El Dimarts, 1 de setembre de 2015, a les 23:34:10, Adam Reichold va escriure: > Hello again, > > Am 01.09.2015 um 00:16 schrieb Albert Astals Cid: > > El Dissabte, 25 de juliol de 2015, a les 14:26:48, Adam Reichold va escriure: > >> Hello, > >> > >> the attached patch adds an option to strip the encryption from a > >> document by doing a complete rewrite but ignoring the original > >> encryption parameters and removing the encryption entry form the trailer > >> dictionary. It also exposes this via the two Qt frontends. > >> > >> This might be useful for a separate command-line utility but for now my > >> primary use case is submitting a rewritten but unencrypted document to > >> CUPS for printing instead of converting to PostScript. > > > > Pushed. > > I am very sorry, but during further testing, I noticed that the raw > stream copying done in "PDFDoc::writeObject" breaks documents generated > using this flag. It seems this yields the encrypted data being written > instead of the decrypted data. > > The attached patch tries to solve this by passing the "stripEncryption" > flag to the "writeObject" method to suppress the raw stream copying, > which however yields a horribly large patch for what is effectively a > single-line change due to the recursive calls to "writeObject" from > various places. > > While this changes does yield correctly decrypted files which CUPS > prints properly, it also "strips" various other stream filters like > "Flate" yielding much larger documents. I think the better solution > would be to reliably detect this situation within "writeObject" to > suppress the raw stream copying on a case-by-case basis, but I have not > yet managed to implement this. I would be grateful for any advice on how > to do this.
I've reverted the change altogether until we find a way to solve this. Cheers, Albert > > Best regards, Adam. _______________________________________________ poppler mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/poppler
