Using pythontex inside itemize

Question

\newcommand{\test}[2]{%
    \begin{itemize}
    \pyc{strs = u"#1".split(u"#2");}
    \pyc{for s in strs: print("\item " + s);}
    \end{itemize}
}

When using this command, I get the folowing output:

perhaps missing \item.

MWE:

\documentclass[a5paper]{extbook}

\usepackage{geometry}
\usepackage{polyglossia}
\usepackage{fontspec}

\usepackage[pyfuture=all]{pythontex}

\setmainlanguage{russian}
\setotherlanguage{english}
\defaultfontfeatures{Ligatures=TeX,Mapping=tex-text}
\setmainfont{Times New Roman}


\newcommand{\test}[2]{%
    \begin{itemize}
    \pyc{strs = u"#1".split(u"#2");}
    \pyc{for s in strs: print("\item " + s);}
    \end{itemize}
}

\begin{document}

\test{test1, test2, test3, тест4}{,}

\end{document}

Another option would be to print the \begin{itemize} and \end{itemize} from Python as well, so that the entire command contents comes from Python, and you don't end up with only parts of it existing. — G. Poore, Mar 25 '17 at 15:15

score 1 · Accepted Answer · answered Mar 24 '17 at 13:59

The problem is that, when the file is compiled before running PythonTeX, the itemize environment is empty, so \@noitemerr is executed.

Define a pyitemize environment that disables it.

\documentclass[a5paper]{extbook}

\usepackage{geometry}
\usepackage{polyglossia}
\usepackage{fontspec}

\usepackage[pyfuture=all]{pythontex}

\setmainlanguage{russian}
\setotherlanguage{english}
\setmainfont{Times New Roman}

\makeatletter
\newenvironment{pyitemize}
 {\def\@noitemerr{}\itemize}
 {\enditemize}
\makeatother

\newcommand{\test}[2]{%
    \begin{pyitemize}
    \pyc{strs = u"#1".split(u"#2");}
    \pyc{for s in strs: print("\item " + s);}
    \end{pyitemize}
}

\begin{document}

Some text before

\test{test1, test2, test3, тест4}{,}

\end{document}

You can do the same without PythonTeX.

\documentclass[a5paper]{extbook}

\usepackage{geometry}
\usepackage{polyglossia}
\usepackage{fontspec}
\usepackage{xparse}

\setmainlanguage{russian}
\setotherlanguage{english}
\setmainfont{Times New Roman}

\ExplSyntaxOn
\NewDocumentCommand{\test}{mm}
 {
  \begin{itemize}
  \seq_set_split:Nnn \l_tmpa_seq { #2 } { #1 }
  \seq_map_inline:Nn \l_tmpa_seq { \item ##1 }
  \end{itemize}
 }
\ExplSyntaxOff

\begin{document}

Some text before

\test{test1, test2, test3, тест4}{,}

\end{document}

By the way, the line

\defaultfontfeatures{Ligatures=TeX,Mapping=tex-text}

is redundant: it does twice a setting that's already on by default.

I know it's possible without pythontex, even without xparse with pure LaTeX, like this:
\def\menu#1{% \begin{itemize} \gdef\firstelement{1} \foreach \e in {#1}{% \ifnum\firstelement=0\fi \item \e% \gdef\firstelement{0}% } \end{itemize} }

Thing is, I know python, and I'm not that good at LaTeX macros, it's easier to debug and produce python code for me. So I'm trying to replace simple macros with python scripts, and then go futher with more complex examples. — Alexander Biryukov, Mar 24 '17 at 14:18
@Sanya_rnd No problem at all! I just suggested a possibly simpler approach. — egreg, Mar 24 '17 at 14:22

Using pythontex inside itemize

1 Answers1