#!/usr/bin/perl

#============================================================================
#
# NAME
#
#     cleanHtml.prl
#
#
# DESCRIPTION
#
#     Clean up the exported HTML files generated by Microsoft Word
#     or Mathematica.  It has been specialized to format simple text 
#     with equations.  Put all equations on lines by themselves.
#
# EXAMPLE
#
#       (1)  Edit equationExample.doc with Microsoft Word,
#
#       (2)  Select "File->Save As Web Page".  Select Web Page, Filtered.
#            Save as equationExample.htm
#            You will also get a directory of images called 
#            equationExample_files
#
#       (3)  From the command line, run
#                $ perl cleanHtml.prl
#
#            cleanHtml Version 1.3 -  Perl utility program which cleans up 
#            Microsoft Word
#            and Mathematica exported HTML containing mathematical equations.
#            Copyright (C) 2002-2008 by Sean Erik O'Connor.  All Rights 
#            Reserved.
#
#            cleanHtml comes with ABSOLUTELY NO WARRANTY; for details see the
#            GNU General Public License.  This is free software, and you are 
#            welcome to redistribute it under certain conditions; see the GNU 
#            General Public License for details.
#
#            Enter the parent directory containing the files.  
#            (RETURN = current directory).
#            =>
#            Enter the name of the .HTM file 
#            (You can leave off the .htm extension).
#            => equationExample
#            Enter the directory containing the .gif image files 
#            (RETURN = equationExample_files).
#            =>
#            Enter the new image file directory name (RETURN = WebPageImages). 
#            => equationExampleImages
#            Enter the starting image number (RETURN = 1).
#            =>
#            Enter the image numbering increment (RETURN = 2):
#            =>
#            Done.
#            Your HTML file is ./equationExample.html
#            Your renumbered equation .gif images are in equationExampleImages
#
#           (1)  Edit equationExample.nb with Mathematica
#           (2)  Select "File->Save As Special->HTML"
#                In "Save As Filename" use the name, equationExample.htm
#                You will get a directory of images called Images.
#
#           (3)  From the command line, run
#                    $ perl cleanHtml.prl
#
#           ...
#           Enter the directory containing the .gif image files 
#            (RETURN = equationExample_files): 
#           => Images
#           ...
#
#       (4)  Using PaintShopPro or another image converter, batch
#       convert the *.gif formulas to *.png files.
#
#       (5) Run cleanWebSite.prl which will convert the src links to each
#       *.gif equation in the *.html file to point to the *.png image 
#       equations.
#
#
# DEBUGGING
#
#       perl -d cleanWebPage.prl     Turn on the Perl debugger.
#       DB> b 225                    Break at specific line.
#       DB> c                        Continue.
#       DB> n                        Step to next line.
#       DB> s                        Step into subroutine.
#       DB> p $_                     Print a line.
#       DB> p $line =~ s/\.html/ ;   Print 1 if we get a match.
#
#
# AUTHOR
#
#     Sean E. O'Connor         5 Jun 2002  Version 1.0 released.
#                             10 Jun 2002  Version 1.1 released.
#                             27 Apr 2006  Version 1.3 released.
#
# LEGAL
#
#     cleanHtml Version 1.3
# 
#     A Perl utility program which cleans up Microsoft Word exported HTML
#     containing mathematical equations.
#
#     Copyright (C) 2002-2008 by Sean Erik O'Connor.  All Rights Reserved.
#
#     This program is free software; you can redistribute it and/or
#     modify it under the terms of the GNU General Public License
#     as published by the Free Software Foundation;  version 2
#     of the License.
#     
#     This program is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
#     
#     You should have received a copy of the GNU General Public License
#     along with this program; if not, write to the Free Software
#     Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
#     
#     The author's address is artifex@seanerikoconnor.freeservers.com. 
#
#===========================================================================

# Print the legal notice first.

printLegalNotice() ;




