LaTeX2Python (tex2py)

Tex2py converts LaTeX into a Python parse tree, using TexSoup. This allows you to navigate latex files as trees, using either the default or a custom hierarchy. See md2py for a markdown parse tree.

Note tex2py currently only supports Python3.

created by Alvin Wan

Installation

Install via pip.

pip install tex2py

Usage

LaTeX2Python offers only one function tex2py, which generates a Python parse tree from Latex. This object is a navigable, "Tree of Contents" abstraction for the latex file.

Take, for example, the following latex file. (See pdf)

chikin.tex

\documentclass[a4paper]{article}
\begin{document}

\section{Chikin Tales}

\subsection{Chikin Fly}

Chickens don't fly. They do only the following:

\begin{itemize}
\item waddle
\item plop
\end{itemize}

\section{Chikin Scream}

\subsection{Plopping}

Plopping involves three steps:

\begin{enumerate}
\item squawk
\item plop
\item repeat, unless ordered to squat
\end{enumerate}

\subsection{I Scream}

\end{document}

Akin to a navigation bar, the TreeOfContents object allows you to expand a latex file one level at a time. Running tex2py on the above latex file will generate a tree, abstracting the below structure.

          <Document>
          /        \
  Chikin Tales   Chikin Scream
      /            /     \
 Chikin Fly  Plopping   I Scream

At the global level, we can access the title.

>>> from tex2py import tex2py
>>> with open('chikin.tex') as f: data = f.read()
>>> toc = tex2py(data)
>>> toc.section
Chikin Tales
>>> str(toc.section)
'Chikin Tales'

Notice that at this level, there are no subsections.

>>> list(toc.subsections)
[]

The main section has two subsections beneath it. We can access both.

>>> list(toc.section.subsections)
[Chikin Fly, Chikin Scream]
>>> toc.section.subsection
Chikin Fly

The TreeOfContents class also has a few more conveniences defined. Among them is support for indexing. To access the ith child of an <element> - instead of <element>.branches[i] - use <element>[i].

See below for example usage.

>>> toc.section.branches[0] == toc.section[0] == toc.section.subsection
True
>>> list(toc.section.subsections)[1] == toc.section[1]
True
>>> toc.section[1]
Chikin Scream

You can now print the document tree. (There is some weirdness with branches beyond titles, so for only titles, we have the following:

           ┌Chikin Tales┐
           │            └Chikin Fly
 [document]┤
           │             ┌Plopping
           └Chikin Scream┤
                         │        
                         │        
                         └I Scream

Additional Notes

• Behind the scenes, tex2py uses TexSoup. All tex2py objects have a source attribute containing a TexSoup object.