Re: [Plplot-devel] Bug in notcrossed() function of plfill.c

[Phil R. <p.d...@gm...>]

Hi Alan
I will try to do this. Unfortunately the fail case of mine was very well wrapped in my own wrapper code so it might be a challenge to generate a simple test.

Phil

-----Original Message-----
From: "Alan W. Irwin" <ir...@be...>
Sent: ‎24/‎06/‎2015 04:03
To: "Phil Rosenberg" <p.d...@gm...>
Cc: "plp...@li..." <plp...@li...>
Subject: Re: [Plplot-devel] Bug in notcrossed() function of plfill.c

Hi Phil:

I have been looking carefully at the parts of the fill code that are
affected by notcrossed, and I think we are dealing with a fairly
complex issue here that is larger than just the criterion that is used
to decide on the PL_NEAR_PARALLEL status.  So to make sure that the
final fix really addresses the PLplot failure you discovered that
started this thread, I would greatly appreciate it if you sent me a
test case that replicates that failure for the current master tip
version of PLplot.  Ideally, if your original test case requires
access to data files and/or data processing to determine the data that
are being plotted, it would be good to figure out (say with a debugger
or appropriate print statement) the exact arguments to plfill that
cause the issue, and simply use those exact arguments for a
self-contained test case that triggers the exact same failure
that you originally discovered.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________