# Read arguments from the command line.

if (@ARGV == 6)
{
    $parentDirectoryName  = $ARGV[ 0 ] ;
    $htmFileName          = $ARGV[ 1 ] ;
    $oldImageDirectory    = $ARGV[ 2 ] ;
    $newImageDirectory    = $ARGV[ 3 ] ;
    $startingImageNumber  = $ARGV[ 4 ] ;
    $imageNumberIncrement = $ARGV[ 5 ] ;
}

# Or else prompt for arguments.

elsif (@ARGV == 0)
{

    ($parentDirectoryName, $htmFileName, 
     $oldImageDirectory,   $newImageDirectory,
     $startingImageNumber, $imageNumberIncrement) = getUserInput() ;
}

else
{
    print "Usage:\n" ;
    print "perl cleanHtml <parentDirectoryName>\n" ;
    print "               <htm file name>\n" ;
    print "               <old image directory>\n" ;
    print "               <new image file directory>\n" ;
    print "               <starting image number>\n" ;
    print "               <image number increment>\n" ;
}




# Add directory paths to the file names.

$fileName = $parentDirectoryName . "/" . $htmFileName ;




#  Create a new scratch file for writing, of same name but
#  a .html extension.

$newFileName = $fileName ;
$newFileName =~ s/\.htm/\.html/ ;





# Open the input file for reading and the output file for writing.

open( FILE, $fileName ) 
      || die( "Cannot open file $fileName" ) ;

open( NEWFILE, ">$newFileName" ) 
      || die( "Cannot open file $newFileName" ) ;




#  Print a custom HTML header with style sheet linkage.

printHTMLHeader( NEWFILE ) ;




# Set starting states.

$textState      = "" ;     
$imageState     = "" ;     
$newImageNumber = $startingImageNumber ;




#  Scan through each line of the file,
#  copying lines to the output file.

while ( <FILE> )
{
    # Process a single Mathematica image line which may be inside
    # a Mathematica text section and skip to the next line.

    if ( ($imageState eq "" || 
          $textState  eq "insideMathematicaText") &&
         detectMathematicaImageLine( $_ ) )
    {
        processMathematicaImageLine( $_, 
                                     $parentDirectoryName, 
                                     $oldImageFileDirectory, 
                                     $oldImageFileName, 
                                     $newImageFileName ) ;
        next ;
    }


    # Process the first line of an MS Word image specification.

    if ( detectMSWordImageLine( $_ ) )
    {
        $imageState = "insideImage" ;

        processMSWordImageLine( $_ ) ;
    }
 


    # Process the second (ending) line of an MS Word image specification.

    elsif ($imageState eq "insideImage" && 
           detectMSWordImageEndingLine( $_ ))
    {
        $imageState = "" ;

        processMSWordImageEndingLine( $_,
                                     $parentDirectoryName, 
                                     $oldImageFileDirectory, 
                                     $oldImageFileName, 
                                     $newImageFileName ) ;
    }


    # Process the beginning of a single MS Word text line.
    # Don't process a text line with an image in it.

    if ( $textState eq "" &&
           detectMSWordSingleTextLine( $_ ) &&
          !detectMSWordImageLine( $_ ) )   # Image line can resemble text.
    {
        processMSWordSingleTextLine( $_ ) ;
    }


    # Process the beginning of a multiple MS Word text line.
    # Don't process a text line with an image in it.

    elsif (  $textState eq "" &&
             detectMSWordMultilineText( $_ ) &&
            !detectMSWordImageLine( $_ ) )   # Image line can resemble text.
    {
        $textState = "insideText" ;

        processMSWordMultilineText( $_ ) ;
    }

    # End of a multiple MS Word text line.  Write it and seal off with 
    # paragraph end delimiter.

    elsif ( $textState eq "insideText" &&
            detectMSWordMultilineTextEnd( $_ ) ) 
    {
        $textState = "" ;

        processMSWordMultilineTextEnd( $_ ) ;
    } 

    # Write intermediate MS Word text lines to file.

    elsif ($textState eq "insideText")
    {
        print NEWFILE $_ ;
    }


    # Beginning of Mathematica multiline text.  Emit paragraph begin delimiter.

    if ( $textState eq "" && 
         detectMathematicaMultilineText( $_ ) )
    {
        $textState = "insideMathematicaText" ;

        print NEWFILE "\n<p>\n\n" ;
    }

    # End of Mathematica multiline text.  Write it and seal off with 
    # paragraph end delimiter.

    elsif ( $textState eq "insideMathematicaText" && 
            detectMathematicaMultilineTextEnd( $_ ) )
    {
        $textState = "" ;

        print NEWFILE "\n</p>\n\n" ;
    } 

    # Write intermediate Mathematica text lines to file.

    elsif ($textState eq "insideMathematicaText")
    {
        processMathematicaIntermediateText( $_ ) ;
    }

