I've got a datafile with NaNs. The plot is ok, but when I try to get GPVAL_DATA_Y_MIN I obtain 0.0 while the minimum value is in fact 420.0; when I substitute the NaNs by random numbers, then I get the right minimum, namely 420.0; so all I need is to enforce gnuplot to ignore NaNs when evaluating GPVAL_DATA_Y_MIN... is it possible ?

1.3    455.2
1.7    645.6
2.0    420.0
2.2    NaN
2.8    498.5

I've got a datafile with NaNs. The plot is ok, but when I try to get GPVAL_DATA_Y_MIN I obtain 0.0 while the minimum value is in fact 420.0; when I substitute the NaNs by random numbers, then I get the right minimum, namely 420.0; so all I need is to enforce gnuplot to ignore NaNs when evaluating GPVAL_DATA_Y_MIN... is it possible ?

1.3    455.2
1.7    645.6
2.0    420.0
2.2    NaN
2.8    498.5

• gnuplot

• What is your gnuplot version? Please post the complete (minimized) script to reproduce this issue. So far, I couldn't reproduce it with gnuplot>=5.0 – theozh Commented Mar 23 at 8:47
• With a extremely minimized version, the NaN value is in fact ignored and I retrieve the right y min value... odd... with my full version I get the right y min value only if I replace NaN by a number... otherwise I get zero... I'll have to check out both my script and my datafile and... waste time :( – ifffam Commented Mar 23 at 16:34
• so, I assume you must have some other settings in your extended script, which might make GPVAL_DATA_Y_MIN=0.0. Without having your script it's difficult to make any tests and conclusions. – theozh Commented Mar 24 at 8:50

1 Answer 1

Just add

set datafile missing NaN

at the beginning of the script and fet about everything :)

Granted, you'll not get into the deepness of the mystery, but life moves on, time is money and much more :)