#!/usr/bin/perl
###################################################################
#  Building a Web Site Index by Keywords to Search
#  with JavaScript SiteSearch Engine, see:
#  http://members.tripod.com/~schick/sitesuche.html
#
#     Version 2.2 29May99, Perl Version 5.001 Win32
#       written by Arnold Schick (Copyright)
#
#   This Perl script is free for domenstic use
#   and is provided "as it is" (using at your own risk)
#
#  This Perl script key.cgi is to copy into
#  your cgi-bin directory on your local machine
#  (file: http://members.tripod.com/~schick/key.txt)
#  AND the related Perl script: spage.cgi
#  (file: http://members.tripod.com/~schick/spage.txt)
#
#  The HTML document with the FORM element for these
#  scripts are to copy into your Web subdirectory
#  (file: http://members.tripod.com/~schick/collect.html)
#
#  Names: Perl script     : key.cgi
#         Perl script     : spage.cgi
#         HTML document   : collect.html
#  call:  /cgi-bin/key.cgi from HTML document     
#
#  note: this script create files which are to
#        name within a FORM of that related
#        HTML document, deleting of files is
#        also possible.
#
#
#  note: Don't place this script into a public cgi-bin directory   
#        on a public Web server, it's a maintain tool of           
#        your own Web Site files on a local machine (PC)           
#                                                                  
#        if you use Apace Web server on WIN32, use the directive   
#        #!perl in place of #!/usr/bin/perl at top of this script  
#
#
#        For offline using, see below (nothing for beginners)  
###################################################################


$now = time();
$now = localtime($now);
@date = split(/ /,$now);

$content_length = $ENV{'CONTENT_LENGTH'};
read(STDIN, $posted, $content_length);

$posted =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C", hex($1))/eg;
$posted =~ s/\+/ /g;

#  for offline use: uncomment the following line and try that Perl script on command line, like: c:\perl5\perl.exe key.cgi > content.html 
#$posted="dirname=d:\\my_web_root&sort=newest&path=no&index=on&indexfile=c:\\tmp\my_index.js&home=/&ftype=htm&ftype=html";

#  the parameters can be changed for your own conditions ie. parameters: dirname=... indexfile=... home=...  ftype=... ftpye=...
#  path uses the values: no (no path), server (takes value from home), or local (takes value from dirname) for file listing
#  sort uses the values: newest, abc, zyx, or oldest  for file and subdirectory listing


@fields = split (/&/, $posted);

@ftype=();

foreach $name (@fields) {
   ($n, $w) = split(/=/, $name);
   if ($n eq 'ftype') {
     @ftype = (@ftype, $w);
   } else {   $tokens{$n} = $w;  }
}

@allInfos=();
if ( ($tokens{'temp_file'}) and (-e $tokens{'temp_file'}) ) {      #find selected information from the Start Page
  open(INFILE, "<$tokens{'temp_file'}");
    comp;
    @allInfos = <INFILE>;
  close(INFILE);

  if (@allInfos) {
    $c=0;
    foreach $line (@allInfos) {
      $line =~ tr/\n\r\f\t\b/ /s;
      $line =~ tr/ //d;
      ($if, $lf, $sp) = split(/,/, $line);
      push(@indexfiles, $if); 
      push(@localfiles, $lf); 
      push(@serverpath, $sp);
      $c++;
    }
    push(@indexfiles, $tokens{'indexfile'} );
    push(@localfiles, $tokens{'dirname'} );
    push(@serverpath, $tokens{'home'} ) ;

    @indexfiles = reverse sort keys %{{map {$_,1} @indexfiles }};  #remove duplicates
    @localfiles = reverse sort keys %{{map {$_,1} @localfiles }};  #remove duplicates
    @serverpath = reverse sort keys %{{map {$_,1} @serverpath }};  #remove duplicates
  }

  open(OUTFILE, ">$tokens{'temp_file'}");            #save selected information from the Start Page
   for ($b=0; $b <= $c; $b++) {
     print OUTFILE "$indexfiles[$b],", "$localfiles[$b],", "$serverpath[$b]\n";
   }
  close(OUTFILE);

