Calculating data inside pst-plot/postscript and conditionally coloring the plot

Question

I am getting lots of simulated data in tables with 13 columns, from where I want to plot the following: The x axis should be column number 12 (E), which is fine with plotNoX=12. The y axis should be column number 10 (pops) divided by column number 13 (degeneracy).

I think the problem can be done in postscript, but I don´t know how to access the data inside.

MWE with original data and scale graph:

!TeX program = xelatex

\documentclass{scrartcl}
\usepackage{filecontents}
\begin{filecontents}{Boltzmann-Plot.txt}
,    E_J,     J,   component, E_v,       v,  N,  id,   numlines,  pops,  errors,  specie,        E,         degeneracy
0,   204.32,  3.5, 1,         32314.13,  0,  3,  155,  6,         2.0503590454,   0.0549585571,  32518.45,  8
1,   339.81,  4.5, 1,         32314.13,  0,  4,  157,  6,         1.9458976329,   0.0468620867,  32653.94,  10
2,   102.37,  2.5, 1,         32314.13,  0,  2,  153,  6,         1.909089931,    0.0630666458,  32416.5,   6
3,   508.54,  5.5, 1,         32314.13,  0,  5,  159,  6,         1.6624777745,   0.047496399,   32822.67,  12
4,   203.53,  2.5, 2,         32314.13,  0,  3,  154,  6,         1.4258811354,   0.0373668708,  32517.66,  6
5,   507.3,   4.5, 2,         32314.13,  1,  5,  158,  6,         1.3639953283,   0.0228045864,  32821.43,  10
6,   710.18,  6.5, 1,         32314.13,  1,  6,  161,  6,         1.303716013,    0.0202061018,  33024.31,  14
7,   34.19,   1.5, 1,         32314.13,  1,  1,  151,  5,         1.2909918307,   0.0810218945,  32348.32,  4
8,   338.79,  3.5, 2,         32314.13,  1,  4,  156,  6,         1.2815394104,   0.0191664948,  32652.92,  8
9,   944.35,  7.5, 1,         32314.13,  2,  7,  163,  6,         1.2091683402,   0.0533139809,  33258.48,  16
10,  08.71,   5.5, 2,         32314.13,  2,  6,  160,  6,         1.1679741692,   0.0403405467,  33022.84,  12
11,  942.65,  6.5, 2,         32314.13,  2,  7,  162,  6,         0.9963321806,   0.023514499,   33256.78,  14
12,  101.8,   1.5, 2,         32314.13,  2,  2,  152,  5,         0.9200611356,   0.0756242618,  32415.93,  4
13,  1210.61, 8.5, 1,         32314.13,  2,  8,  165,  6,         0.911434806,    0.0380645658,  33524.74,  18
\end{filecontents}
\usepackage[]{pst-plot}
\pagestyle{empty}
\readdata[ignoreLines=5]{\Boltzmann}{Boltzmann-Plot.txt}

\begin{document}
\begin{pspicture}
\begin{psgraph}[
Ox=3.3,dx=0.1,Dx=0.1,xlabelFactor={ \cdot 10^4},
Oy=-1,ylogBase=10,
comma]{->}(3.2,-1)(3.5,1.1){12cm}{9cm}
\pstScalePoints(1e-4,1){}{log}
\listplot[plotNoMax=12,plotNo=9,plotNoX=12,yMinValue=0.00001,plotstyle=dots]{\Boltzmann}
\end{psgraph}
\end{pspicture}
\end{document}

edit: removed a aspect

edit2: for physical correctness, pops must be divided by degeneracy, not multiplied

The new pst-plot package has a command that allows you to read a column. However, it is not \psreadColumnData as claimed in the manual, rather it is \psreadDataColumn. This works nicely in that it allows you to read columns into different macros. Yet combining them with PSTricks will probably require some knowledge of postscript (see e.g. here), which I do not have. — , Feb 14 '18 at 01:50
Why not just read the entire file and compute the new data set? Then you can write that back to a file and use that new file to plot the data. — Werner, Feb 14 '18 at 05:36

score 3 · Accepted Answer · 2018-02-15T19:51:40.017

With an up-to-date pst-plot you can also use the ignoreLines optional argument for \psreadDataColumn.

