Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!sun-barr!newstop!male!central!csccat!larry
From: larry@csccat.cs.com (Larry Spence)
Newsgroups: comp.lang.postscript
Subject: Re: Spline data points to Bezier curve
Summary: Look before you leap
Message-ID: <4314@csccat.cs.com>
Date: 5 Jun 91 19:50:08 GMT
References: <1991Jun3.095901.3793@desire.wright.edu>
Reply-To: larry@csccat.UUCP (Larry Spence)
Organization: Computer Support Corporation. Dallas,Texas
Lines: 25

In article <1991Jun3.095901.3793@desire.wright.edu> cse0080@desire.wright.edu writes:
>Help!
>
>I need to know how to generate spline curves in PostScript when
>all the data I have on hand is an array of data points that lie on 
>the curve itself.  (IE: no control points for the curveto operator)
>
>Anyone know a good way to do this?

The method will depend on what flavor of spline was used, uniform or
non-uniform (I'm assuming that it's cubic and non-rational).  If it's
non-uniform, you'll need to know what form of parametrization was used, 
plus the end conditions (there are some weird ones out there!).  You might 
take a look at Farin's book on curves and surfaces as a starting point.

That may sound nitpicky, but I had to convert a bunch of Kanji character
outlines that were in interpolating-spline form to PostScript, and found
that all sorts of bizarre assumptions were made about the end curvatures,
etc.  I found out _after_ I had written a straightforward conversion
algorithm, of course.  Duhhhhh.... %(

-- 
Larry Spence
larry@csccat.cs.com
...{uunet,texsun,cs.utexas.edu,decwrl}!csccat!larry