} # end while FILE



#  Print the end of the HTML file.

printHTMLTrailer( NEWFILE ) ;




# Rename the image file directory.

rename( $parentDirectoryName . "/" . $oldImageFileDirectory,
        $parentDirectoryName . "/" . $newImageFileDirectory ) ;



# Delete MS Word HTM file.

deleteFile( $fileName ) ;



#  Close both files.

close( FILE ) ;
close( NEWFILE ) ;


print "Done.\n" ;
print "Your HTML file is $newFileName\n" ;
print "Your renumbered equation .gif images are in $newImageFileDirectory\n" ;



#============================================================================
#
# NAME
#
#     deleteFile
#
#
# DESCRIPTION
#
#     Delete a file by hook or by crook.
#
# EXAMPLE
#
#     deleteFiles( "C:/junk.obj" ) ;
#
#============================================================================

sub deleteFile
{
    my( $fileName ) ;   # Define local variables
    ($fileName) = @_ ;  # and split up arguments.


    #  If delete doesn't work try a rename and then delete.

    if (!unlink( $fileName ))
    {
        print "Retrying delete $fileName.\n" ;

        $tempFileName = "junk" . time . ".txt" ;
        rename $fileName, $tempFileName ;


        #  Otherwise, just rename the file.

        if (!unlink( $tempFileName ))
        {
            rename $tempFileName, "C:/Temp" . $tempFileName ;

            print "Can't delete $fileName.  " ;
            print "Renamed it C:/Temp/$tempFileName.\n" ;
        }
    }
}



#============================================================================
#
# NAME
#
#     printLegalNotice
#
#
# DESCRIPTION
#
#     Print the legal notice.
#
# EXAMPLE
#
#
#============================================================================

sub printLegalNotice
{
    $legalNotice = 

    "\n" .
    "cleanHtml Version 1.3 -  Perl utility program which cleans up Microsoft Word\n" .
    "and Mathematica exported HTML containing mathematical equations.\n" .
    "Copyright (C) 2002-2008 by Sean Erik O'Connor.  All Rights Reserved.\n" .
   "\n" .
    "cleanHtml comes with ABSOLUTELY NO WARRANTY; for details see the\n" .
    "GNU General Public License.  This is free software, and you are welcome\n" .
    "to redistribute it under certain conditions; see the " .
    "GNU General Public License\n" .
    "for details.\n\n" ;

    print $legalNotice ;
}




#============================================================================
#
# NAME
#
#     getUserInput
#
#
# DESCRIPTION
#
#     Prompt the user for information.
#
# EXAMPLE
#
#
#============================================================================

sub getUserInput
{

#  Enter the parent directory, name of the .htm file from MS Word, the
#  name of the directory containing the .gif image files, the new
#  name desired and the starting image number.

print "Enter the parent directory containing the files.\n" ;
print "(RETURN = current directory)\n" ;
print "=> " ;
chop( $parentDirectoryName = <STDIN>) ;

if ($parentDirectoryName eq "") # Use the current directory, if none is given.
{
    $parentDirectoryName = "." ;
}

print "Enter the name of the .HTM file\n" ;
print "You can leave off the .htm extension\n" ;
print "=> " ;
chop( $htmFileName = <STDIN>) ;


if ( !($htmFileName =~ /\.htm/) )
{
    $htmFileName = $htmFileName . ".htm" ;
}

# Strip off the .htm, if any.

$temp = $htmFileName ;
$temp =~ s/\.htm//g ;

