PDFdev is a service provided by PDFzone.com | http://www.pdfzone.com
_____________________________________________________________

> I am working on an application to decode and reencode the encoded
> data
> in the pdf..and am using third party encoding/decoding libraries
> for
> this.
> now the problem is..I am not able to decode the lzw buffer in
> pdf..
> i get the error bad compressed format..but if i rip off initial
> 2
> bytes the buffer gets decoded. Also I am able to decode the
> buffer
> even if i rip off till 5 bytes..is the lzw buffer contained in
> some
> header like flate buffer or whats going wrong..?

Are you carefully following the strict rules for white space
to skip after the "stream" keyword?
>
> One more issue...i decode the flate buffer..reencode it..the new
> encoded data is not same as that of the pdf encoded data..

This is normal. There are many different valid Flate streams from
the same input data.

> so i am not able to reopen the regenerated pdf file

That should not cause it. You probably have other issues like
- wrong whitespace after "stream"
- failing to rebuild xref table correctly
- failing to update Length keyword correctly

You can start, though, by making your program dump a copy of the
decompressed
data to a log file. You can look to see if it is reasonable, and whether
it has changed.

Aandi


To change your subscription:
http://www.pdfzone.com/discussions/lists-pdfdev.html

Reply via email to