#!/usr/local/bin/perl
#
#	Index2html --  Generate new 00index.html files.
#
#    This program also creates the djgpp/00allindex.html file.
#    For safety and system reliability, index.pl and longindex.pl are kept
#    in 2 separate files, disrecarding their similarity.
#
    $Version= "V1.4 PL 01  (17 Jun 1998)";
#
#   Written by
#	Tommi Saarinen	(tsa@x2ftp.oulu.fi)
#	Jouko Valta	(jopi@x2ftp.oulu.fi)
#
#   Revision History
#	V1.1 pl 01  -- 11 Apr 1997
#	 Updated "Copyright 1997".
#
# 	V1.2 pl 00  -- 15 Aug 1997
#	 Newlines and NO.SPAM added.
#	 Alternate TXT index file names.
#	 Test if target is File, Dir or even non-existent.
#	 Option to write alternate file. (For DJGPP).
#
# 	V1.2 pl 01  -- 28 Nov 1997
#	 Protect any HTML metacharacters in the description.
#
# 	V1.3 pl 00  -- 21 Mar 1998
#	 Add 'Included on CDROM' markers via an option.
#	 CD-ROM database checking taken from CdTest V1.1 pl 00.
#
# 	V1.4 pl 00  -- 12 Jun 1998
#	 Distinquish icons between Text, Binary, and unknown files.
#	 Icons can be selected according to the file type, with default value
#	  from binary test based on the actual data contained on the file.
#	 Relative subdirectories (mainly for djgpp directory).
#
# 	V1.4 pl 01  -- 17 Jun 1998
#	 Combined main version with the separate parallel CD-ROM patch.
#
#
#


$Usage = "\
    Options:\
	-h	 Show help page and exit\
	-CD	 Specify alternate path for CDROM database\
	-cd	 Add 'included on cdrom' markers\
	-nc	 Suppress cdrom markers\
	-f file	 Specify input file\
	-w file	 Write html index 'file'\n\n";


# -----------------------------------------------------------------------------

## Declare CDROM Constants

$CD_NONE	= 0;
$CD_YES		= 1;
$CD_FREE	= 2;
$CD_NONPROFIT	= 3;
$CD_GPL		= 4;
$CD_DEMO	= 5;				# Demo version, not for CD
$CD_PERM	= 6;				# Prior permission required
$CD_NO		= 7;
$CD_CONFL	= 8;		# Claimed yes, but contents seem suspicious
$CD_ASKED	= 16;


@CDPrefix = (   "",     "ON_", "ON_", "ON_", "",    "OFF_", "OFF_", "OFF_",
	        "OFF_", "",    "",     "",   "",    "",     "",     "",
	        "ASKED_"
);
@CDString = (   "N/A",  "Yes", "Yes", "Yes", "N/A", "No",   "No",   "No",
		"No",   "N/A", "N/A", "N/A", "N/A", "N/A",  "N/A",  "N/A",
	        "N/A"
);


$FIND		= "/bin/find";
$SORT		= "/bin/sort";

$CDROMDIR	= "/ftp-service/cdrom/pub/msdos";	# The X2 CDROM Database


$CDMarkers = 0;
$CDFileIcon = "../x2www/cdfile32.gif";


# -----------------------------------------------------------------------------

# Defaults

$YEAR =	`/bin/date +%Y`;		# Copyright year (in 4 digits)
$RECENT = "00recent.txt";      		# Recent index file
$MAINHTML = "../00index.html";		# The programming/ main index.