Green ist y10*y12:

\documentclass{scrartcl}
\usepackage{filecontents}
\begin{filecontents*}{Boltzmann-Plot.txt}
    ,    E_J,     J,   component, E_v,       v,  N,  id,   numlines,  pops,  errors,  specie,        E,         degeneracy
    0,   204.32,  3.5, 1,         32314.13,  0,  3,  155,  6,         2.0503590454,   0.0549585571,  32518.45,  8
    1,   339.81,  4.5, 1,         32314.13,  0,  4,  157,  6,         1.9458976329,   0.0468620867,  32653.94,  10
    2,   102.37,  2.5, 1,         32314.13,  0,  2,  153,  6,         1.909089931,    0.0630666458,  32416.5,   6
    3,   508.54,  5.5, 1,         32314.13,  0,  5,  159,  6,         1.6624777745,   0.047496399,   32822.67,  12
    4,   203.53,  2.5, 2,         32314.13,  0,  3,  154,  6,         1.4258811354,   0.0373668708,  32517.66,  6
    5,   507.3,   4.5, 2,         32314.13,  1,  5,  158,  6,         1.3639953283,   0.0228045864,  32821.43,  10
    6,   710.18,  6.5, 1,         32314.13,  1,  6,  161,  6,         1.303716013,    0.0202061018,  33024.31,  14
    7,   34.19,   1.5, 1,         32314.13,  1,  1,  151,  5,         1.2909918307,   0.0810218945,  32348.32,  4
    8,   338.79,  3.5, 2,         32314.13,  1,  4,  156,  6,         1.2815394104,   0.0191664948,  32652.92,  8
    9,   944.35,  7.5, 1,         32314.13,  2,  7,  163,  6,         1.2091683402,   0.0533139809,  33258.48,  16
    10,  08.71,   5.5, 2,         32314.13,  2,  6,  160,  6,         1.1679741692,   0.0403405467,  33022.84,  12
    11,  942.65,  6.5, 2,         32314.13,  2,  7,  162,  6,         0.9963321806,   0.023514499,   33256.78,  14
    12,  101.8,   1.5, 2,         32314.13,  2,  2,  152,  5,         0.9200611356,   0.0756242618,  32415.93,  4
    13,  1210.61, 8.5, 1,         32314.13,  2,  8,  165,  6,         0.911434806,    0.0380645658,  33524.74,  18
\end{filecontents*}
\usepackage{pst-plot}
\pagestyle{empty}
\readdata[ignoreLines=5]{\Boltzmann}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{12}{,}{\xData}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{10}{,}{\yDataOne}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{13}{,}{\yDataTwo}{Boltzmann-Plot.txt}
\def\DATA{
  /X [\xData] def 
  /Y1 [\yDataOne] def  
  /Y2 [\yDataTwo] def             % save as array
  0 1 X length 1 sub {            % on stack is the loop variable
    /Index ED                     % save it 
    X Index get 
    Y1 Index get Y2 Index get mul % y10*y13
  } for 
}   

\begin{document}
\begin{pspicture}
\begin{psgraph}[
    Ox=3.3,dx=0.1,Dx=0.1,xlabelFactor={ \cdot 10^4},
    Oy=-1,ylogBase=10,comma]{->}(3.2,-1)(3.5,1.5){12cm}{9cm}
    \pstScalePoints(1e-4,1){}{log}
    \psset{plotstyle=dots,dotscale=1.5}
    \listplot[linecolor=green]{\DATA}
    \listplot[plotNoMax=12,plotNo=9,plotNoX=12,linecolor=red]{\Boltzmann}
    \listplot[plotNoMax=12,plotNo=12,plotNoX=12,linecolor=blue]{\Boltzmann}
\end{psgraph}
\end{pspicture}
\end{document}

Coloring the dots depending to column v use:

\documentclass{minimal}

\usepackage{pst-plot}
\pagestyle{empty}
\readdata[ignoreLines=5]{\Boltzmann}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{12}{,}{\xData}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{10}{,}{\yDataOne}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{13}{,}{\yDataTwo}{Boltzmann-Plot.txt}
\psreadDataColumn[ignoreLines=1]{6}{,}{\ColValue}{Boltzmann-Plot.txt}
\makeatletter
\def\specialPlot{%
  \begin@SpecialObj
  \addto@pscode{
    /X [\xData] def 
    /Y1 [\yDataOne] def  
    /Y2 [\yDataTwo] def             % save as array
    /Color [\ColValue] def
    0 1 X length 1 sub {            % on stack is the loop variable
      /Index exch def                     % save it 
      X Index get 1.e-4 mul
      Y1 Index get Y2 Index get mul log % log(y10*y13) 
      \tx@ScreenCoor 
      gsave
      3 0 360 arc 
      Color Index get /No exch def
      No 0 eq { 1 0 0 setrgbcolor } if 
      No 1 eq { 0 1 0 setrgbcolor } if 
      No 2 eq { 0 0 1 setrgbcolor } if 
      fill
      stroke
      grestore
    } for
  }
  \use@pscode
  \end@SpecialObj
}  
\makeatother
\begin{document}

\begin{pspicture}
\begin{psgraph}[
    Ox=3.3,dx=0.1,Dx=0.1,xlabelFactor={ \cdot 10^4},
    Oy=-1,ylogBase=10,comma]{->}(3.2,-1)(3.5,1.5){12cm}{9cm}
    \psset{plotstyle=dots,dotscale=1.5}
    \specialPlot
\end{psgraph}
\end{pspicture}
\end{document}

