[pdf-devel] Encoded PS Stream

[Matt Davis]

Hello,
I have had experience developing PDF tools.  I am the original writer
and maintainer of PDFresurrect (which needs a serious rewrite), and I
am working on a new tool.  Granted, all of my PDF parsing routines
were homegrown, therefore producing no dependencies in my previous
project.  I think it is way awesome that GNU PDF aims to be that
"go-to" library for dealing with PDF creation and parsing!

Anyways, I have a question that has really been troubling me.  I am
decoding a PS stream (and maybe this question is more appropriate on a
PS related list).  Anyways, there are no explicit space characters in
the PS document; instead all of the spacing is done by moving the text
cursor.  The amount the position moves can vary greatly, and I am
having a hard time deducing spaces from these values.  Any tips would
be helpful. Section 9.4 (Text Objects) in the PDF3000 spec, which has
some spacing help, does help some, but I feel like my code is just
guessing when a space is.  I have a hard coded value that looks at
last versus current character position and if that distance is large
enough, I assume it is a space.  I don't like that solution, since I
do not know how large to make the 'hard coded threshold'.  Any
tips/help would be appreciated.

Thanks!

-Matt