The following script has some strange problems. When I load it after starting gnuplot, I get the following error:
gnuplot> load "test_plot_data.gp"
"test_plot_data.gp" line 46: warning: selected_columns is not a string variable
"lh_01":"rh_01"
"test_plot_data.gp" line 62: undefined variable: with
gnuplot>
When I load it again, it correctly selects the volumes from the specified columns “lh_01”:“rh_01” until it reaches “pause -1” in the loop. When it continues, the column names “lh_02”:“rh_02” are updated correctly, but it still outputs the values from the previous run (“lh_01”:“rh_01”).
When I load this script afterwards in the same Gnuplot session, I always get the values of “lh_02”:"rh_02", although the column names are always set correctly. Do I need to make some special settings to avoid this behaviour?
reset
set datafile missing "NaN"
set datafile columnheader
# Set up for subjects and column usage
$Data << EOD
# TP1
gp HS Age lh_01 rh_01 lh_02 rh_02
2 0 37.74 7030.26 7353.44 70.26 73.44
2 0 26.42 7652.64 8566.62 76.64 85.62
2 0 30.20 6938.74 7489.41 69.74 74.41
2 0 27.68 6619.97 6224.74 66.97 62.74
2 1 18.87 7472.23 8635.48 74.23 86.48
2 0 34.56 6612.52 7347.70 66.52 73.70
2 0 33.54 7049.64 7137.09 70.64 71.09
3 1 21.47 8327.11 8745.67 83.11 87.67
1 0 22.00 7924.66 8694.33 79.66 86.33
1 0 24.96 7605.58 8072.01 76.58 80.01
3 1 12.10 8883.32 9230.87 88.32 92.87
2 0 24.72 8271.73 8465.82 82.73 84.82
# TP 2
gp HS Age lh_01 rh_01 lh_02 rh_02
2 0 38.84 7069.54 7380.43 70.54 73.43
2 0 33.14 7705.56 8076.87 77.56 80.87
2 0 39.78 7346.33 7712.80 73.33 77.80
2 0 28.98 6963.54 7171.83 69.54 71.83
2 1 23.71 6830.13 8113.97 68.13 81.97
2 0 39.93 7438.96 7020.43 74.96 70.43
2 0 44.92 6943.02 7204.57 69.02 72.57
3 1 26.82 8245.33 8335.07 82.33 83.07
1 0 24.93 7396.14 7853.01 73.14 78.01
1 0 29.79 7421.25 7810.83 74.25 78.83
3 1 14.55 8672.95 8937.76 86.95 89.76
2 0 27.30 8668.1 8506.81 86.1 85.81
EOD
LABELS = "lh_01 rh_01 lh_02 rh_02"
# Search string for hemispheres (adjust based on your data)
search_string = 'lh'
# Loop through selected labels and group labels
do for [selected_label in LABELS] {
if ( strstrt( selected_label, search_string ) > 0 ) {
# Strip hemisphere label
selected_label = substr( selected_label, strlen( search_string ) + 1, strlen( selected_label ) )
# Set column names
left = "lh".selected_label
right = "rh".selected_label
# Prepare selected columns for plotting
selected_columns = sprintf("\"%s\":\"%s\"", left, right)
print selected_columns
# Store temporary table
set table $Temp
plot $Data using @selected_columns with table
unset table
print $Temp
pause -1
}
}
### End of script
It's not clear to me why you are taking a detour via sprintf() and @ macro.
Simply take the string variables left and right into your plot command, which is plotting the columns with that headers.
plot $Data u left:right with table