4

MWE:

\documentclass{scrartcl}
\usepackage{tikz}
\usetikzlibrary{datavisualization.formats.functions}

\begin{document}

  \begin{tikzpicture}
    \datavisualization[
                       scientific axes={clean},
                       all axes = grid,
                       x axis = {
                                 min value = 0,
                                 ticks = {
                                          major = {
                                                   also at = 1.5 as $T_0$,
                                                   at = 2 as $T_0 + T_1$
                                                  },
                                          stack
                                         },
                                 label = $t$
                                },
                       y axis = {
                                 min value = 0,
                                 ticks = {major = {also at = 4.5 as $K$}},
                                 label = $y(t)$
                                },
                       visualize as smooth line/.list = {
                                                         curve,
                                                         line 1,
                                                         line 2
                                                        },
                       line 2 = {style = dashed}
                      ]
    data[
         set = curve,
         format = function
        ] {
           var x : interval[1.5 : 6];
           func y = 9/2 * (1 - exp(3 - 2 * \value x));
          }
    data[set = line 1] {
                        x,   y
                        1.5, 0
                        2,   4.5
                       }
    data[set = line 2] {
                        x, y
                        2, 4.5
                        2, 0
                       };
  \end{tikzpicture}

\end{document}

at = 2 as $T_0 + T_1$ completely disabling the other ticks and ignores also at = 1.5 as $T_0$:

figure 1

If I change the line 16 to also at = 2 as $T_0 + T_1$ than I get

figure 2

So, how can one get the custom tick text without disabling/overpainting the automatically generated ticks? The second question is how can be T_0 and T_0 + T_1 placed without overlapping?

Thank you for your help and effort in advance!

Su-47
  • 2,508
  • How much text would a TikZ tick TeX if a TikZ tick could TeX text? – Peter LeFanu Lumsdaine Nov 27 '19 at 19:56
  • Hello @Peter LeFanu Lumsdaine! How should I understand your statement? – Su-47 Dec 05 '19 at 20:25
  • Just as a joke: several words in the title sounded similar, and fit into the format of “How much wood would a woodchuck chuck, if a woodchuck could chuck wood?” – Peter LeFanu Lumsdaine Dec 05 '19 at 20:59
  • Hello @Peter LeFanu Lumsdaine! This is funny, but what about my question? – Su-47 Dec 12 '19 at 15:01
  • I think with the change: major = {at = 1.5 as $T_0$, also at = 2 as $T_0 + T_1$ } , To be correct. – M.Ahmadi Dec 30 '19 at 16:06
  • Why not using pgfplots for more options... See my edited answer. – cis Dec 30 '19 at 20:52
  • I'm not completely sure, but I believe the "datavisualization" library is a holdover from older TikZ libraries. It was replaced by pgfplots. – cis Dec 31 '19 at 15:45
  • Hello @M.Ahmadi! Thank you for your comment. If one use your suggestion, than all other ticks, except of 1.5 and 2 are removed. I wish to let the automatically generated ticks and only replace/add new ticks (with labels) by 1.5 and 2, like nearly done in the below answer. I also don't want to set all ticks manually. – Su-47 Jan 09 '20 at 20:13
  • Hello @cis! Thank you for your answer and comments! I really like to stay by datavisualization, because I have no experience with pgfplots and like said in the TikZ & PGF manual on page 337 is pgfplots an alternative for the datavisualization not a replacement. – Su-47 Jan 09 '20 at 20:19
  • @Su-47 Yes, well, that may be, but almost nobody uses datavisualization anymore. Compare on this page: datavisualization: 76 questions. pgfplots: 7196 questions. I would also have to learn the dv-syntax here and before I do that, I would rather work out how to do with pgfplots. PS: It is also questionable whether dv is still maintained or developed at all (considering pgfplots). – cis Jan 10 '20 at 14:48

3 Answers3

3

€dit³

A suggestion with pgfplots:

I think, you want something like that...

· One methode is setting extra x ticks, see "T_0+T_1"; same game with "K" by using extra y ticks.

· Another method is to add special labels as a draw annotation, see "T_0", "T_1" and "T_a", "T_b":

If you want to use different tick lenghts for your labels, you could prefer the second method.

The following shows both ways.

enter image description here

\documentclass[border=5pt, tikz]{standalone}
\usepackage{pgfplots}
\pgfplotsset{compat=newest}
\usetikzlibrary{backgrounds}
\begin{document}
\pgfdeclarelayer{foreground}
\pgfsetlayers{background,main,foreground}
\begin{tikzpicture}
\pgfmathsetlengthmacro\MajorTickLength{
\pgfkeysvalueof{/pgfplots/major tick length}*0.5}
\pgfmathsetlengthmacro\AxisShift{-4.5pt}
\begin{axis}[ 
xmin=-0, 
xmax=6.0, 
ymin=0,
ymax=4.5,
%axis lines=middle,
axis y line=middle, 
axis x line=bottom,
axis line style={-, gray},
major tick length=\MajorTickLength,
every tick/.style={gray},
tick align=outside,
y axis line style={xshift=\AxisShift},
every y tick/.style={xshift=\AxisShift},
yticklabel style={xshift=\AxisShift},
x axis line style={yshift=\AxisShift},
every x tick/.style={yshift=\AxisShift},
x tick label style={yshift=\AxisShift, fill=white},
ylabel={\rotatebox{90}{$y(t)$}},
xlabel={$t$},
x label style={at={(ticklabel* cs:0.975)}, inner sep=5pt, yshift=4*\AxisShift, anchor=north},
y label style={at={(ticklabel* cs:0.55)}, inner sep=5pt, xshift=3.5*\AxisShift, anchor=east},
xtick={0,...,6},
ytick={1,...,4},
extra y ticks={0,4.5},
extra y tick labels={0,$K$},
grid=major, 
extra x ticks={2},
extra x tick labels={$T_0 + T_1$},
extra x tick style={major tick length=-5.25*\AxisShift, on background layer},
%enlarge y limits={abs=0.4,upper},
]
% Curve
\addplot [domain=1.5:6, samples=222, thick]{9/2 *(1 -exp(3 -2*x)};
% Line 1 as parametric plot
\addplot[dashed, variable=\t, samples=2, thick]({2},{\t}) ;       
% Line 2 as table-plot
\addplot[no marks, thick,] table[x=X,y=Y] {
X     Y
1.5   0
2     4.5
};
% Some special labels ===================
\begin{pgfonlayer}{foreground}
\foreach \T/\Name in {0.5/T_1, 1.5/T_0, 3.85/T_a, 4.6/T_b} {\edef\temp{\noexpand
\draw [gray, yshift=\AxisShift] (\T,0) -- (\T,2.75*\AxisShift) node[below, text=black, inner sep=2pt]{$\Name$}; 
}\temp}
\end{pgfonlayer}
% ================================
\end{axis}   