$TXTINDEX = "00index.txt";
$HTMLINDEX= "00index.html";


# HTML Index File Generation Date

$date = `/bin/date`;
chop $date;

@mons = (
    "Jan", "Feb", "Mar", "Apr", "May", "Jun",
    "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"
);


### process any FOO=bar switches
eval '$'.$1.'$2;' while $ARGV[0] =~ /^([A-Za-z_]+=)(.*)/ && shift;

### Command line options
while ($_ = $ARGV[0], /^-/) {
	shift;
	last if /^--$/;
	if (/^-h/)	{ print $Usage; exit 2; }

   if (1) {
	if (/^-CD/)	{ print STDERR "-CD ignored\n"; shift; next; }
	if (/^-cd/)	{ print STDERR "-cd ignored\n"; next; }
	if (/^-nc/)	{ print STDERR "-nc ignored\n"; next; }
   }
   else {
	if (/^-CD/)	{ if ($ARGV[0]) {
				++$CDMarkers;
				$CDROMDIR = $ARGV[0]; shift;
			  }
			  next;
			}
	if (/^-cd/)	{ ++$CDMarkers; next; }
	if (/^-nc/)	{ $CDMarkers = 0; next; }
   }

	if (/^-f/)	{ $TXTINDEX  = $ARGV[0]; shift; next; }
	if (/^-w/)	{ $HTMLINDEX = $ARGV[0]; shift; next; }
	if (/^-/)	{ die "\nUnsupported argument near $_\n Stopped"; }
}


# Directory for which a HTML-index should be generated
$DIR = $ARGV[0];
$DIR || die "No directory specified. Stopped";


# ----------------------------------------------------------------------------
# Read CDROM Database
#

if ($CDMarkers) {

    Read_CDMarkers();

}  # cdmarkers


# ----------------------------------------------------------------------------

$errs = 0;

# Try all ascii index file names
open (I, "$DIR/$TXTINDEX") ||
open (I, "$DIR/00_index.txt") ||
open (I, "$DIR/00index") ||
open (I, "$DIR/00INDEX") ||
   die "Cannot read index file for '$DIR/$TXTINDEX'";

open(H, ">$DIR/$HTMLINDEX") || die "Cannot open '$DIR/$HTMLINDEX'";
#select(H);

# rm -f ${DIR}/${HTMLINDEX};


printf STDERR "Processing:  $DIR/$TXTINDEX ...";

# Print headers

print_html_headers();


# Directory Listing

$LastDir = "";			# All files listed are on the CWD by default.

line: while (<I>) {
    chop;

    if (/^\s*$/) {   # Empty line
	print H "\n";
	next;
    }

    if (/\s*(\S+)\s+(.+)/) {
	$file = $1;
	# Protect any HTML metacharacters in the description.
	$short = html_encode($2);

	if ($LastDir && ! -e"$DIR/$file") {
	    if ( -e "$DIR/$LastDir/$file") {
		$file = "$LastDir/$file";
	    }
	}

	($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
	 $atime,$mtime,$ctime,$blksize,$blocks)
		= stat("$DIR/$file");

($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst)=localtime($mtime);

#	$datestring=$mday.'-'.$mons[$mon].'-'.$year.' '.$hour.':'.$min;

	if ($size < 1024) {
	    $size_string="bytes";
	} elsif($size/1024 < 1024) {
	    $size=$size/1024;
	    $size_string="kB   ";
	} else {
	    $size=$size/(1024*1024);
	    $size_string="MB   ";
	}


	# Align filename

	if (length($file) < 8) {
	    $s = "\t";
	}
	else {
	    $s = "";
	}


	#
	# Select icon type and print one directory entry.
	#

	if ( -f _) {
	    $LinkType="internal-gopher-unknown";

	    #  \s matches whitespace or newline
	    if ($file =~ /(\S+)\.([0-9A-Za-z\+\-\_\~\^\$\*\%\!\|]+)\s*$/) {
		$MainName = $1;
		(($MainType = $2)) || die "Filename extraction failed for '$file'";

