Vous (Owen) avez ?crit : > On Sat, 15 Aug 2009 12:21:21 -0400 > > John Culleton <john at wexfordpress.com> wrote: > > My old nemesis LSI is still insisting on the above PDF format; > > otherwise they will run the cover file through a process that > > rasterizes the file. When it comes to access to Amazon it is their > > bat and their ball and you have to play the game their way. > > *WARNING* I know zilch about PDF specs > > > have you tried editing the first line in the pdf file to read > %PDF-X/1a:2001 (or whatever a PDF-X/1a:2001 file has as its first line) > > I just edited a PDF file that had %PDF-1.4 as its first line. Acrobat > Reader flashed up a message (I haven't done speed reading) but it said > something about "Rebuilding" > > Ocular just opened it > > Maybe that's all they check?
You can?t insert random data in a PDF file because by doing it, you invalidate the xref table (kind of index). While the file can still be processed at a cost of rebuilding the xref table by analysing the content of the file etc. the file becomes basically corrupted. About PDF-X/1a, the code that Thach did write for the GSOC program should find his way in main development branch, Trunk, in coming months. -- Pierre Marchand