} elsif ($tokens{'temp_file'}) {
    open(OUTFILE, ">$tokens{'temp_file'}");          #save selected information from the Start Page
      print OUTFILE "$tokens{'indexfile'},", "$tokens{'dirname'},", "$tokens{'home'}\n";
      print OUTFILE ",,";
    close(OUTFILE);
  }
 
print "Content-type: text/html", "\n\n";
print "<HTML>\n<HEAD>\n";
print "<META name=\"keywords\" content=\"Mein Inhaltsverzeichnis, My Table of Content, TOC\">\n";
print "<script language='javascript'>\n";
print "<!-- \n function oC() \{ alert ('\<' + 'META  NAME=\"keywords\"  CONTENT=\"key1, key2,...\"\>'); \}\n";
print "function oT() \{ alert ('\<' + 'TITLE>my Page<\/TI' + 'TLE>'); \}\n \/\/-->\n";
print "</script>\n"; 
print "<title>Files in Subdirectory $tokens{'dirname'}</title></head>\n";
print "<body>\n";

if ( (!$tokens{'indexfile'}) and ($tokens{'index'} or $tokens{'appindex'}) ) {
  print "missing Index File<br>\n";
  goto fin;
}

if ($tokens{'delindex'}) { 
  open(OUTFILE, ">$tokens{'indexfile'}")  || (print "Can't open $tokens{'indexfile'}: $!\n");
    print OUTFILE "\n";
  close(OUTFILE);
}

if ($tokens{'all'}) { @ftype=(); $ftype[0] = "*"; }

print "<font face='Arial' size=+1>\n";
print "Collection of Files in $tokens{'dirname'} by selected Filetype: @ftype<br></font>\n";
print "<font face='Arial' size=-1>\n";

$tokens{'dirname'} =~ s/\\$|\/$//;
$tokens{'dirname'} =~ s/\:$/\:\\/;

if ($tokens{'dirname'} and -d "$tokens{'dirname'}") {

  @files =();
  opendir(DIR,$tokens{'dirname'}) || (print "Can't open $tokens{'dirname'}: $!\n");
     @files = readdir DIR;
  closedir(DIR); 
} else { print "missing Subdirectory Name or given Name <b>$tokens{'dirname'}</b> is not a Subdirectory<br>\n"; }

if ($tokens{'dirname'} !~ m/.\\$/) { $tokens{'dirname'} = ($tokens{'dirname'} . "\\"); }


# define an array for all wanted filetypes and grep
$latest   = pop @ftype;
@typelist = ();

foreach $m ( @ftype ) {
  (@typelist) = (@typelist, "/\\.$m\$\/i ||");
}
(@typelist) = (@typelist, "/\\.$latest\$\/i");

@dateien = ();
foreach $m ( @files ) {           # find files for listing and find subdirectoríes
   if ((-f "$tokens{'dirname'}$m") and (grep { eval "@typelist" } $m) ) {
      (@dateien) = (@dateien,  $m);
   }
   if ((-d "$tokens{'dirname'}$m") and ($m ne ".") and ($m ne ".." )) {
      (@dirs) = (@dirs, $m);
   }
}
@files = @dateien;


if (@files or @dirs) {

#  prepare something to sort by date, use hashes fset and dset for date sorting
foreach $m (@files) {
     $p = "$tokens{'dirname'}" . "$m";
     (($fset{$m}) = (stat($p))[9]) or (print "can't stat file $m<br>\n");
}
foreach $m (@dirs) {
     $p = "$tokens{'dirname'}" . "$m";
     (($dset{$m}) = (stat($p))[9]) or (print "can't stat directory $m<br>\n");
}

