#! /usr/local/bin/perl
#############################################################################
#
#                 NOTE: This file under revision control using RCS
#                       Any changes made without RCS will be lost
#
#              $Source: U:\\mhansen\\sites\\marchansen\\bin\\RCS\\view.cgi,v $
#            $Revision: 1.4 $
#                $Date: 2002-11-09 11:13:45-06 $
#              $Author: mhansen $
#              $Locker: mhansen $
#               $State: Exp $
#
#              Purpose: CGI to display tech note files as HTML. 
#
#          Description: Takes ASCII text files formatted in the "tech note"
#                       format and outputs HTML. Prepends standard header. 
#                       Appends standard footer. Generates and prepends index.
#                       Converts titles and section headings to different 
#                       display Format. Converts embedded URL's and other tech 
#                       note references to links.
#
#           Directions: Usage: view.cgi/[tech_note]
#                       If tech_note argument is supplied it must be in the 
#                       default tech note directory. Argument is passed as 
#                       extra path information, NOT with ? delimiter
#
#                       If no argument is supplied then it is assumed that
#                       the tech note is "index.txt" in the current directory.
#                       Presumably, there is an index.cgi in the same directory
#                       which is a link to view.cgi. This feature permits 
#                       simpler paths for tech note URLs.
#               
#     Default Location: http:/www.marchansen.com/bin/view.cgi
#
#           Invoked by: httpd, typically as UID nobody
#
#       Copyright (C) 1997  Marc Hansen
#
#       This program is free software; you can redistribute it and/or
#       modify it under the terms of version 2 of the GNU General Public
#       License as published by the Free Software Foundation available at:
#       http://http://www.gnu.org/copyleft/gpl.html
#
#       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.
#
#
#############################################################################
#
#
#       REVISION HISTORY:
#
#       $Log: view.cgi,v $
#       Revision 1.4  2002-11-09 11:13:45-06  mhansen
#       Fixed bug in index table
#
#       Revision 1.3  2000-03-04 00:52:35-06  mhansen
#       Added link conversion for ftp://
#
#       Revision 1.2  2000-02-19 16:16:33-06  mhansen
#       Added index
#
#       Revision 1.1  1998-05-09 13:22:43  mhansen
#       Change references to other tech notes to links
#
#       Revision 1.0  1997-09-13 16:27:36  mhansen
#       Initial revision
#
#




#
# Constants
#
$BASE ="/htdocs/userdirs/mhansen";          # Base path to httpd docroot from 
                                            #   perl interpreters point of view.
                                            
$DEFAULTNOTE="index.txt";                   # Default file name for tech note

$DIR="/tech";                               # Default directory for tech note


# 
# Get file to display
if ($ENV{PATH_INFO} =~ /^\/(.*)/) {         # If we have extra path info
    $file = $1;                             #   we assume it is path to tech
}                                           #   note and relative to $DIR.
else {
    $file = "$DEFAULTNOTE";                 # If we have no arguments then we
    $DIR = ".";                             #   assume file name is "index.txt"
}                                           #   and directory is current dir.

if ($DIR eq "."){                           # Set path to file
    $path = "$file"
} else {
    $path = "$BASE/$DIR/$file"
}



#
# Print HTTP header so we can see error messages
print "Content-type: text/html\n\n";


#
# Check for file


# Get tech note number from file name
$file =~ /^(\d+)_/;
$number = $1;



##
# Open tech note and read through it to build index, 
#   reformat, and insert links.
# Push note body on to array @text
# Push note index on to array @index
#


# Default format for body is <pre>
push(@text,"<pre>");                       

# Open file
open(NOTE, "< $path") ||                    
fatal("Cannot open file '$DIR/$file'");

# Begin loop through tech note
foreach $line (<NOTE>) {

    # Look for comment lines with name value pairs
    # which look like '#       key: ...'
    if ($line =~ /^#\s+([^:]+):\s*(.*)\s*$/){
        ($key,$value) = ($1,$2);
        $value =~ s/^\s+|\s+$//g;
        $value =~ s/\$$//g;

        if ($key =~ /Title/ ) {$title=$value}
        if ($key =~ /Number/ ) {
            $tnum=$value;
            $tnum =~ /^TN(\d+)/;
            $number = $1;
        }
    }