print "Enter the directory containing the .gif image files.\n" ;
print "RETURN = " . $temp . "_files\n" ;
print "=> " ;
chop( $oldImageFileDirectory = <STDIN>) ;

if ($oldImageFileDirectory eq "") # Use the MS Word default image file directory.
{
    $oldImageFileDirectory = $htmFileName ;
    $oldImageFileDirectory =~ s/\.htm/_files/ ;
}

print "Enter the new image file directory name.\n" ;
print "RETURN = WebPageImages\n" ;
print "=> " ;
chop( $newImageFileDirectory = <STDIN>) ;

if ($newImageFileDirectory eq "") # Default
{
    $newImageFileDirectory = "WebPageImages" ;
}

print "Enter the starting image number\n" ;
print "RETURN = 1\n" ;
print "=> " ;
chop( $startingImageNumber = <STDIN>) ;

if ($startingImageNumber eq "")
{
    $startingImageNumber = 1 ;
}

print "Enter the image numbering increment.\n" ;
print "RETURN = 2\n" ;
print "=> " ;
chop( $imageNumberIncrement = <STDIN>) ;

if ($imageNumberIncrement eq "")
{
    $imageNumberIncrement = 2 ;
}

return ( $parentDirectoryName, $htmFileName, 
         $oldImageDirectory,   $newImageDirectory,
         $startingImageNumber, $imageNumberIncrement )  ;
}



#============================================================================
#
# NAME
#
#     printHTMLHeader
#
#
# DESCRIPTION
#
#     Print my technical report XHTML 1.0 header.
#
# EXAMPLE
#
#
#============================================================================


#  Print an HTML header with style sheet linkage.

sub printHTMLHeader
{
    my( $FILEHANDLE ) ;   # Define local variables
    ($FILEHANDLE) = @_ ;  # and split up arguments.

    $xmlHeader =

    '<!DOCTYPE html' . "\n" .
    '         PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN' . "\n" .
    '         "DTD/xhtml1-strict.dtd">' . "\n\n" ;


    $commentHeader =

    "<!-- ===================================================================\n" .
    "|\n" .
    "| NAME\n" .
    "| \n" .
    "|     enterFileNameHere.html\n" .
    "|\n" .
    "| DESCRIPTION\n" .
    "|\n" .
    "|     Enter description here.\n" .
    "|     It is a hand-crafted XHTML 1.0 strict W3C validated document.\n" .
    "|\n" .
    "| AUTHOR\n" .
    "|\n" .
    "|    Sean Erik O'Connor\n" .
    "|\n" .
    "=================================================================== -->\n\n" ;


    $xmlHeader2 = 

    '<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">' . "\n\n" ;



    $documentHead =

    '<head>' . "\n" .
    '   <title>Document title goes here.</title>' ."\n\n" .
    '   <meta name="description" ' ."\n" .
    '         content="content goes here." />' ."\n\n" .
    '   <meta name="author"' ."\n" .
    '         content="Sean Erik O\'Connor" />' ."\n\n" .
    '   <meta name="copyright"' ."\n" .
    '         content="Copyright (C) 2000-2008 by Sean Erik O\'Connor.' . "\n" .
    '  All Rights Reserved." />' ."\n\n" .
    '   <meta name="rating"' ."\n" .
    '         content="general" />' ."\n\n" .
    '     <meta name="keywords" ' ."\n" .
    '           content="Sean Erik O\'Connor,' ."\n" .
    '                   keywords go here,' ."\n" .
    '                   mathematical software,' ."\n" .
    '                   source code, C software, ' ."\n" .
    '                   free software, freeware, shareware" />' ."\n\n" .
    '   <meta http-equiv="Content-type" content=\'text/html; charset="UTF-8"\' />' . 
    "\n\n" .
    '   <!-- Cascading style sheet for this document defines the fonts for all ' ."\n" .
    '        our text in this document.' ."\n" .
    '   -->' ."\n" .
    '   <link rel="StyleSheet" href="../../../StyleSheet.css" type="text/css" />' ."\n" .
    '</head>' ."\n\n" ;


