这是我在这里的第一个问题,因此我提前为所犯的所有错误道歉,但这是我的 MWE,本质上是:
\documentclass[10pt,a4paper]{article}
\usepackage{fontspec}
\usepackage{lmodern}
\usepackage[safe]{tipa}
\usepackage{hyperref}
\usepackage{graphicx}
\usepackage[english,ngerman]{babel}
\usepackage[linguistics]{forest}
\usepackage{avm}
\usepackage{rotating}
\usepackage{pdflscape}
\begin{document}
\begin{sidewaysfigure}[htbp]
\centering
\scalebox{.8}{
\begin{forest}for tree={l=2cm}
[\emph{realisation-rule}
[MORPHOTACTICS,draw,calign=fixed edge angles,calign secondary angle=0
[{\begin{avm}
\[
mud & \{\normalfont\textit{agr} \}\\
ms & \{ \[\asort{pid}
cat & verb\], ...\}\\
mph & \< \[pc & $-4$\]\> \]
\end{avm}}, name= 1, tier=word]
[{\begin{avm}
\[mud & \{\normalfont\textit{agr}\}\\
ms & \{ \[\asort{pid}
cat & \[\asort{adj}
type & B\]\], ...\}\\
mph & \<\[pc & $-2$\], \[pc & $-1$\]\>\]
\end{avm}}, l=7cm]
[{\begin{avm}
\[mud & \{\normalfont\textit{agr}\}\\
ms & \{ \[\asort{pid}
cat & \[\asort{adj}
type & A\]\], ...\}\\
mph & \< \[pc & -1\]\>\]
\end{avm}}, name=2, tier=word]
]
[EXPONENCE,draw
[QUAL,draw, name=qual
[{\begin{avm}
\[mud & \{ \[\normalfont\textit{agr}\\cl & 7\]\}\\
mph & \< ... \[ph & \<\normalfont ca\>\\
pc & $-1$ $\vee$ $-2$
\] ... \>\]
\end{avm}}, tier=word]
[\dots]
]
[CONC,draw, name=conc
[{\begin{avm}
\[mud & \{ \[\normalfont\textit{agr}\\cl & 7\]\}\\
mph & \< ... \[ph & \<\normalfont ci\>\\
pc & $-1$ $\vee$ $-4$
\] ... \>\]
\end{avm}}, tier=word]
[\dots]
]
]
]
\draw (1.north) to (qual.south);
\draw (2.north) to (conc.south);
\end{forest}
}
\caption{Nyanja pre-prefixation}\label{fig:Nyanja}
\end{sidewaysfigure}
\end{document}
由于树太宽,具有 l=7 选项的节点位于其左右兄弟节点的下方。这就是为什么森林算法认为没有必要将其置于母节点正下方的中心,而这正是我想要的,因为现在,它的边缘与另一个节点相交。而且它看起来也不好看。
也许这张现在看起来的图片用红色画出了预期的位置可以更好地解释这一点:
答案1
或许:
\documentclass[10pt,a4paper]{article}
\usepackage{fontspec}
\usepackage{lmodern}
\usepackage[safe]{tipa}
\usepackage{hyperref}
\usepackage{graphicx}
\usepackage[english,ngerman]{babel}
\usepackage[linguistics]{forest}
\usepackage{avm}% Copied from https://nlp.stanford.edu/~manning/tex/avm.sty
\usepackage{rotating}
\usepackage{pdflscape}
\begin{document}
\begin{sidewaysfigure}[htbp]
\centering
\scalebox{.8}{
\begin{forest}for tree={l=4cm}
[\emph{realisation-rule}
[MORPHOTACTICS,draw,calign=fixed edge angles,calign secondary angle=0,l=3cm
[{\begin{avm}
\[
mud & \{\normalfont\textit{agr} \}\\
ms & \{ \[\asort{pid}
cat & verb\], ...\}\\
mph & \< \[pc & $-4$\]\> \]
\end{avm}}, name= 1, tier=word]
[{\begin{avm}
\[mud & \{\normalfont\textit{agr}\}\\
ms & \{ \[\asort{pid}
cat & \[\asort{adj}
type & B\]\], ...\}\\
mph & \<\[pc & $-2$\], \[pc & $-1$\]\>\]
\end{avm}}, l=2cm]
[{\begin{avm}
\[mud & \{\normalfont\textit{agr}\}\\
ms & \{ \[\asort{pid}
cat & \[\asort{adj}
type & A\]\], ...\}\\
mph & \< \[pc & -1\]\>\]
\end{avm}}, name=2, tier=word]
]
[EXPONENCE,draw,l=3.3cm
[QUAL,draw, name=qual,
[{\begin{avm}
\[mud & \{ \[\normalfont\textit{agr}\\cl & 7\]\}\\
mph & \< ... \[ph & \<\normalfont ca\>\\
pc & $-1$ $\vee$ $-2$
\] ... \>\]
\end{avm}}, tier=word]
[\dots]
]
[CONC,draw, name=conc
[{\begin{avm}
\[mud & \{ \[\normalfont\textit{agr}\\cl & 7\]\}\\
mph & \< ... \[ph & \<\normalfont ci\>\\
pc & $-1$ $\vee$ $-4$
\] ... \>\]
\end{avm}}, tier=word]
[\dots]
]
]
]
\draw (1.north) to (qual.south);
\draw (2.north) to (conc.south);
\end{forest}
}
\caption{Nyanja pre-prefixation}\label{fig:Nyanja}
\end{sidewaysfigure}
\end{document}
但使用\scalebox
只是一种丑陋的解决方法。恕我直言,应该考虑使用不同类型的树进行彻底的重新设计。