HTML to LaTeX translator [ad hoc]

#!/bin/gawk -E

{
    parsed = $0

    if( $0 ~ /<section .*>/) {
        if( $0 ~ /<section [a-zA-Z ]+>/) {
            parsed = gensub(/<section ([a-zA-Z ]+)>/,"\\\\chapter*{\\1}","g",parsed);
        }
        if( $0 ~ /<section [0-9]+ &ndash; .*>/) {
            parsed = gensub(/<section [0-9]+ \&ndash; (.*)>/,"\\\\chapter{\\1}","g",parsed);
        }
        if( $0 ~ /<section [0-9]+.[0-9]+ &ndash; .*>/) {
            parsed = gensub(/<section [0-9]+.[0-9]+ \&ndash; (.*)>/,"\\\\section{\\1}","g",parsed);
        }
        if( $0 ~ /<section [0-9]+.[0-9]+.[0-9]+ &ndash; .*>/) {
            parsed = gensub(/<section [0-9]+.[0-9]+.[0-9]+ \&ndash; (.*)>/,"\\\\subsection{\\1}","g",parsed);
        }
    }

    # parsed = gensub(/(<code>[^<>]*)\\\\([^<>]*<\/code>)/, "\\1\\\\\\\\\\\\\\2", "g", parsed);
    # parsed = gensub(/(<code>[^<>]*)\\([^<>\\]*<\/code>)/, "\\1\\\\\\\\\\2", "g", parsed);
    # parsed = gensub(/(<code>[^<>]*)\}([^<>]*<\/code>)/, "\\1\\\\}\\2", "g", parsed);
    # parsed = gensub(/(<code>[^<>]*)\{([^<>]*<\/code>)/, "\\1\\\\{\\2", "g", parsed);
    # parsed = gensub(/(<code>[^<>]*)\#([^<>]*<\/code>)/, "\\1\\\\#\\2", "g", parsed);
    parsed = gensub(/<code>([^<>]*)<em>([^<>]*)<\/em>([^<>]*)<\/code>/, "<code>\\1\\2\\3</code>", "g", parsed);
    parsed = gensub(/<code>/, "\\\\lstinline|", "g", parsed);
    parsed = gensub(/<\/code>/, "|", "g", parsed);

    parsed = gensub(/<p>/                   , "\\\\par "                                               , "g", parsed);
    parsed = gensub(/<B>/                   , "\\\\textbf{"                                            , "g", parsed);
    parsed = gensub(/<\/B>/                 , "}"                                                      , "g", parsed);
    parsed = gensub(/<b>/                   , "\\\\textbf{"                                            , "g", parsed);
    parsed = gensub(/<\/b>/                 , "}"                                                      , "g", parsed);
    parsed = gensub(/<a [^>]*>/             , ""                                                       , "g", parsed);
    parsed = gensub(/<\/a>/                 , ""                                                       , "g", parsed);
    parsed = gensub(/<table .*>/            , "\\\\begin{table}[H]\\\\centering\\\\begin{tabular}{ccc}", "g", parsed);
    parsed = gensub(/<\/table>/             , "\\\\end{tabular}\\\\end{table}"                         , "g", parsed);
    parsed = gensub(/<td>/                  , ""                                                       , "g", parsed);
    parsed = gensub(/<\/td>/                , " \\& "                                                  , "g", parsed);
    parsed = gensub(/<tr>/                  , ""                                                       , "g", parsed);
    parsed = gensub(/<\/tr>/                , "\\\\\\\\"                                               , "g", parsed);
    parsed = gensub(/<ul>/                  , "\\\\begin{itemize}"                                     , "g", parsed);
    parsed = gensub(/<\/ul>/                , "\\\\end{itemize}"                                       , "g", parsed);
    parsed = gensub(/<li>/                  , "\\\\item "                                              , "g", parsed);
    parsed = gensub(/<font colors?="(\w+)">/, "\\\\textcolor{\\1}{"                                    , "g", parsed);
    parsed = gensub(/<\/font>/              , "}"                                                      , "g", parsed);
    parsed = gensub(/<pre>/                 , "\\\\begin{lstlisting}"                                  , "g", parsed);
    parsed = gensub(/<\/pre>/               , "\\\\end{lstlisting}"                                    , "g", parsed);

    if( $0 ~ /.*<em>.*<SUB>.*<\/em>.*/) {
        parsed = gensub(/<em>(.*<SUB>.*)<\/em>/,"$ \\1 $","g",parsed);
        parsed = gensub(/<SUB>([^<>]+)<\/SUB>/,"_{\\1}","g",parsed);
    }
    if( $0 ~ /.*<em>[a-zA-Z ]{3,}<\/em>.*/) {
        parsed = gensub(/<em>([\w ]{3,})<\/em>/, "\\\\emph{\\1}", "g", parsed);
    }
    if( $0 ~ /.*<em>[a-zA-Z]{1,2}<\/em>.*/  ) {
        parsed = gensub(/<em>(\w{1,2})<\/em>/,"$ \\1 $","g",parsed);
    }
    if( $0 ~ /.*<em>([^<>]*[0-9=\#\&][^<>]*)<\/em>.*/ ) {
        parsed = gensub(/<em>([^<>]*[0-9=\#\&][^<>]*)<\/em>/,"$ \\1 $","g",parsed);
    }
    parsed = gensub(/<em>/, "\\\\emph{", "g", parsed);
    parsed = gensub(/<\/em>/, "}", "g", parsed);


    parsed = gensub(/\&acute;/, "'", "g", parsed);
    parsed = gensub(/\&aacute;/, "á", "g", parsed);
    parsed = gensub(/\&eacute;/, "é", "g", parsed);
    parsed = gensub(/\&iacute;/, "í", "g", parsed);
    parsed = gensub(/\&ccedil;/, "ç", "g", parsed);
    parsed = gensub(/\&Ccedil;/, "Ç", "g", parsed);
    parsed = gensub(/\&atilde;/, "ã", "g", parsed);
    parsed = gensub(/\&Atilde;/, "Ã", "g", parsed);
    parsed = gensub(/\&ecirc;/, "ê", "g", parsed);
    parsed = gensub(/\&lt;/, "<", "g", parsed);
    parsed = gensub(/\&amp;/, "\\\\\\&", "g", parsed);
    parsed = gensub(/\&#060;=/, "\\\\leq", "g", parsed);
    parsed = gensub(/%/, "\\\\%", "g", parsed);

    printf("%s\n",parsed);
}