Hello: Is there a way to load/parse input and retrieve a PDDocument *without* having the input stream closed "automatically"? Or alternately, is there a way to hook into parser's processing of the end-of-document *before* the stream is closed.
I need to process a 'print stream', which is a set of valid PDF documents concatenated into a single large file. I'd like to record the byte offsets into the large file for each of the 'sub-files' ... and was hoping to get this info in a single pass using the position reported by the input stream's channel, but now I find that the PDFParser closes its input stream when finished. I don't see a way to get visibility on a [still-opened] stream, unless I sub-class PDFParser or write one-off code to scan the input file for beginning-/ending-of-document markers during a separate pass. Any thoughts? Regards, Brad Harper