     $bodyLine = 

    '<body class="technicalReport">' ."\n\n" ;



    print $FILEHANDLE $xmlHeader ;
    print $FILEHANDLE $commentHeader ;
    print $FILEHANDLE $xmlHeader2 ;
    print $FILEHANDLE $documentHead ;
    print $FILEHANDLE $bodyLine ;
}




#============================================================================
#
# NAME
#
#     printHTMLTrailer
#
#
# DESCRIPTION
#
#     Print my technical report XHTML 1.0 trailer.
#
# EXAMPLE
#
#
#============================================================================

sub printHTMLTrailer
{
    my( $FILEHANDLE ) ;   # Define local variables
    ($FILEHANDLE) = @_ ;  # and split up arguments.

    $footnote =

    '    <!-- And now a blue bar followed by a return to home button and' . "\n" .
    ' a copyright footnote to ' ."\n" .
    '         preserve my legal rights. ' ."\n" .
    '      -->' ."\n" .
    '    <p>' ."\n" .
    '    <img src="../../../WebPageImages/blueBar.png"' ."\n" .
    '         alt="Blue Bar Separator." ' ."\n" .
    '         height="5" width="1024" />' ."\n" .
    '    </p>' ."\n\n" .
    '    <!-- Quick return to my home page. -->' ."\n" .
    '    <p>' ."\n" .
    '    <a href="../../../index.html">' ."\n" .
    '       <img class="icon" ' ."\n" .
    '            src="../../../WebPageImages/home.png"' ."\n" .
    '            alt="Home Page Button." ' ."\n" .
    '            width="96" height="72" />' ."\n" .
    '       home' ."\n" .
    '    </a>' ."\n" .
    '    </p>' ."\n\n" .
    '    <p class="footnote">' ."\n" .
    '    Copyright &copy; 2000-2008 by Sean Erik O\'Connor.  ' ."\n" .
    '    All Rights Reserved.' ."\n" .
    '    &nbsp;&nbsp;&nbsp;' ."\n" .
    '    Last updated 00 Nov 3000.' ."\n" .
    '    </p>' ."\n\n" ;


    $htmlEnd =

    '</body>' ."\n" .
    '</html>' ."\n" ;


    print $FILEHANDLE $footnote ;
    print $FILEHANDLE $htmlEnd ;
}




#============================================================================
#
# NAME
#
#     detectMathematicaImageLine
#
#
# DESCRIPTION
#
#     Detect the beginning of a single image line with multiple images as we 
#     see in Mathematica, such as
#    
#         Step 2)Check if<IMG SRC="Images/mathematica_gr_27.gif"  
#         BORDER="0"  ALT="[Graphics:Images/mathematica_gr_27.gif]" 
#         WIDTH="46" HEIGHT="23"  ALIGN="absmiddle" >...
#
# EXAMPLE
#
#
#============================================================================

sub detectMathematicaImageLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

  return ($line =~ 
          /<IMG SRC="Images\/.*?_gr_[0-9]+\.gif.*WIDTH="[0-9]+"\sHEIGHT="[0-9]+"/) ;
}




#============================================================================
#
# NAME
#
#     processMathematicaImageLine
#
#
# DESCRIPTION
#
#
#     Process a single image line with multiple images as we 
#     see in Mathematica, such as (it's all one line):
#    
#        <H4>
#
#        Step 2)&nbsp;&nbsp;Check if&nbsp;&nbsp;
#        <IMG SRC="Images/math_gr_27.gif"  BORDER="0"  
#        ALT="[Graphics:Images/math_gr_27.gif]"  WIDTH="46" HEIGHT="23"  
#        ALIGN="absmiddle" >
#        <IMG SRC="Images/math_gr_28.gif"  BORDER="0"  
#        ALT="[Graphics:Images/math_gr_28.gif]"  WIDTH="24" HEIGHT="23"  
#        ALIGN="absmiddle" > 
#        is a primitive root of p.
#        <IMG SRC="Images/math_gr_29.gif"  BORDER="0"  
#        ALT="[Graphics:Images/math_gr_29.gif]"  WIDTH="15" HEIGHT="23"  
#        ALIGN="absmiddle" >
#
#        </H4>
#
#        <PRE><CODE><B><IMG SRC="Images/math_gr_25.gif"  BORDER="0"  
#        ALT="[Graphics:Images/math_gr_25.gif]"  hspace="45"  
#        WIDTH="245" HEIGHT="51"  ALIGN="absmiddle" ></B></CODE></PRE><PRE><SAMP>
#        <IMG SRC="Images/math_gr_26.gif"  BORDER="0"  
#        ALT="[Graphics:Images/math_gr_26.gif]"  hspace="47"  WIDTH="55" HEIGHT="25"  
#        ALIGN="absmiddle" ></SAMP></PRE>
#
# EXAMPLE
#
#============================================================================