   if ($tokens{'sort'} =~ m/newest/i) { @files = reverse sort{ $fset{$a} <=> $fset{$b} } keys %fset;  @dirs = reverse sort{ $dset{$a} <=> $dset{$b} } keys %dset; }
   if ($tokens{'sort'} =~ m/oldest/i) { @files = sort{ $fset{$a} <=> $fset{$b} } keys %fset;          @dirs = sort{ $dset{$a} <=> $dset{$b} } keys %dset;         }
   if ($tokens{'sort'} =~ m/abc/i)    { @files = sort{ uc($a) cmp uc($b) } @files;                    @dirs = sort{ uc($a) cmp uc($b) } @dirs;                    }
   if ($tokens{'sort'} =~ m/zyx/i)    { @files = reverse sort{ uc($a) cmp uc($b) } @files;            @dirs = reverse sort{ uc($a) cmp uc($b) } @dirs;            }

   if (!$tokens{'home'} == "") {
     if ($tokens{'home'} !~ m/.\/$/) { $tokens{'home'} = ($tokens{'home'} . "/"); }
     if ($tokens{'home'} !~ m/^\//)  { $tokens{'home'} = ("/" . $tokens{'home'}); }
     $tokens{'home'} =~ s/\/\//\//g;
   }

   if ($tokens{'path'} =~ m/local/i)  { $path = $tokens{'dirname'}; }
   if ($tokens{'path'} =~ m/server/i) { $path = $tokens{'home'}; }
   if ($tokens{'path'} =~ m/no/i) { $path = ""; }

   if ($tokens{'appindex'}) { 
      open(INFILE, "<$tokens{'indexfile'}") || ( print "Can't open Index File $tokens{'indexfile'}: $!, new one created<br>\n" );
        while (<INFILE>) {
          chomp;
          if ( index($_,"new makeArray(") ) {
            $last_index = substr($_, index($_,"(")+1,
                                     index($_,")") - index($_,"(")-1);
            last;
          }
        }
      close(INFILE);
      $last_index++;
   }

   if ($tokens{'index'}) { 
     open(OUTFILE, ">$tokens{'indexfile'}") || (print "Can't open Index File $tokens{'indexfile'}: $!\n");
       print OUTFILE "\/\/ JavaScript SiteSearch Index File, created: @date\n";
     close(OUTFILE);
     $tokens{'appindex'} = "on";
     $last_index = 1;
   } 

  print "<small>$tokens{'sort'} first</small><br>Date: @date<br><br><small>to navigate to these pages save this page to $tokens{'dirname'} and load it into the Web Browser<br>or an HTML Editor to maintain <a href='javascript:oT()'>titles</a> or <a href='javascript:oC()'>keywords</a> within the \&lt;HEAD&gt;&lt;/HEAD&gt; section.<br></small><br>\n";

  print "<TABLE border='0' cellpadding='0' cellspacing='0' width='100%'>\n";
  if (@dirs) {
    print "<tr><td valign='bottom'><font face='Arial' size='-1'>&nbsp;&nbsp;&nbsp;&nbsp;<b>Subdirectory</b></font></TD><TD valign='bottom'><font face='Arial' size='-2'><b>last modified</b></font></TD></tr>\n";
    $a = 1;
    foreach $i ( @dirs ) {
     if ($a < 10)  { $an = "\&nbsp; \&nbsp; " . $a;
       } else { if (($a > 9) and ($a < 100)) { $an = "\&nbsp; " . $a;
         } else { $an = $a; }
     }
    $p = "$tokens{'dirname'}" . "$i";
    (($mtime) = (stat($p))[9]) or (print "no subdirectory stats<br>\n");
    print "<TR><TD valign='top'><font face='Arial' size=-1>$an.&nbsp;<a href='$path$i'>$path$i</a>&nbsp;&nbsp;</TD><TD valign='top'><font size=-1><tt>", scalar localtime $mtime, "</tt></font></TD></TR>\n";
    $a++; 
    }
  }

