MoinMoin: attachment:latex2wiki.py of ReimarBauer

   1 """Translate a subset of LaTeX into MoinMoin wiki syntax.
   2 
   3 This program was originally written by Maxime Biais <maxime@biais.org>
   4 and then modified by Allen Downey <downey@allendowney.com>.  The
   5 original is available at http://wiki.loria.fr/wiki/Latex2wiki
   6 
   7 The primary limitation of this program is that it only recognizes
   8 Latex patterns if they appear on a single line, so {\tt this pattern}
   9 would get translated, but {\tt this
  10 pattern} would not.
  11 
  12 Other limitations include:
  13 
  14 1) It doesn't distinguish between itemize, enumerate and
  15 description; everything becomes itemize.
  16 
  17 2) It only recognizes the subset of Latex I use.
  18 
  19 3) It's not particularly efficient, but for the files I have
  20    translated, it doesn't matter.
  21 
  22 """
  23 
  24 #    Copyright (C) 2003, Maxime Biais <maxime@biais.org>
  25 #    Modified version Copyright 2005 Allen B. Downey
  26 #    Modified version for my latex text  2008 Reimar Bauer   
  27 #
  28 #    This program is free software; you can redistribute it and/or modify
  29 #    it under the terms of the GNU General Public License as published by
  30 #    the Free Software Foundation; either version 2 of the License, or
  31 #    (at your option) any later version.
  32 #
  33 #    This program is distributed in the hope that it will be useful,
  34 #    but WITHOUT ANY WARRANTY; without even the implied warranty of
  35 #    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  36 #    GNU General Public License for more details.
  37 #
  38 #    You should have received a copy of the GNU General Public License
  39 #    along with this program; if not, write to the Free Software
  40 #    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  41 #
  42 
  43 import sys, re
  44 
  45 # rewrites is a list of rewrite rules.
  46 # The first entry in each line is a regular expression.
  47 # The second line is the format string
  48 # used to rewrite the line if the regexp matches.
  49 
  50 rewrites = [
  51     # comments
  52     (r"%.*", ""),
  53     # simple Latex commands we ignore
  54     (r"\\[^ {]*$", ""),
  55     # other Latex commands we ignore
  56     (r"\\label{.*}", ""),
  57     (r"\\index{.*}", ""),
  58     (r"\\documentclass{.*}", ""),
  59     (r"\\usepackage{.*}", ""),
  60     (r"\\newcommand{.*}", ""),
  61     (r"\\setlength{.*}", ""),
  62     # some characters that have to be quoted in Latex
  63     # that don't have to be quoted in wiki
  64     (r"(.*)\\#(.*)", "%s#%s"),
  65     (r"(.*)\\\$(.*)", "%s$%s"),
  66     (r"(.*)\\&(.*)", "%s&%s"),
  67     (r"(.*)\\_(.*)", "%s_%s"),
  68     # text formats
  69     (r"(.*)\\emph{([^}]*)}(.*)", "%s'''%s'''%s"),
  70     (r"(.*){\\sf ([^}]*)}(.*)", "%s`%s`%s"),
  71     (r"(.*){\\bf ([^}]*)}(.*)", "%s'''%s'''%s"),
  72     (r"(.*){\\em ([^}]*)}(.*)", "%s''%s''%s"),
  73     # urls (a url at the beginning of a line gets a special rule
  74     (r"^\\url{([^}]*)}(.*)", "%s %s"),
  75     (r"(.*)\\url{([^}]*)}(.*)", "%s %s %s"),
  76     (r"(.*){\\tt ([^}]*)}(.*)", "%s{{{%s}}}%s"),
  77     (r"(.*)\\texttt{([^}]*)}(.*)", "%s{{{%s}}}%s"),
  78     (r"(.*)\\textit{([^}]*)}(.*)", "%s''%s''%s"),
  79     (r"(.*)\\textbf{([^}]*)}(.*)", "%s'''%s'''%s"),
  80     
  81     (r"(.*)\\colorbox{silver}{([^}]*)}(.*)", "%s{{{%s}}}%s"),
  82     (r"(.*)\\colorbox{lightgray}{\\parbox{14.4cm}{(.*)", "%s{{{{{#!wiki blue %s"),
  83     #(r"(.*)\\includegraphics([^\[]*)([^\[]*)", "%s{{attachment:%s}} %s"),
  84     (r"(.*)\\includegraphics\[([$\]]*)([^\[]*)", "%s{{attachment:%s}} %s"),
  85     # arrow is a special command that appears in some of my files
  86     (r"(.*)\\arrow (.*)", "%s-->%s"),
  87     # footnotes get translated into parenthetical comments
  88     # (which some people think is a preferable style anyway)
  89     (r"(.*)\\footnote{(.*)}(.*)", "%s (footnote: %s) %s"),
  90     # all items become bulleted items (no unemerations)
  91     (r"\\item (.*)", " * %s"),
  92     # verbatim becomes code display
  93     (r"\\begin{verbatim}", "{{{{{"),
  94     (r"\\end{verbatim}", "}}}}}"),
  95     # listings becomes python code display
  96     (r"\\begin{lstlisting}(.*)", "{{{{{#!python %s" ),
  97     (r"\\end{lstlisting}", "}}}}}"),
  98     # all other latex environments are ignored
  99     (r"\\begin{.*}", ""),
 100     (r"\\end{.*}", ""),
 101     # turn Latex quotation marks into plain old quotation marks
 102     (r"(.*)``(.*)''(.*)", "%s\"%s\"%s"),
 103     # headings and title page entries
 104     (r"\\paragraph{(.*)}", "==== %s ===="),
 105     (r"\\subsubsection{(.*)}", "==== %s ===="),
 106     (r"\\subsection{(.*)}", "=== %s ==="),
 107     (r"\\section{(.*)}", "== %s =="),
 108     (r"\\chapter{(.*)}", "= Chapter %s ="),
 109     (r"\\title{(.*)}", "= %s ="),
 110     (r"\\author{(.*)}", "%s"),
 111     (r"\\date{(.*)}", "%s"),
 112     (r"\\caption{(.*)}", "(caption: %s)"),
 113    
 114     ]
 115 
 116 
 117 def apply_rewrites(line, rewrites):
 118     """apply each rewrite rule repeatedly and return the result"""
 119     for reg, format in rewrites:
 120         # keep applying each rule until it doesn't match.
 121         # compiling the reg would probably improve performance.
 122         while True:
 123             m = re.search(reg, line)
 124             if not m: break
 125             line = format % m.groups()
 126     return line
 127 
 128 
 129 def main(script, infile, outfile=None):
 130     in_stream = open(infile)
 131     if outfile:
 132         out_stream = open(outfile, 'w')
 133     else:
 134         out_stream = sys.stdout
 135 
 136     for line in in_stream:
 137         line = line.rstrip()
 138         line = apply_rewrites(line, rewrites)
 139         print >> out_stream, line
 140 
 141 if __name__ == '__main__':
 142     main(*sys.argv)