\end{tikzpicture}
\end{document}
cis
  • 8,073
  • 1
  • 16
  • 45
2

This does something of that sort. In order for one additional tick not wiping out the previous one, you need to the /.list key. In order to avoid overlaps, you can use shifts. So altogether

major also at/.list={ 1.5 as    $T_0$,
           2 as [node style={yshift=-2em}] $T_0 + T_1$},

Full MWE:

\documentclass{scrartcl}
\usepackage{tikz}
\usetikzlibrary{datavisualization.formats.functions}

\begin{document}

  \begin{tikzpicture}
    \datavisualization[
       scientific axes={clean},
       all axes = grid,
       x axis = {
          min value = 0,
          ticks = {
               major also at/.list={ 1.5 as $T_0$,
               2 as [node style={yshift=-2em,alias=pft}] $T_0 + T_1$},
             stack
            },
          label = $t$
         },
       y axis = {
          min value = 0,
          ticks = {major = {also at = 4.5 as $K$}},
          label = $y(t)$
         },
       visualize as smooth line/.list = {
          curve,
          line 1,
          line 2
         },
       line 2 = {style = dashed}
      ]
    data[
         set = curve,
         format = function
        ] {
           var x : interval[1.5 : 6];
           func y = 9/2 * (1 - exp(3 - 2 * \value x));
          }
    data[set = line 1] {
                        x,   y
                        1.5, 0
                        2,   4.5
                       }
    data[set = line 2] {
                        x, y
                        2, 4.5
                        2, 0
                       };
   \draw[very thin] (pft.north) -- ++ (0,1em);
  \end{tikzpicture}

\end{document}

enter image description here

  • Hello @Schrödinger's cat! Thank you for your answer! Is it possible to have a pin to T_0+T_1? I tried it with 2 as [{tick text padding=2.5em}] $T_0 + T_1$. But the pin overpaints the number 2. The number 2 can be removed, if needed. The solution of cis is acceptable too, but with the datavisualization please. Thank your for your effort in advance! – Su-47 Jan 13 '20 at 09:41
  • @Su-47 OK, I added this pin. It is a bit manual, but you said that the proposal by cis is fine, so... –  Jan 13 '20 at 17:10
  • Hello @Schrödinger's cat! Thank you for the update! No, I meant the same result like suggested by cis, only with datavisualization. – Su-47 Jan 14 '20 at 14:50
  • @Su-47 Sorry, I do not understand what you mean. Please tell me explicitly what you would like to add. –  Jan 14 '20 at 14:53
  • Hello @Schrödinger's cat! Your updated answer is what I wished, thank you! Sorry for my bad english. – Su-47 Jan 14 '20 at 16:47
0

I discovered a possibility to solve this task with an on board method. [{tick text padding = 'some value some measurement'}] and clean ticks do the trick, see my MWE:

\documentclass{scrartcl}
\usepackage{tikz}
\usetikzlibrary{datavisualization.formats.functions}

\begin{document}


\begin{tikzpicture}
    \datavisualization[
                       scientific axes={clean},
                       all axes = grid,
                       x axis = {
                                 min value = 0,
                                 ticks = {
                                          major = {
                                                   also at = 1.5 as $T_0$,
                                                   also at = 2 as [{tick text padding = 1.5em}] $T_0 + T_1$ % <--- changed !!!!!
                                                  },
                                          %stack <--- removed !!!!!
                                         },
                                 label = $t$
                                },
                       y axis = {
                                 min value = 0,
                                 ticks = {major = {also at = 4.5 as $K$}},
                                 label = $y(t)$
                                },
                       clean ticks, % <--- added !!!!!
                       visualize as smooth line/.list = {
                                                         curve,
                                                         line 1,
                                                         line 2
                                                        },
                       line 2 = {style = dashed}
                      ]
    data[
         set = curve,
         format = function
        ] {
           var x : interval[1.5 : 6];
           func y = 9/2 * (1 - exp(3 - 2 * \value x));
          }
    data[set = line 1] {
                        x,   y
                        1.5, 0
                        2,   4.5
                       }
    data[set = line 2] {
                        x, y
                        2, 4.5
                        2, 0
                       };
  \end{tikzpicture}


\end{document}

and the result

result

Su-47
  • 2,508