Graphics, Figures & Tables ⇒ tikZ | Drawing Circles by new Commands

[kasa]

What is the problem with this source? The result is far to be nice.

Code: Select all

\documentclass{article}
\usepackage{tikz}

\newcommand{\Andr}[4]{  
% #1 n
% #2 angle
% #3 node size
% #4 \fill or \draw
\begin{tikzpicture}[rotate=#2,scale=2]
 \pgfmathsetmacro{\n}{3*#1 - 1}
 \pgfmathsetmacro{\m}{2*#1 - 1}
 \foreach \i in {1,...,\n}
 { \path (360/\n*\i:1cm) node (X\i) {};
    #4 (X\i) circle (#3pt);
 }
  \foreach \i in {1,...,\n}
 {
  \foreach \j in {#1,...,\m}
  {   \pgfmathsetmacro{\p}{1+ mod(\i+\j-1,\n)}
      \draw (X\i) -- (X\p);
   }
  }  
\end{tikzpicture}
}
  
\begin{document}
   \Andr{4}{0}{3}{\draw} 
   \Andr{6}{0}{2}{\draw}
   \Andr{2}{0}{3}{\fill}
   \Andr{3}{0}{2}{\fill}
\end{document}

[cgnieder]

Hi kasa,

welcome to the LaTeX community!

Your problem description is not very clear but I guess I know what your problem is. If I'm right it is due to the fact that node(X\i) {} not only stores a coordinate but actually creates a non-empty node, i.e., not a single point! If you change this to coordinate (X\i) the output looks like it could be what you want.

Regards

[kasa]

You are right. Thanks.
But the following example works nice even with node. Why?
(The edges link nodes, and not the centers of the circles, as in the case of coordinates. And in my first example the nodes and edges are distorsioned, the edges link different points on and in the circles. Why?)

Code: Select all

\documentclass{article}
\usepackage{tikz}

\newcommand{\Andr}[4]{  
% #1 n
% #2 angle
% #3 node size
% #4 \fill or \draw
\begin{tikzpicture}[rotate=#2, scale=2]
 \pgfmathsetmacro{\n}{3*#1 - 1}
 \pgfmathsetmacro{\m}{2*#1 - 1}
 \foreach \i in {1,...,\n}
 {
    \path (360/\n*\i:1cm)  node (X\i) {};
    #4 (X\i) circle (#3pt);
 }
  \foreach \i in {1,...,\n}
 {
   \foreach \j in {#1,...,\m}
   { 
          \draw (X\i) -- (X\j);   
   }
}  
\end{tikzpicture}
}
  
\begin{document}
\Andr{2}{90}{2}{\draw}\qquad\Andr{4}{90}{2}{\draw}

\Andr{6}{90}{2}{\draw}\qquad\Andr{3}{90}{2}{\fill}
\end{document}

[cgnieder]

Huh, that's strange... judging the code I can see no apparent reasons why the two macros give different outcomes. Maybe someone else can help here. (I'll try to investigate anyway when I find the time.)

Regards

[kasa]

Zap! I found the solution!! If I put
\pgfmathtruncatemacro{\p}{1+mod(\i+\j-1,\n)}
instead of
\pgfmathsetmacro{\p}{1+ mod(\i+\j-1,\n)}
the outcome will be nice.