Re: [Plplot-devel] Reducing postscript rounding inconsistencies

[Alan W. I. <ir...@be...>]

On 2006-06-02 09:21-0700 Alan W. Irwin wrote:

> On 2006-06-02 17:14+0100 Andrew Ross wrote:
>
>>
>> This should now be fixed for both the ps and psttf drivers. It's just a
>> result of the way fprintf (and the C++ streams) round numbers to a given
>> precision. I've added a macro for use in calls to printf etc to set any
>> small negative numbers to zero hence avoiding them being printed as -0.000.
>>
>> Andrew
>
> Thanks, Andrew, for eliminating this particular source of rounding noise.

P.S. I just did the following test on my Ubuntu system:

(for postscript in *.ps; do echo $postscript; sed 's?-0.000?0.000?g'
<$postscript |diff - ../examples/$postscript  ; done) |less

I was in the standard examples results directory I built in mid-May, and it
turned out that replacing all -0.000 by 0.000 in those old postscript files
gave identical results to today's build (which included your above fix). So
as a result of your fix (a) there should be no -0.000 results left anywhere
in any postscript file, and (b) that is the only change introduced by your
fix.

Good work!

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________