save to pdf" is not able, to generate one file with different pages sizes. I then need to generate a separate pdf-file of the pages with the for e.g. bigger page size and combine it afterwards.
In that project I'm working there will over all something between 1000 and 1500 pages from maybe 10 books with hunderts of bookmarks plus additionally separete pfd files to combine. So it's not an option to treat every bookmark before combining the pdf-files.
My system is FM11.02.384 / Acrobat Pro 11.0.3 / Windows 7 64 Bit

Answer 1

Never having had a need to do it, and Adobe documentation for Acrobat being what it is, I can't tell if AcroPro is supposed to be able to merge bookmarks when merging PDFs. It would be an unsurprising feature, and you do say it works on PDFs generated by means other than FM.
Here's a hack that might work, if the variable-page-size rendering turns out to be out of reach. You can batch* this.
Open each component PDF and re-save it with some otherwise harmless optimization (or even none). This will probably cause AcroPro to re-write the internal structures of each PDF, which might then be amenable to merged bookmarks when you re-open them for merging.
* We do a similar thing for a slightly different problem. We have a software-in-a-box XML publishing app that, at just some seats, is generating PDFs that are tagged for byte serving (fast web view), but are not in fact encoded for it. This causes AcroReader plugin and entire browser crashes for web end users. The s/w vendor can't even tell us where the option is for controlling that feature, because they get the PDF gen code from some other 3rd party. So we routinely batch process all generated PDFs, with, oddly enough, NO "commands" selected. The re-save alone fixes it.