sub processMathematicaImageLine
{   
    # Define local variables  # and split up arguments.

    my( $line, 
        $parentDirectoryName, $oldImageFileDirectory, 
        $oldImageFileName, $newImageFileName ) ;

      ( $line, 
        $parentDirectoryName, $oldImageFileDirectory, 
        $oldImageFileName, $newImageFileName ) = @_ ;


    # This is the beginning of an image specification.

    $startOfImagePattern = "<IMG SRC=" ;




    # Split the line apart with image information in each piece.

    @images = split( $startOfImagePattern, $line ) ;
    $numPieces = @images ;



    # Process each piece of the line.

    for ($i = 0 ;  $i < $numPieces ;  ++$i)
    {
        $image  = $images[ $i ] ;



        # Do a non-greedy match between the "Images/" and the ".gif" to pick up the 
        # file name prefix, "mathematica" and avoid
        # running into the middle of the ALT=...  .gif line.
        # Pick up the width and height sequence of digits, and any text after the 
        # ALIGN=...> section to the end of the line.

        if ( $image =~ 
             /"Images\/(.*?)_gr_([0-9]+)\.gif.*WIDTH="([0-9]+)"\s+HEIGHT="([0-9]+)/ )
        {
            # Extract the file name prefix.

            $fileNamePrefix = $1 ;


            # Pull out the image number.

            $oldImageNumber = $2 ;
    

            # Pull out the image width and height.
        
            $imageWidth  = $3 ;
            $imageHeight = $4 ;
                

            # Pull out any text near the end. 

            $image =~ /ALIGN="absmiddle"\s+>(.*)$/ ;
            $text = $1 ;


            #  Strip off nuisance HTML code.

            $text =~ s/<PRE>//g ;
            $text =~ s/<CODE>//g ;
            $text =~ s/<B>//g ;
            $text =~ s/<SAMP>//g ;
            $text =~ s/<\/PRE>//g ;
            $text =~ s/<\/CODE>//g ;
            $text =~ s/<\/B>//g ;
            $text =~ s/<\/SAMP>//g ;
            $text =~ s/&nbsp;//g ;



            # Rename the old image file name to equationNNN.gif

            $oldImageFileName = sprintf( $fileNamePrefix . "_gr_%d.gif", 
                             $oldImageNumber ) ;
            $newImageFileName = sprintf( "equation%03d.gif", 
                                         $newImageNumber ) ;

            rename( $parentDirectoryName . "/" . $oldImageFileDirectory . "/" . 
                    $oldImageFileName, 
                    $parentDirectoryName . "/" . $oldImageFileDirectory . "/" . 
                    $newImageFileName ) ;



            # Write the last image line.

            $equationNumber = sprintf( "%03d", $newImageNumber ) ;
            $string = "\n<img src=\"$newImageFileDirectory\/$newImageFileName\" " .
                      "alt=\"equation$equationNumber.\" \/>\n" ;

            print NEWFILE $string ;



            # Print text if any.

            print NEWFILE $text ;

        
            # Step up the image number.

            $newImageNumber += $imageNumberIncrement ;
        } 
        # Non-image text.  Write it out.
        else
        {
            #  Strip off nuisance HTML code.