  if (@files) {

    print "<tr><td valign='bottom'><font face='Arial' size='-1'><br>&nbsp;&nbsp;&nbsp;&nbsp;<b>Filename</b></font></TD><TD valign='bottom'><font face='Arial' size='-2'><b>last modified</b></font></td></tr>\n";
    $a = 1; 
    foreach $i ( @files ) {
     if ($a < 10)  { $anz = "\&nbsp; \&nbsp; " . $a;
       } else { if (($a > 9) and ($a < 100)) { $anz = "\&nbsp; " . $a;
         } else { $anz = $a; }
     }
     $p = "$tokens{'dirname'}" . "$i";
     (($mtime) = (stat($p))[9]) or (print "no file stats<br>\n");
     print "<TR><TD valign='top'><font face='Arial' size=-1>$anz.&nbsp;<a href='$path$i'>$path$i</a>&nbsp;&nbsp;</TD><TD valign='top'><font size=-1><tt>", scalar localtime $mtime, "</tt></font></TD></TR>\n";
     $a++; 
    }
  }

  print "</TABLE><br>\n";

} else { print "no files found<br><br>"; goto fin;}


$k += $last_index;
$anzahl = 0;
$anz    = 0;
$a      = 0;
$b      = 0;
@lines  = ();
@allkws = ();

if ($tokens{'index'} or $tokens{'appindex'}) {
  print "<FORM action='/cgi-bin/spage.cgi' method='post'><br>\n";
}

print "&nbsp;&nbsp;&nbsp;&nbsp;<b>List of Files with TITLE and HEADer Information:</b><br>\n";
print "<table border='0' cellspacing='1' cellpadding='0' width='100%'>\n";

foreach $i ( @files ) {

  if ($tokens{'index'} or $tokens{'appindex'}) {
    $checkbox = "<input type=checkbox name='file' value=\"$path$i\" checked>";
    } else { $checkbox = "<font face='Arial' size=-1>==&gt;</font>";
  }

  $fname = $tokens{'dirname'} . $i;

  $anzahl ++;
  if ($anzahl < 10)  { $anz = "\&nbsp;\&nbsp;\&nbsp;\&nbsp;" . $anzahl;
    } else { if (($anzahl > 9) and ($anzahl < 100)) { $anz = "\&nbsp;\&nbsp;" . $anzahl;
       } else { $anz = $anzahl; }
  }

  if (-T $fname) {
    open (EINGABE, $fname) || print "can't open $i: $!\n";
      chomp;
      @page = <EINGABE>;
    close(EINGABE) ;

    $ln = 0;
    foreach $line (@page) {                          # process each line within the HEADer

      $pos = index(uc($line), "</HEAD");
      $posT= index(uc($line), "</TITLE");

      # looking for the KEYWORDS:
      $kwline = $line;
      if (($kwline =~ s/\<META.*keywords.*content=//ig) or ($ln == -1)) {
        $ln = index($kwline,">");
        if ($ln gt -1 ) { $a++; }
        $kwline =~ tr/ \n\r\f\t\b/ /s;                # change some whitespaces to one blank
        $kwline =~ s/'|"|>|,$|^\s+|\s+$|\<HEAD//ig;
        (@keyw) = (@keyw, $kwline);
        $b += split (/,/ ,$kwline);
      }

      # looking for the AUTHOR:
      if ($kwline =~ s/\<META.*author.*content=//ig) {
        $kwline =~ tr/ \n\r\f\t\b/ /s;                # change some whitespaces to one blank
        $kwline =~ s/'|"|>|,$|^\s+|\s+$|\<HEAD//ig;
        (@keyw) = (@keyw, "," . $kwline);
        $b++;
      }

