valid/invalid points [was: Patch for new "stats" command]

[Ethan M. <merritt@u.washington.edu>]

On Sunday 08 November 2009, Philipp K. Janert wrote:
> We'd like to hear feedback and suggestions. Is this
> useful? Are we missing anything?

- I don't know how to interpret this behaviour:
gnuplot> stats '-' using (1)
input data ('e' ends) > 1   
input data ('e' ends) > 2   
input data ('e' ends) > NaN
input data ('e' ends) > 4
input data ('e' ends) > 5

* FILE:
  Records:      4
  Invalid:      0
  Blank:        0
  Data Blocks:  1

Why is the NaN not listed as either "invalid" or "blank"?
Same thing happens for '?' or Inf or junk.
I don't think this is being tracked correctly.

NB: We exchanged Email off-list about making df_readline() more consistent
about returning DF_UNDEFINED, DF_MISSING, and so on.  Totally aside from
keeping statistics, does anyone object to making
      plot 'foo' 
      plot 'foo' using 1:2
      plot 'foo' using ($1):($2)
all consistently return DF_MISSING and DF_UNDEFINED?
Right now all three behave differently.