Re: The need for an integrated PDF reader/Implementation ofNSPDFImageRep

[stefan]

Hi,

> I was just tossing this around in my brain, but I wanted to mention it.
>
> Currently, one way to implement NSPDFImageRep is by using ghostscript.
> Using
> ghostscript to implement this would have some advantages and
> disadvantages:
>
> Advantages:
>
> 1) The PDF code in ghostscript is tested and works.
> 2) ghostscript is GPL and copyright is owned by the FSF
>
> Disadvantages:
>
> 1) ghostscript is an external program
> 2) slow
>
> Would it be better to use ghostscript or to implement our own internal PDF
> reader based on it?

i think implementing our own renderer is the better choice. I've already
made some investigations for my Preview-Application ViewIt
(http://mac.wms-network.de/gnustep/viewit/viewit.html). It currently uses
an xpdf/ghostscript based combination for PDF rendering. My plans are to
implement a mechanism that produces custom (PDF)ImageReps by using the
code from xpdf. xpdf does rendering very well and it is modular so that it
is easy to write new output devices. xpdf uses this output devices to
'visualize' PDF content. An output device must implement some methods that
can be translated (nearly) directly to postscript, so it may be possible
to use display-postscript here.

ciao
Stefan Kleine Stegemann