Use a foreach statement to simplify TikZ drawing

Question

Below, the two diagrams should be the same, but actually they are not!

\documentclass[margin=2pt]{standalone}
\usepackage{tikz}
\usetikzlibrary{positioning,fit}
\newcommand{\PL}[3][]{
    \node[thick,circle,draw,minimum size=0.5cm,inner sep=0,outer sep=0,label=left:#3,#1](#2){};
    \draw[thick,#1] (#2.south west) -- (#2.north east);
}
\begin{document}
\begin{tikzpicture}
    \PL{L1}{1}
    \PL[below=0.1 of L1]{L2}{2}
    \PL[below=0.1 of L2]{L3}{3}
    \PL[below=0.1 of L3]{L4}{4}
    \PL[below=0.1 of L4]{L5}{5}
    \PL[below=0.1 of L5]{L6}{6}
    \PL[below=0.1 of L6]{L7}{7}
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\begin{tikzpicture}
    \PL{L1}{1}
    \foreach \i [evaluate=\i as \j using \i - 1] in {2,...,7} {
        \PL[below=0.1cm of L\j]{L\i}{\i}
    }
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\end{document}

How do I fix the \foreach statement diagram?

Answer 1

If you do not evaluate \j to an integer, you get numbers like 1.0, in which .0 is interpreted as the east anchor. So all I did was to replace [evaluate=\i as \j using \i - 1] by [evaluate=\i as \j using {int(\i - 1)}] to get

\documentclass[margin=2pt]{standalone}
\usepackage{tikz}
\usetikzlibrary{positioning,fit}
\newcommand{\PL}[3][]{
    \node[thick,circle,draw,minimum size=0.5cm,inner sep=0,outer sep=0,label=left:#3,#1](#2){};
    \draw[thick,#1] (#2.south west) -- (#2.north east);
}
\begin{document}
    \begin{tikzpicture}
    \PL{L1}{1}
    \PL[below=0.1 of L1]{L2}{2}
    \PL[below=0.1 of L2]{L3}{3}
    \PL[below=0.1 of L3]{L4}{C}
    \PL[below=0.1 of L4]{L5}{R}
    \PL[below=0.1 of L5]{L6}{RC}
    \PL[below=0.1 of L6]{L7}{RH}
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\begin{tikzpicture}
    \PL{L1}{1}
    \foreach \i [evaluate=\i as \j using {int(\i - 1)}] in {2,...,7} {
        \PL[below=0.1cm of L\j]{L\i}{\i}
    }
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\end{document}

ADDENDUM: Since @AndréC added an answer which is IMHO not really to the point of the original question, I add something that is to the point of the original question as well as some sort of a response to AndréC.

1. You can avoid all this by using remember.... And you can, of course, attach the labels of your left diagram, just declare \i to be the count.
2. Using a path picture to strike out a node is one way, what you are doing is IMHO at least equally good, and one can also use strike out that comes with shapes.misc. None of this, however, is IMHO at the heart of the question.

So here is an example (but if I was you I would keep your way of striking the node out, probably using a path picture is more elegant than strike through, though).

\documentclass[tikz,border=3.14mm]{standalone}
\usetikzlibrary{positioning,fit,shapes.misc}
\newcommand{\PL}[3][]{
    \node[thick,strike out,
    draw,minimum size=0.35cm,inner sep=0,outer sep=0,#1](#2-inner){};
    \node[draw,circle,inner sep=0,fit=(#2-inner),outer sep=1pt,label=left:#3](#2){};
}
\begin{document}
\begin{tikzpicture}
    \PL{L1}{1}
    \foreach \X [count=\i starting from 2,remember=\i as \j (initially 1)] 
    in {2,3,C,R,RC,RH} {
        \PL[below=0.1cm of L\j]{L\i}{\X}
    }
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\end{document}

Answer 2

It is not necessary to use a LaTeX command to trace nodes. All you have to do is declare tikz styles. To draw the diagonal bar, I used the notion of path picture bounding box described on page 173 of the 3.0.1a manual.

\documentclass[margin=5mm]{standalone}
\usepackage{tikz}
\usetikzlibrary{positioning,fit}
%\newcommand{\PL}[3][]{
%    \node[thick,circle,draw,minimum size=0.5cm,inner sep=0,outer sep=0,label=left:#3,#1](#2){};
%    \draw[thick,#1] (#2.south west) -- (#2.north east);
%}

\begin{document}
%    \begin{tikzpicture}
%    \PL{L1}{1}
%    \PL[below=0.1 of L1]{L2}{2}
%    \PL[below=0.1 of L2]{L3}{3}
%    \PL[below=0.1 of L3]{L4}{4}
%    \PL[below=0.1 of L4]{L5}{5}
%    \PL[below=0.1 of L5]{L6}{6}
%    \PL[below=0.1 of L6]{L7}{7}
%    \node[fit=(L1)(L7), draw] {};
%\end{tikzpicture}

\begin{tikzpicture}
    [slash/.style={
        draw,thick,circle,minimum size=.5cm,
        inner sep =0pt,outer sep=0pt,
        label={left:#1},
        path picture={
            \draw(path picture bounding box.south west)--(path picture bounding box.north east);}
    }]
    \node[slash=1](L1){};
    \foreach \i [evaluate=\i as \j using int(\i - 1)] in {2,...,7} {
        \node[below=0.1 of L\j,slash=\i](L\i){};
    }
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\end{document}

Update (to answer to marmot):

@marmot: I know for a fact that my answer is not the heart of the question. And that was not my point.

To simplify even more the loop:

\documentclass[margin=5mm]{standalone}
\usepackage{tikz}
\usetikzlibrary{positioning,fit}
\begin{document}

\begin{tikzpicture}
    [slash/.style={
        draw,thick,circle,minimum size=.5cm,
        inner sep =0pt,outer sep=0pt,
        label={left:#1},
        path picture={
            \draw(path picture bounding box.south west)--(path picture bounding box.north east);}
    }]
    \node[slash=1](L1){};
    \foreach \j [count=\i] in {2,...,7} {
        \node[below=0.1 of L\i,slash=\j](L\j){};
    }
    \node[fit=(L1)(L7), draw] {};
\end{tikzpicture}
\end{document}