    # Look for section headings
    # lines begin with nnn.n
    elsif ($line =~ /^$number\.(\d+)(\s+)(.*)\s*$/){
    
        # Push on to index array
        push(@index, "<tr><td align=\"right\" width=\"10\"><font size=\"-1\"><a href=\"#$1\">$number.$1</a></font></td><td align=\"left\"><font size=\"-1\">&nbsp;&nbsp;$3</font></td></tr>\n");
        
        # Push on to body array
        push(@text, "<a name=\"$1\"></a></pre><h4>$number.$1&nbsp;&nbsp;&nbsp;&nbsp;$3</h4><pre>\n");
    }


    # If the line is not a comment line, then push it on to body array
    elsif ($line !~ /^#/) {
    
        # Translate characters that are not allowed in HTML 
        # to entity equivilants, must do & to &amp; first 
        $line =~ s/&/&amp;/g;

        #Translate any < or > symbols to &lt or &gt
        $line =~ s/</&lt;/g;
        $line =~ s/>/&gt;/g; 

        # Change references to email addresses to 'mailto:' links
        $line =~ s/([\w\.\/-]+\@\w+\.[\w\.\/-]+)/a("mailto:$1",$1)/eg;

        # Change http/s URL references to links
        $line =~ s/(https?:[#@~\w\.\/:\+\?=-]+[\/\w])/a($1,$1)/eg;

        # Change ftp URL references to links
        $line =~ s/(ftp:[#@~\w\.\/:\+\?=-]+[\/\w])/a($1,$1)/eg;

        # Change references to other tech notes to links
        $line =~ s/TN(\d\d\d)\.(\d+)/a("\/tn$1\/#$2","TN$1.$2")/eg;
        $line =~ s/TN(\d\d\d)([^\.])/a("\/tn$1","TN$1")/eg;

        # Push rewritten $line on to body array
        push(@text,$line);
        
    }
    
} # End of loop to read through tech note

# Close tech note file
close(NOTE);

# Close format tag for body of tech note
push(@text,"</pre>");




##
# Output tech note as HTML
#

#
# Open HTML document
print "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2 Final//EN\">\n";
print '<!-- Dynamically generated by view.cgi $Revision: 1.4 $ -->';
print "\n<html>\n";
print "<head>\n";
print "  <title>\n";
print "    $title\n";
print "  </title>\n";
print "  <meta name=\"description\" content=\"$title\">\n";
print "  <meta name=\"keywords\" content=\"$title\">\n";
print "  <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">\n";
print "</head>\n";
print "<body>\n";


#
# Prepend standard html header for this site
&include("/include/header.html");


# 
# Put whole page in a table to contain width
print "<table width=595 border=0 ><tr><td>\n";


#
# Document title
print "<h2 align=\"center\">$tnum - $title</h2>\n";


#
# Tech note index 
# goes in its own table inside whole page table
print "<table width=595 border=0 >\n";
print join('',@index); 
print "</table>\n"; 


#
# Tech note body
print join('',@text); 

#
# Close out "whole page" table
print "</td></tr></table>\n"; 


#
# Append standard html footer for this site
&include("/include/footer.html");


#
# Close HTML document
print "</body>\n";
print "</html>\n";

#
# Done
exit (0);



###################
#
#   Function: include ($file)
#    Purpose: copies file to stdout
#  Arguments: $file - path to file, relative to httpd docroot
#     Output: to stdout
# Depends on: Expects global variable $BASE set to base path 
#             to prepend to httpd docroot. 
#
sub include 
{
    my($file) = @_;
    open(INCLUDE, "< $BASE/$file") || fatal("Cannot include file '$file'");
    while (<INCLUDE>) {print $_}
    close(INCLUDE);
    return 0;   
}



###################
#
#   Function: a ($href, $text)
#    Purpose: rewrite args as HTML anchor to simplify sytax in regex
#  Arguments: $href - href path
#             $text - text of link
#     Output: returns string <a href=$href>$text</a>
#
sub a 
{
    my($href, $text) = @_;
    return '<a href="'.$href.'">'.$text.'</a>';
}


###################
#
#   Function: fatal ($msg)
#    Purpose: Print error message to stdout, close HTML page and exit
#  Arguments: $msg - Error message to print
#     Output: to stdout
#
sub fatal
{
    my($msg) = @_;
    print($msg);
    print "\n</body>\n</html>\n";
    exit 1;
}