      # looking for the TITLE:
      if ($posT > -1) { 
        $pos1 = index(uc($line), "<TITLE>");
        $pos2 = index(uc($line), "</TITLE>");
        if ($pos1 > -1) {
          $title = substr($line,$pos1+7,$pos2-$pos1-7);
          $title =~ tr/ \n\r\f\t\b/ /s;               # change some whitespaces to one blank
          $title =~ s/^\s+|\s+$|\"//g;                # change whitespaces at the beginning and at the end, also dquotes
        }
      }

      # looking for the end tag of HEAD:
      if ($pos > -1) {

        if (!@keyw) {
          @keyw = ("no keywords"); 
        }

        (@lines) = ( @lines, $title );
        if (!$title)  { $title = "no title"; }
        print "<tr><td valign='bottom'>$checkbox</td><td valign='bottom'><font face='Arial' size=-1>File&nbsp;<a href='$path$i'>$anz:</a>&nbsp;Page&nbsp;Title:&nbsp;</font></td><td valign='bottom'><font face='Arial' size=-1><b>$title</b></font></td></tr>\n";
        print "<tr><td valign='top'><font face='Arial' size=-1>==&gt;</font></td><td valign='top'><font face='Arial' size=-1>File&nbsp;<a href='$path$i'>$anz:</a>&nbsp;Keywords:&nbsp;</font></td><td valign='top'><font face='Arial' size=-1><i>@keyw</i></font></td></tr>\n";

        foreach $m (@keyw) {
          $m =~ s/^\s+|\s+$//g;                      # change whitespaces at the beginning and at the end
          $m =~ s/, | ,|,/\",\"/g;                   # change commas between keywords with quotes for index file
        }

        if ($title !~ m/no title/s)  {               # indexing if title pesent 
          if ($tokens{'appindex'}) { 
            open(OUTFILE, ">>$tokens{'indexfile'}") || (print "Can't open Index File $tokens{'indexfile'}: $!\n");
               print OUTFILE "indexedPage[$k]=new index(\"$tokens{'home'}$i\",\"$title\",\"@keyw\");\n";
            close(OUTFILE);
          }
          $k++;                                      # counts index on index file
        }

        (@allkws) = (@allkws, @keyw);
        @keyw = ();
        $title= "";
        last;                                        # HEADer processed, jump to the next file

      }
    }
  if ($pos == -1) { print "<tr><td colspan=2><font face='Arial' size=-1>==&gt;File <a href='$path$i'>$anz:</a> <small>has no HEAD section</small></font></td></tr>\n"; }
  }
}

print "</table><font face='Arial' size=-1>\n";

foreach $key (@allkws) {
   @key   = split (/,/ , $key);
  (@keys) = ( @keys, @key);
}

if ($tokens{'appindex'}) {
  @allLines=();
  open(INFILE, "<$tokens{'indexfile'}") || (print "Can't open Index File $tokens{'indexfile'} $!\n");
    chomp;
    @allLines = <INFILE>;
  close(INFILE);

  if ($allLines[0] =~ m/^var indexed/ig) {  $f = shift(@allLines); }
  if ($allLines[1] =~ m/^var indexed/ig) {  $f = shift(@allLines); }

  foreach $line (@allLines) {                # clean up read file
    $line =~ s/^\s+//;                       # remove whitespaces at beginning
    $line =~ s/\\+/\\\\/g;                   # change all backslahes with double-backslashes   
  }

  $number = $k-1;

  open(OUTFILE, ">$tokens{'indexfile'}") || (print "Can't open Index File $tokens{'indexfile'} $!\n");
    print OUTFILE "var indexedPage = new makeArray($number);\n"; 
  close(OUTFILE);

  open(OUTFILE, ">>$tokens{'indexfile'}") || (print "Can't open Index File $tokens{'indexfile'} $!\n");
    print OUTFILE "@allLines\n"; 
  close(OUTFILE);
}

