Re: Should NaN be reported as "UNDEFINED" ?

[Daniel J S. <dan...@ie...>]

Daniel J Sebald wrote:

> I think that is good for 4.2.  But I would like to see eventually flexibility with defining both "missing" and "undefined", with possibly the ability to have more than one definition.  For example,
> 
> set missing "Inf" "+Inf" "-Inf" "inf" "+inf" "-inf"
> set undefined "N/A" "outdated"

I have updated the patch in [ 1508316 ] for this (just for "datafile missing").  I think users will be happy to have multiple strings.

There are some string linked-list helper routines I added in gadgets.c and gadgets.h.  This may not be the best location for those.  In fact, if someone already knows of some simple character string linked-lists definitions inside gnuplot, they should be reused instead.

Here is some example behavior (I put an Inf and -Inf in the data file):

gnuplot> set pm3d map
gnuplot> set dgrid3d 100,100,1
gnuplot> splot 'example.dat' u 1:2:3
         Axis range undefined due to improper data values. NaN? Inf?

gnuplot> set datafile missing "NaN"
gnuplot> splot 'example.dat' u 1:2:3
         Axis range undefined due to improper data values. NaN? Inf?

gnuplot> set datafile missing "NaN" "Inf" "-Inf"
gnuplot> show datafile missing

        The following in datafile are interpreted as missing values

          NaN
          Inf
          -Inf

gnuplot> splot 'example.dat' u 1:2:3

Dan