		## Binary file: zip, lzh, lha, arj, arc, exe, com, ...
		#
		# An option may be used to flag these all as "binary" or
		# to distinquish between  these types even further:
		#  archive, audio, image, executable, ...
		#

		if ($MainType =~ /^arc|^arj|^lha|^lzh|^zip|^zoo|^rar|^exe|^com|^hlp|^rtf|^pif|^dll|^cur|^fnt|^pal|^sys/i ||
		    $MainType =~ /^\S\S_|^dsm|^s3m|^voc|^wav|^dat|^3ds|^bbm|^bmp|^lbm|^map|^gif|^jpg|^pcx|^pic|^ppm|^raw|^tga/i) {

		    $LinkType="internal-gopher-binary";
		}

		## Text file: txt, doc, inf, ini, diz, c, h, pas, ...
		#
		# Note: *.doc is assumed text unless further separation is performed.
		#

		if ($MainType =~ /^txt|^e?ps|^faq|^man|^htm|^swg/i ||
		    $MainType =~ /^diz|^nfo|^log|^lsm|^www|^bbs/i ||
		    $MainType =~ /^asm|^bas|^cc*$|^cpp|^hh*$|^hpp|^hxx|^inc|^mak|^prj|^pas|^tpu|^s$/i ||
		    $MainName =~ /^\!*read$|readme|read1st|descript|howto/i ||
		    $MainType =~ /^me|^now|^1st/i) {

		    $LinkType="internal-gopher-text";
		}

		# Unspecified files and the .doc format.
		else {
		    if ( -T _) {
			$LinkType="internal-gopher-text";
		    }
		    elsif ( -B _) {
			$LinkType="internal-gopher-binary";
		    }
		}
	    }  # file name

	    # Extensionless files
	    else {
		if ( -T _) {
		    $LinkType="internal-gopher-text";
		}
		elsif ( -B _) {
		    $LinkType="internal-gopher-binary";
		}
	    }

	    ## Default is Unknown filetype

	    printf H ("<A HREF=\"%s\"><IMG ALIGN=ABSBOTTOM ALT=\"[FILE]\" BORDER=0 SRC=\"$LinkType\">%s</A>%s\t%02d-%s-%d %02d:%02d   %5.1f %s\t%s\n",
		      $file, $file, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string, $short);

	}  # file

	elsif ( -d _) {
	    printf H ("<A HREF=\"%s\"><IMG ALIGN=ABSBOTTOM ALT=\"[DIR]\" BORDER=0 SRC=\"internal-gopher-menu\">%s</A>%s\t%02d-%s-%d %02d:%02d   %5.1f %s\t%s\n",
	    $file, $file, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string, $short);

	    $LastDir = $file;
	}  # directory

	else {
	    print H "$_\n";
	    printf STDERR "BAD LINE >>$_<<\n";
	    ++$errs;
	}  # any dir entry

    }  # line
}  # while


# Print footers

print H "</PRE>";
print H "<P><HR>";

print H "Compilation Copyright $YEAR by X2 Support Group";
print H "<P><HR>";
print H "<ADDRESS>";
print H "Index2html $Version by X2 Support Group (x2ftp\@x2ftp.oulu.fi.NO.SPAM)";
print H "</ADDRESS>";
print H "</BODY>";
printf H ("</HTML>\n");

close H;
chmod 0664, "$DIR/$HTMLINDEX";

printf STDERR "done\n";

exit $errs;


# -----------------------------------------------------------------------------
#
# Print headers
#

sub print_html_headers {

print H "<HTML>";
print H "<HEAD>";
print H "<TITLE>x2ftp.oulu.fi - ${DIR}</TITLE>\n";
print H "<!-- Formatted to HTML for x2ftp\@x2ftp.oulu.fi.NO.SPAM  $date -->\n";
print H "</HEAD>";

print H "<BODY>";
print H "<H1 ALIGN=CENTER>x2ftp.oulu.fi</H1>";
print H "<H2 ALIGN=CENTER>X2 Support Group Game Development Archives<BR>\n";
print H "Directory: /pub/msdos/programming/${DIR}</H2>\n";
print H "<HR>\n";
printf H ("<A HREF=\"../%s\">What's new</A> on X2FTP<BR>\n", $RECENT);

if ($MAINHTML && -f $MAINHTML && -s $MAINHTML) {
    print H 'Go back to <A HREF="$MAINHTML">/pub/msdos/programming</A> Index';
}
else {
    print H 'Go back to <A HREF="..">/pub/msdos/programming</A>';
}

print H "<P><PRE>";
print H "      Name\t   Last modified\tSize\t   Description";
print H "<HR>";

