1/wow cool .. does your pdf implementations includes correct non latin fonts handling (font subsampling),
The reason I chose AggPas (instead of say FPCanvas) is because AggPas has far superior anti-aliased and sub-pixel graphics rendering. (see here) Saying that, a FPCanvas/FPImage version will probably be implemented too.
font-subset embedding (partial font embedding) will be implemented soon.
@marcinh:fpReport has been implemented and is being used internally here. I was only tasked in implementing it. Michael is in charge of making it public.