Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!sdd.hp.com!decwrl!ucbvax!bloom-beacon!EXPO.LCS.MIT.EDU!keith
From: keith@EXPO.LCS.MIT.EDU (Keith Packard)
Newsgroups: comp.windows.x
Subject: Re: Need help with arcs in R4
Message-ID: <9008160316.AA18330@xenon.lcs.mit.edu>
Date: 16 Aug 90 03:16:20 GMT
References: <6304@gssc.UUCP>
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 18


> It appears that the arc drawing code in R4 mi has changed drasticly.
> When I run the xbench arc tests there appears to be some start/path 
> combinations that don't draw at all and others that may be drawing in 
> the wrong direction. The server I'm using uses only mi code for drawing 
> arcs so I'm assuming that the problem is in miPolyArc(). Does anyone know 
> of any fixes to miarc.c? Has anyone got miPolyArc() to work in their server?

Yes, the R4 arc code had a wide range of bug fixes, but the basic algorithms
have remained the same since R3.  Certainly the interpretation of the protocol
values hasn't changed.

I have never seen the troubles you are having, and that code has seen some
rough testing.  If you can put together a sample program which draws incorrect
arcs, please send it to xbugs@expo.lcs.mit.edu and we'll have a closer look.

Keith Packard
MIT X Consortium