0

I want to control the size of a piece of text in my Latex document (for an application for research funding), and ignore the latex code. I want to count also the whitespaces. I really want to measure the size of the text generated in the PDF. I coded this macro getting inspiration from here.

\documentclass{article}
\usepackage{stringstrings}

\newcommand{\numchars}[1]{%
   \convertchar[q]{#1}{ }{}
   {\noindent 
    \textit{(\stringlength{\thestring} signs)}

    #1}}


\begin{document}


\numchars{Longtemps, je me suis couché de bonne heure. Parfois, à peine ma bougie éteinte, mes yeux se fermaient si vite que je n’avais pas le temps de me dire : " Je m’endors. " \label{À la recherche du temps perdu}}


\end{document}

I have three issues here:

  1. Exotic characters like "é" count as several characters
  2. It looks like beyond 500 characters the code fails
  3. Any Latex code, any citation for instance, will generate an error
the world is not flat
  • 571
  • 1
    Try to convert your test file to a regular text with pandoc -t asciidoc test.tex. You could then pipe the output to wc -c. If you want to exclude new line characters insert sed -z "s:\n::g" in the pipe. If you also want to exclude spaces, add \s to the pattern. for instance: pandoc -t asciidoc test.tex | sed -z "s:\n\|\s::g" | wc -c. – Celdor Feb 15 '24 at 13:15
  • 1
    There are approaches that use outside software. If you wanted to stay within LaTeX to perform the operation, https://tex.stackexchange.com/questions/44618/dynamically-count-and-return-number-of-words-in-a-section/587567#587567 provides an approach. By the way, the problem you are looking at is far too complex for stringstrings (which I wrote). Don't use it. – Steven B. Segletes Feb 15 '24 at 19:48

0 Answers0