☰

在元组方程中嵌入简单的 tikz 图

tag-icon • tag-icon tikz-pgf math-mode equations
在元组方程中嵌入简单的 tikz 图

我想使用 tikz 绘制一棵简单的树，并将该树嵌入等式中：

\documentclass{article}
\usepackage[a4paper,margin=25mm,showframe]{geometry}
\usepackage{graphicx} % <--- For including graphics
\usepackage{listings} % <--- For typesetting verbatim code
\usepackage{tikz}     % <--- For drawing the diagram
\usepackage{array,amsmath}     % <--- For drawing the diagram

\usetikzlibrary{calc,positioning,shapes.geometric}

\begin{document}

\def\mytree{\tikz[inner sep=1pt] {
                  \node (T1) at (0,0) {$T_1$};
                  \node (T2) at (1,0) {$T_2$};
                  \node (INLJ) at (0.5,0.5) {\textsf{IndexJoin}};

                  \node (T3) at (1.5,0.5) {$T_3$};
                  \node (SMJ) at (1,1) {\small{\sf HashJoin}};

                  \node (T4) at (2,1) {$T_4$};
                  \node (HJ) at (1.5,1.5) {\sf HashJoin};

                  \draw (T1)--(INLJ)--(T2);
                  \draw (INLJ)--(SMJ)--(T3);
                  \draw (SMJ)--(HJ)--(T4);
}}

\[ \left( \mytree, 10.111  \right)\]

\end{document}

结果如下图所示： 在此处输入图片描述

你可以看到大量空白元组的两个元素之下。我该如何修复这个问题？

另一个问题：有没有办法让“[逗号] 10.111”部分更美观，比如通过垂直对齐它与 tikz 图片的中间？

==============

编辑

感谢 Phelype 的帮助，我现在能够减少空格。然而，出现了一个相关的问题。我想\def事先创建树的两个变体，然后使用它们——但 latex 给出了编译错误。如果我创建\def一个变体，立即使用它，然后\def使用另一个变体，则不会出现问题。为什么会这样？

代码正常（如果我同时使用这两棵树就会出错）：

\def \tree1 {\tikz[inner sep=1pt, baseline=(T3.north)] {
                  \node (T1) at (0,0) {$T_1$};
                  \node (T2) at (1,0) {$T_2$};
                  \node (INLJ) at (0.5,0.5) {\small{\sffamily IndexJoin}};

                  \node (T3) at (1.5,0.5) {$T_3$};
                  \node (SMJ) at (1,1) {\small{\sffamily HashJoin}};

                  \node (T4) at (2,1) {$T_4$};
                  \node (HJ) at (1.5,1.5) {\small{\sffamily HashJoin}};

                  \draw (T1)--(INLJ)--(T2);
                  \draw (INLJ)--(SMJ)--(T3);
                  \draw (SMJ)--(HJ)--(T4);}}

\[ \left[ \tree1 , \{T1, T2, T3, T4\}; 10.111  \right]\]

\def \tree2 {\tikz[inner sep=1pt, baseline=(T3.south)] {
                  \node (T1) at (0,0) {$T_1$};
                  \node (T2) at (1,0) {$T_2$};
                  \node (INLJ) at (0.5,0.5) {\small{\sffamily IndexJoin}};

                  \node (T3) at (1.5,0.5) {$T_3$};
                  \node (SMJ) at (1,1) {\small{\sffamily HashJoin}};

                  \draw (T1)--(INLJ)--(T2);
                  \draw (INLJ)--(SMJ)--(T3);}}

\[ \left[ \tree2 , \{T1, T2, T3, T4\}; 10.111  \right]\]

答案1

我会baseline=(current bounding box.center)用蒂克兹图片。（这并不重要，我也会使用tikzpicture环境而不是\tikz命令。）

除非我错过了什么，否则这两棵树几乎完全相同，只是第二棵树有两条额外的边。与其定义两个几乎相同的命令，为什么不创建一个\mytree和\mytree*变体呢？这很容易做到，使用\NewDocumentCommandfrom解析。这样做的话，输出结果如下：

在此处输入图片描述

代码如下：

\documentclass{article}
\usepackage[a4paper,margin=25mm,showframe]{geometry}
\usepackage{graphicx} % <--- For including graphics
\usepackage{listings} % <--- For typesetting verbatim code
\usepackage{tikz}     % <--- For drawing the diagram
\usepackage{array,amsmath}     % <--- For drawing the diagram

\usetikzlibrary{calc,positioning,shapes.geometric}

\usepackage{xparse}

\NewDocumentCommand\mytree{s}{
  \begin{tikzpicture}[inner sep=1pt, baseline=(current bounding box.center)]
      \node (T1) at (0,0) {$T_1$};
      \node (T2) at (1,0) {$T_2$};
      \node (INLJ) at (0.5,0.5) {\textsf{IndexJoin}};

      \node (T3) at (1.5,0.5) {$T_3$};
      \node (SMJ) at (1,1) {\small{\sf HashJoin}};

      \node (T4) at (2,1) {$T_4$};
      \node (HJ) at (1.5,1.5) {\sf HashJoin};

      \draw (T1)--(INLJ)--(T2);
      \draw (INLJ)--(SMJ)--(T3);
      \IfBooleanT{#1}{\draw (SMJ)--(HJ)--(T4);}
  \end{tikzpicture}
}

\begin{document}

\[ \left[ \mytree*, \{T1, T2, T3, T4\}; 10.111  \right]\]

\[ \left[ \mytree, \{T1, T2, T3, T4\}; 10.111  \right]\]

\end{document}

in表示该宏有一个可选的-参数；也就是说，{s}它可以用作或。在行的定义中\NewDocumentCommand\mytree{s}{...}\mytree*\mytree\mytree*\mytree

\IfBooleanT{#1}{...}

表示如果#1（存在且）是*则执行{...}。还有变体\IfBooleanTF和\IfBooleanF。参见解析文档以了解更多详细信息。

相关内容