$n = $k - $last_index;
$an=@dirs;
if ($an == 1)     { $an = "1 Subdirectory"; } else { $an = "$an Subdirectories"; }
if ($anzahl == 1) { $anzahl = "1 File"; }     else { $anzahl = "$anzahl Files"; }
if ($n == 1)      { $n = "1 Web document"; }  else { $n = "$n Web documents"; }
if ($a == 1)      { $a = "1 Web document"; }  else { $a = "$a Web documents"; }
if ($b == 1)      { $b = "1 keyword"; }       else { $b = "$b keywords"; }


print "<br><b>$an found<br>$anzahl found<br>&nbsp; &nbsp; &nbsp;$n with titles found<br>&nbsp; &nbsp; &nbsp;$a with total $b found</b><br>\n";

if ($tokens{'appindex'}) { 
   print "<b>&nbsp; &nbsp; &nbsp;$n with titles indexed, $n added to $tokens{'indexfile'}.</b><br>\n";
}
if ($tokens{'delindex'}) {
   print "<small>existing JavaScript SiteSearch Index File $tokens{'indexfile'} deleted -if $tokens{'indexfile'} was there<br></small>\n";
}
if ($tokens{'index'}) { 
   print "<small>new JavaScript SiteSearch Index File $tokens{'indexfile'} created, existing file was deleted -if $tokens{'indexfile'} was existing<br></small>\n";
}
if ($tokens{'appindex'}) { 
   print "<small>continue with index: $last_index<br></small><br>\n";
}
if ( (!$tokens{'appindex'}) and (!$tokens{'index'}) ) { 
   print "<hr>\nYou can save this page on the Web browser (with Save As on File menu)<br> to: <i><b>$tokens{'dirname'}content.html</b></i><br> and that is then your Table of Content in that work subdirectory.<br>\n";
}
if ($tokens{'index'} or $tokens{'appindex'}) { 
  print "<hr>The produced JavaScript SiteSearch Index File is saved to <b><i>$tokens{'indexfile'}</i></b><br>and can be included into this JavaScript <a href='http://members.tripod.com/~schick/sitesuche.html'>SiteSearch source</a>.<br>\n";
  &create_JavaScript_Search;
  print "<hr>\n";
}
print "</font><p align=right><font face='Arial' size='-2'>created with Perl Script written by Arnold Schick</font></p>\n";
fin : ;
print "<font face='Arial' size='-2'><a href='javascript:history.back()'>back</a></font>\n";
print "</body></html>\n";

# finally, clean up, be sure for the next run:
$posted="";


sub create_JavaScript_Search {
  $spage = $tokens{'indexfile'};              # find dirname:
  $spage =~ s/.\w+$//i;                       # remove all chars from end until . (filetype)
  $spage =~ s/.\.$//;                         # remove period
  $spage =~ s/.\w+$//i;                       # remove filename
  $spage = $spage . "\\search.html";          # append new filename
  print "<br>Or <b>Take the checked files and create the Site Search Page with this tool:</b><br>\n";
  if ($last_index > 1) {
    print "<input type=hidden name=present value=\"$last_index\">\n";
  }
  print "<input type=hidden name=home value=\"$tokens{'home'}\">\n";
  print "<input type=hidden name=indexfile value=\"$tokens{'indexfile'}\">&nbsp; &nbsp; &nbsp; Search Page: &nbsp; &nbsp; &nbsp; &nbsp;</font>\n";
  print "<input type=text name=spage value=\"$spage\" size=24><br>\n<font face='Arial' size=-1>&nbsp; &nbsp; &nbsp; Results to Frame: &nbsp;</font>\n";
  print "<input type=text name=frame value=\"\" size=7> &nbsp; <font face='Arial' size=-1><small>(Name of Frame -if frames are inuse)</small><br>\n";
  print "<input type=submit value='                Create JavaScript Search Page             '><br>\n";
  print "</FORM><br>\n";
}
################ End of this Perl Script #############################################