Hi,
I'm Tom Veatch, unemployed furniture maker, plumber, software developer,
contemplating working on a program to build graphical scenes out of
abstract descriptions thereof and converting them to a visible/printable
output form such as PDF. I spent the last day learning about PDF and this
last two hours reading everything at gnupdf.org. It seems the project is
inactive for a couple of years since the Contact pages' developers' links
are broken as is the flyspray link with a lot of debug noise on it. I can
understand that the project could be perhaps depressingly ambitious since
PDF 1.7 is so enormous itself. I wanted to do two things in this email,
one to encourage the developers, and two to offer a simple user story which
might be an attractive development target or test case.
So congratulations on working on this project; I do think it will help the
world. Not only is PDF everywhere but its capabilities are expanding
greatly, and a PDF viewer could function as a multimedia display system, a
powerpoint display, a lecture presentation form, an internet UI, many
things that it isn't presently much doing, but that PDF 1.7 includes the
possibilities for. I am rooting for you!
The user story I have is like this. Consider creating a 3D wire frame
model of something, then making a PDF view of it. So I'd like to build a
software model of an object, say something rather like a PDF path object in
a 3D object coordinate space, being a set of 1 or more subpaths, each
subpath being a sequence of one or more straight-line or cubic-bezier curve
segments. It could constitude a mesh-like model of the outlines of some
physical object, for example. Then I'd like to place that object within a
world coordinate system at some location. Then I'd like to establish a
camera coordinate system with location and orientation thereof, and with
some window in its view, and finally I'd like to transform the object path
segments from object coordinates into their positions in the camera
coordinate system, then determine if they are clipped out or in to the
viewing window, and then draw them in 2D PDF graphics. Fiinally, prepend
and append the PDF header, xref table, etc.,etc., thereby generating an
actual PDF file representing a displayable view of my object. I wasn't
really thinking about sketchup PDF writing, but you could think of it that
way if you wanted to.
So shouldn't that be a nice simple example, a tutorial even, for using
GNUpdf library code to generate PDF files from the object description? I
thought so.
But I didn't find graphics commands in the GNUpdf library functions, nor a
tutorial of what a GNUpdf program might look like and how it would go about
creating a path ("0 0 100 100 re" in a PDF file, how would that be
generated by GNUpdf library functions? On the other hand it seems to me
that that would be a nice baseline simple case of how to use such a library
to do stuff people might be interested in. So I thought I'd offer it to
you as a thought experiment.
If you have suggestions or ideas for me please let me know.
Meanwhile I think I'm going to be writing code that generates PDF myself,
just for this project of mine. It doesn't have to do everything, but I
hope it will do something!
Again I am glad you folks are there and I wish you all success in this
project.
Regards,
Tom Veatch
PS Check out http://tomveatch.com/pffs which might be inspiring on the
point of how to get projects paid for...