Comment fill for circles instead of solid dots

Hello Herbert, this is working for me perfectly. The coloring I mentioned in the heading (and did not remove in edit) was for coloring dependend on column "v". So if column v=0, red and so on. Is there is a way to do so? — Niklas, Feb 14 '18 at 13:40

score 1 · Answer 2 · 2018-02-14T18:00:01.017

I'm not sure if this is an answer. At least it is an anchor for the command \psreadColumnData, which does not exist, and the information that one should use \psreadDataColumn instead. These commands work fine.

UPDATE: Accommodated you revision: div instead of mul.

\documentclass{scrartcl}
\usepackage{tikz}
\usetikzlibrary{math}
\usepackage{filecontents}
%,    E_J,     J,   component, E_v,       v,  N,  id,   numlines,  pops,  errors,  specie,        E,         degeneracy
\begin{filecontents}{Boltzmann-Plot.txt}
0,   204.32,  3.5, 1,         32314.13,  0,  3,  155,  6,         2.0503590454,   0.0549585571,  32518.45,  8
1,   339.81,  4.5, 1,         32314.13,  0,  4,  157,  6,         1.9458976329,   0.0468620867,  32653.94,  10
2,   102.37,  2.5, 1,         32314.13,  0,  2,  153,  6,         1.909089931,    0.0630666458,  32416.5,   6
3,   508.54,  5.5, 1,         32314.13,  0,  5,  159,  6,         1.6624777745,   0.047496399,   32822.67,  12
4,   203.53,  2.5, 2,         32314.13,  0,  3,  154,  6,         1.4258811354,   0.0373668708,  32517.66,  6
5,   507.3,   4.5, 2,         32314.13,  1,  5,  158,  6,         1.3639953283,   0.0228045864,  32821.43,  10
6,   710.18,  6.5, 1,         32314.13,  1,  6,  161,  6,         1.303716013,    0.0202061018,  33024.31,  14
7,   34.19,   1.5, 1,         32314.13,  1,  1,  151,  5,         1.2909918307,   0.0810218945,  32348.32,  4
8,   338.79,  3.5, 2,         32314.13,  1,  4,  156,  6,         1.2815394104,   0.0191664948,  32652.92,  8
9,   944.35,  7.5, 1,         32314.13,  2,  7,  163,  6,         1.2091683402,   0.0533139809,  33258.48,  16
10,  08.71,   5.5, 2,         32314.13,  2,  6,  160,  6,         1.1679741692,   0.0403405467,  33022.84,  12
11,  942.65,  6.5, 2,         32314.13,  2,  7,  162,  6,         0.9963321806,   0.023514499,   33256.78,  14
12,  101.8,   1.5, 2,         32314.13,  2,  2,  152,  5,         0.9200611356,   0.0756242618,  32415.93,  4
13,  1210.61, 8.5, 1,         32314.13,  2,  8,  165,  6,         0.911434806,    0.0380645658,  33524.74,  18
\end{filecontents}
\usepackage[]{pst-plot}
\pagestyle{empty}
\readdata[ignoreLines=5]{\Boltzmann}{Boltzmann-Plot.txt}
\psreadDataColumn{12}{,}{\xData}{Boltzmann-Plot.txt}
\psreadDataColumn{10}{,}{\yDataOne}{Boltzmann-Plot.txt}
\psreadDataColumn{13}{,}{\yDataTwo}{Boltzmann-Plot.txt}
\def\DATA{ % tried to adapt from https://tex.stackexchange.com/a/141094/121799
  /X [\xData] def /Y [\yDataOne] def  /Z [\yDataTwo] def %save as array
  0 1 X length 1 sub {            % on stack is the loop variable
    /Index ED                     % save it 
    X Index get Y Index get Z Index get mul       % get x  and y as y/z
  } for }   
\typeout{\xData}    
\typeout{\yDataOne}  
\typeout{\yDataTwo}  
\begin{document}
\begin{pspicture}
\begin{psgraph}[
Ox=3.3,dx=0.1,Dx=0.1,xlabelFactor={ \cdot 10^4},
Oy=-1,ylogBase=10,
comma]{->}(3.2,-2)(3.5,1.1){12cm}{9cm}
\pstScalePoints(1e-4,1){}{log}
\listplot[plotstyle=dots]{\DATA}
%\listplot[plotNoMax=12,plotNo=9,plotNoX=12,yMinValue=0.00001,plotstyle=dots]{\Boltzmann}
\end{psgraph}
\end{pspicture}
\end{document}

Calculating data inside pst-plot/postscript and conditionally coloring the plot

2 Answers2