            $image =~ s/<PRE>//g ;
            $image =~ s/<CODE>//g ;
            $image =~ s/<B>//g ;
            $image =~ s/<SAMP>//g ;
            $image =~ s/<\/PRE>//g ;
            $image =~ s/<\/CODE>//g ;
            $image =~ s/<\/B>//g ;
            $image =~ s/<\/SAMP>//g ;
            $image =~ s/&nbsp;//g ;

            print NEWFILE $image ;
        }

    } # end for i
}




#============================================================================
#
# NAME
#
#     detectMSWordImageLine
#
#
# DESCRIPTION
#
#     Here is a typical MS Word image lines for us to parse:
#
#         <p class=MsoNormal><span style='font-family:\w+'><sub><img width=286 height=42
#            src="./equationsExample_files/image014.gif"></sub></span></p>
# 
#     We want to rewrite it as:
#
#         <img src="TheoryWebPageImages/equation072.png" alt="equation072." />
#
# EXAMPLE
#
#
#============================================================================

sub detectMSWordImageLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

    # Match <img width=<number> <space> height=<number>
    return ($line =~ /<img width=[0-9]*\sheight=[0-9]*/) ;
}



#============================================================================
#
# NAME
#
#     processMSWordImageLine
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub processMSWordImageLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

    # Pull out the image width and height.

    $line =~ /<img width=([0-9]*)\sheight=([0-9]*)/ ;

    $imageWidth  = $1 ;
    $imageHeight = $2 ;
}



#============================================================================
#
# NAME
#
#     detectMSWordImageEndingLine
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub detectMSWordImageEndingLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

    # Match src=<directory>image<number>.gif
    return( $line =~ /src=".*image[0-9]*.*\.gif/ ) ;
}


#============================================================================
#
# NAME
#
#     processMSWordImageEndingLine
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub processMSWordImageEndingLine
{
    # Define local variables  # and split up arguments.

    my( $line, 
        $parentDirectoryName, $oldImageFileDirectory, 
        $oldImageFileName, $newImageFileName ) ;

    ( $line, 
      $parentDirectoryName, $oldImageFileDirectory, 
      $oldImageFileName, $newImageFileName ) = @_ ;



    # Pull out the image number.

    $line =~ /src=".*image([0-9]*).*\.gif/ ;

    $oldImageNumber = $1 ;



    # Rename the old image file name to equationNNN.gif

    $oldImageFileName = sprintf( "image%03d.gif", 
                                 $oldImageNumber ) ;

    $newImageFileName = sprintf( "equation%03d.gif", 
                                 $newImageNumber ) ;

    rename( $parentDirectoryName . "/" . $oldImageFileDirectory . "/" . 
                $oldImageFileName, 
            $parentDirectoryName . "/" . $oldImageFileDirectory . "/" . 
              $newImageFileName ) ;


    # Write the last image line.

    $equationNumber = sprintf( "%03d", $newImageNumber ) ;

    $string = "\n<img src=\"$newImageFileDirectory\/$newImageFileName\" " .
              "alt=\"equation$equationNumber.\" \/>\n" ;

    print NEWFILE $string ;



    # Step up the image number.

    $newImageNumber += $imageNumberIncrement ;
}


#============================================================================
#
# NAME
#
#     detectMSWordSingleTextLine
#
# DESCRIPTION
#
#     For a one line paragraph,
#
#         <p class=MsoNormal>though it converges slowly.  There are better formulas.</p>
#
#     rewrite as,
#
#         <p>
#         though it converges slowly.  There are better formulas.
#         </p>
#
# EXAMPLE
#
#
#============================================================================

sub detectMSWordSingleTextLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

    # Match <p class=MsoNormal>... text ... </p>
    return ( $line =~ /<p class=MsoNormal>.*<\/p>/ ) ;
}



#============================================================================
#
# NAME
#
#    processMSWordSingleTextLine
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub processMSWordSingleTextLine
{
    my( $line ) ;   # Define local variables
    ($line) = @_ ;  # and split up arguments.

