
     1ˆÊÉƒ|M¹¿Ç $(    1""" TeXcheck.py -- rough syntax checking on Python style LaTeX documents.

   Written by Raymond D. Hettinger <python at rcn.com>
   Copyright (c) 2003 Python Software Foundation.  All rights reserved.

Designed to catch common markup errors including:
* Unbalanced or mismatched parenthesis, brackets, and braces.
* Unbalanced or mismatched \\begin and \\end blocks.
* Misspelled or invalid LaTeX commands.
* Use of forward slashes instead of backslashes for commands.
* Table line size mismatches.

Sample command line usage:
    python texcheck.py -k chapterheading -m lib/librandomtex *.tex

Options:
    -m          Munge parenthesis and brackets. [0,n) would normally mismatch.
    -k keyword: Keyword is a valid LaTeX command. Do not include the backslash.
    -d:         Delimiter check only (useful for non-LaTeX files).
    -h:         Help
    -s lineno:  Start at lineno (useful for skipping complex sections).
    -v:         Verbose.  Trace the matching of //begin and //end blocks.
"""

import re
import sys
import getopt
from itertools import izip, count, islice
import glob

cmdstr = r"""
    \section \module \declaremodule \modulesynopsis \moduleauthor
    \sectionauthor \versionadded \code \class \method \begin
    \optional \var \ref \end \subsection \lineiii \hline \label
    \indexii \textrm \ldots \keyword \stindex \index \item \note
    \withsubitem \ttindex \footnote \citetitle \samp \opindex
    \noindent \exception \strong \dfn \ctype \obindex \character
    \indexiii \function \bifuncindex \refmodule \refbimodindex
    \subsubsection \nodename \member \chapter \emph \ASCII \UNIX
    \regexp \program \production \token \productioncont \term
    \grammartoken \lineii \seemodule \file \EOF \documentclass
    \usepackage \title \input \maketitle \ifhtml \fi \url \Cpp
    \tableofcontents \kbd \programopt \envvar \refstmodindex
    \cfunction \constant \NULL \moreargs \cfuncline \cdata
    \textasciicircum \n \ABC \setindexsubitem \versionchanged
    \deprecated \seetext \newcommand \POSIX \pep \warning \rfc
    \verbatiminput \methodline \textgreater \seetitle \lineiv
    \funclineni \ulink \manpage \funcline \dataline \unspecified
    \textbackslash \mimetype \mailheader \seepep \textunderscore
    \longprogramopt \infinity \plusminus \shortversion \version
    \refmodindex \seerfc \makeindex \makemodindex \renewcommand
    \indexname \appendix \protect \indexiv \mbox \textasciitilde
    \platform \seeurl \leftmargin \labelwidth \localmoduletable
    \LaTeX \copyright \memberline \backslash \pi \centerline
    \caption \vspace \textwidth \menuselection \textless
    \makevar \csimplemacro \menuselection \bfcode \sub \release
    \email \kwindex \refexmodindex \filenq \e \menuselection
    \exindex \linev \newsgroup \verbatim \setshortversion
    \author \authoraddress \paragraph \subparagraph \cmemberline
    \textbar \C \seelink
"""

def matchclose(c_lineno, c_symbol, openers, pairmap):
    "Verify that closing delimiter matches most recent opening delimiter"
    try:
        o_lineno, o_symbol = openers.pop()
    except IndexError:
        print "\nDelimiter mismatch.  On line %d, encountered closing '%s' without corresponding open" % (c_lineno, c_symbol)
        return
    if o_symbol in pairmap.get(c_symbol, [c_symbol]): return
    print "\nOpener '%s' on line %d was not closed before encountering '%s' on line %d" % (o_symbol, o_lineno, c_symbol, c_lineno)
    return

def checkit(source, opts, morecmds=[]):
    """Check the LaTeX formatting in a sequence of lines.

    Opts is a mapping of options to option values if any:
        -m          munge parenthesis and brackets
        -d          delimiters only checking
        -v          verbose trace of delimiter matching
        -s lineno:  linenumber to start scan (default is 1).

    Morecmds is a sequence of LaTeX commands (without backslashes) that
    are to be considered valid in the scan.
    """

    texcmd = re.compile(r'\\[A-Za-z]+')
    falsetexcmd = re.compile(r'\/([A-Za-z]+)') # Mismarked with forward slash

    val