From: Alan W. Irwin <irwin@be...> - 2009-08-27 04:04:30
Valery Pipin reported the following warning to me:
*** PLPLOT WARNING ***
Unrecognized cmap1 format #ff0000 100 0
At first glance that is impossible. The line in question is the last
line in cmap1_default.pal, and that is just as perfectly formatted as
the other lines in that file.
However, a closer look at the code in plctrl.c showed a problematic
floating-point test for whether pos[i] (calculated from the above 100 value
by converting to double and multiplying by 0.01) was in the range from 0. to
1. I have now fixed that test as of revision 10351 to account for small
floating-point errors in the calculation of pos[i]. Valery hasn't yet
gotten back to me, but when he does I presume he will tell me that fix
This whole issue shows how important it is to run PLplot on a large variety
of hardware with different floating point errors so that we find problematic
floating-point comparisons like what I just spotted and fixed in plctrl.c.
Anyhow, if you have access to PowerPC's, sparcstations, Dec Alphas, IBM
mainframes or other non-Intel hardware, now would be a good time to do some
thorough checking of PLplot on that hardware.
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); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project