    
    #  Pull out the text from the HTML syntax.
    #  <p class=MsoNormal>... text ... </p>

    $line =~ /<p class=MsoNormal>(.*)<\/p>/ ;
    $lineOfText = $1 ;


    # Don't print empty lines, 
    #      <p class=MsoNormal>&nbsp;</p>
    #      <p class=MsoNormal><span style='font-family:Verdana'>&nbsp;</span></p>
    
    if ( !($_ =~ /<p class=MsoNormal>.*&nbsp;<\/p>/) )
    {
        print NEWFILE "\n<p>\n$lineOfText\n</p>\n" ;
    }
}

#============================================================================
#
# NAME
#
#     detectMSWordMultilineText
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub detectMSWordMultilineText 
{
    my( $line ) ;
    ($line) = @_ ;

    # Match a line like
    #     <p class=MsoNormal><span style='font-family:Verdana'>Since the ring of
    return( $line =~ /<p class=MsoNormal>.*/ && !($line =~ /<\/p>$/) ) ;
}


#============================================================================
#
# NAME
#
#    processMSWordMultilineText
#
#
# DESCRIPTION
#
#     Here is a typical multiline MS Word text line for us to parse:
#
#     <p class=MsoNormal>cleanHtml.prl is a Perl utility to process the exported HTML
#     files generated by Microsoft Word or Mathematica.  It has been specialized to
#     format simple text with equations.</p>
#
#     We want to rewrite it as:
#
#         <p>
#         cleanHtml.prl is a Perl utility to process the exported HTML
#         files generated by Microsoft Word or Mathematica.  It has been 
#         specialized to format simple text with equations.
#         </p>
#
# EXAMPLE
#
#
#============================================================================

sub processMSWordMultilineText
{
    my( $line ) ;
    ($line) = @_ ;


    #  Pull out the text in the line.

    $line =~ /<p class=MsoNormal>(.*)/ ;
    $firstLineOfText = $1 ;

    print NEWFILE "\n<p>\n$firstLineOfText\n" ;
}


#============================================================================
#
# NAME
#
#     detectMSWordMultilineTextEnd
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub detectMSWordMultilineTextEnd
{
    my( $line ) ;
    ($line) = @_ ;

    # Detect ... text ...</p>

    return( $line =~ /.*<\/p>$/ ) ;
}


#============================================================================
#
# NAME
#
#     processMSWordMultilineTextEnd
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub processMSWordMultilineTextEnd
{
    my( $line ) ;
    ($line) = @_ ;


    # Pull out the text in the line.

    $line =~ /(.*)<\/p>/ ;
    $lastLineOfText = $1 ;

    print NEWFILE "$lastLineOfText\n</p>\n" ;
}


#============================================================================
#
# NAME
#
#    detectMathematicaMultilineText
#
#
# DESCRIPTION
#
#    Handle typical Mathematica multiline text:
#
#       <H4>
#
#       The number of distinct primitive polynomials is
#       </H4>
#
# EXAMPLE
#
#
#============================================================================

sub detectMathematicaMultilineText
{
    my( $line ) ;
    ($line) = @_ ;

    return( $line =~ /<H\d+>/ ) ;
}


#============================================================================
#
# NAME
#
#     detectMathematicaMultilineTextEnd
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub detectMathematicaMultilineTextEnd
{
    my( $line ) ;
    ($line) = @_ ;

    return ( $line =~ /<\/H\d+>/ ) ;
}


      
#============================================================================
#
# NAME
#
#     processMathematicaIntermediateText
#
#
# DESCRIPTION
#
#
# EXAMPLE
#
#
#============================================================================

sub processMathematicaIntermediateText
{
    my( $line ) ;
    ($line) = @_ ;


    # Strip off nuisance HTML syntax.

    $line =~ s/<PRE>//g ;
    $line =~ s/<CODE>//g ;
    $line =~ s/<B>//g ;
    $line =~ s/<SAMP>//g ;
    $line =~ s/<\/PRE>//g ;
    $line =~ s/<\/CODE>//g ;
    $line =~ s/<\/B>//g ;
    $line =~ s/<\/SAMP>//g ;
    $line =~ s/&nbsp;//g ;

    print NEWFILE "$line" ;
}