}  # sub print_html_headers


# -----------------------------------------------------------------------------
#
#  Protect any HTML metacharacters.
#

sub html_encode {
    local ($value, $foo) = @_;

    $value =~ s/\&/\&amp;/;
    $value =~ s/\</\&lt;/;
    $value =~ s/\>/\&gt;/;

    return $value;

}  # html_encode



# -----------------------------------------------------------------------------
#
# This part is only used on the CD-ROM Marker version.
#

# ----------------------------------------------------------------------------
# Read CDROM Database
#

sub Read_CDMarkers {

$ReadCDList	= 1;
$TimeCheck	= 0;
$PrintNoMain	= 1;
$PrintTime	= 0;
$TimeLimit	= 4*3600;			# Allow delay (in seconds)

$CheckRecent	= 0;				# Check recent files.
$CheckLimit	= 5*7;				# Find recent files for N days.


    # Check we have valid Icon File.
    if (! -r "$DIR/$CDFileIcon") {		# Test reference to the icon.
	$CDMarkers = 0;
	printf STDERR  "*** Icon file '$DIR/$CDFileIcon' not found!\n";
    }

    else {

	# Set CDROM Search Path

	if ($DIR =~ /^\//) {			# Absolute path
	    $cdpath = $DIR;
	}
	elsif ($DIR eq ".") {			# Current work dir
	    $cdpath = `/bin/cwd`;
	}
	else {					# Relative path
	    $cdpath = `/bin/cwd` . "/$DIR";
	}

	if ($cdpath =~ /msdos\/(\S+)/) {	# Check for default prefix...
	    $cdpath = $1;			# and cut off excess path.
	}

	$cdpath =~ s/\//\/\*/g;

	$FIND_CDROM	= "$FIND $CDROMDIR/\*$cdpath -type f -name ON_\* -print";

	scan_cdrom();
    }

}  # read_cdmarkers


# -----------------------------------------------------------------------------
# Scan the CDROM Database
#
# -----------------------------------------------------------------------------


# Get the directory descriptions for ON_dir and OFF_dir ...

sub scan_cdrom {

open (foo, "$FIND_CDROM |") || die "Can't excute '$FIND_CDROM'";


printf "\nScanning the CDROM Database for '$DIR'.\n";
printf "$FIND_CDROM\n";
printf "\n";


 cdline: while (<foo>) {
     print $_;
     chop;

    if (/^\s*$/) { next cdline; }			# Empty line

    # Takes the last slash by default.
    if (/(\S+)\/(\S+)/) {				# File in any CD dir
	$dir  = $1;
	$file = $2;
	$main = $2;

#	$main = $dir . "/" . $file;
	$main =~ s/ON_//os;


	# Set flags for FTP file
	# Both the parent dir dir and the file must default to 0.

	if ($FileStatus{$dir} || $FileStatus{$main}) {
	    printf "*** Conflicting CDROM values for  $_.\n"; 
	}

	if (/ON_/) {
	    $FileStatus{$main} = $CD_YES;
	}
	elsif (/OFF_/) {
	    $FileStatus{$main} = $CD_NO;
	}
	elsif (/ASKED_/) {
	    printf "ASKED for:  $_.\n";
	    $FileStatus{$main} = $CD_ASKED;
	}
	else {						# Non-prefixed CD file
	    printf "*** Invalid CDROM entry  $_.\n";
	    $FileStatus{$main} = $CD_NONE;
	    ++$CdFilesBad;
	}

    }  # cdline
}  # while
}  # scan_cdrom