#!/usr/local/bin/perl
#
#	LongIndex --  Generate new 00longindex.{txt,html} files.
#
#    For safety and system reliability, index.pl and longindex.pl are kept
#    in 2 separate files, disrecarding their similarity.
#
    $ProgVersion= "1.0 PL 00  (26 Nov 1997)";
#
#   Written by
#	Tommi Saarinen	(tsa@x2ftp.oulu.fi)
#	Jouko Valta	(jopi@x2ftp.oulu.fi)
#
#   Revision History
# 	V0.9 pl 00  -- 18 Aug 1997
#	 Patched from index.pl script.
#
# 	V1.0 pl 00  -- 26 Nov 1997
#	 Initial revision finished.
#
#



$Usage = "\
    Options:\
	-h	  Show help page and exit\
	-f file	  Read text index 'file'\
	-i file	  Read textfiles cross-index 'file'\
	-wh file  Write HTML index to 'file'\
	-wt file  Write TXT index to 'file'\n\n";


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

# 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";
$LONG_TXT = "00longindex.txt";
$LONG_HTML= "00longindex.html";

$LOGHOME	= "/pc/log";

if (! -d $LOGHOME || ! -w $LOGHOME) {
    printf STDERR "Cannot access $LOGHOME. Please, check the setting or run\n create_dosadm_logs script.\n";
    $LOGHOME = "/var/tmp";
}

$INDEXDIR	= $LOGHOME . "/programming";	# Re-formatted TXT files.
$TEXTDIR	= "/src/txt";
$FTPDIR		= "/ftp-service/ftp/pub/msdos";
$CDROMDIR	= "/ftp-service/cdrom/pub/msdos/programming";
$CROSSINDEX	= $TEXTDIR . "/00index.txt";


# Index Files Generation Date

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

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


$errs	= 0;				# Number of bad mainfiles.
$n_files= 0;				# Number of mainfiles found.
$n_ok	= 0;				# Number of index files found.
$n_fail	= 0;				# Number of index files missing.
$opt	= 0;

### 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 (/^-f/)	{ $TXTINDEX  = $ARGV[0]; shift; next; }
	if (/^-i/)	{ $CROSSINDEX= $ARGV[0]; shift; next; }
	if (/^-wh/)	{ $LONG_HTML = $ARGV[0]; $opt |= 1; shift; next; }
	if (/^-wt/)	{ $LONG_TXT  = $ARGV[0]; $opt |= 2; shift; next; }
	if (/^-/)	{ die "\nUnsupported argument near $_\n Stopped"; }
}

if ($opt && $opt != 3) {
    die "Error: Both '-wh' and '-wt' options are required. Stopped";
}


# -----------------------------------------------------------------------------
# Read textfiles cross-index file.
#

open (CI, "$CROSSINDEX") ||
   printf STDERR "Cannot read index file '$CROSSINDEX'";

file: while (<CI>) {
    chop;

    if (/^\s*$|^%/) { next file; }		# Empty line or comment

    #	   category  main     txt  longtext
    if (/^\s*(\S+)\s+(\S+)\s+(\S+)\s+(\S+)\s*/) {
	$RealCategory{$2} = $1;
	$TextName{$2} = $3;
	$IndexName{$2} = $4;
	next file;
    }

    print "Incorrect line '$_'\n";

}  # while

close(CI);


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

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


# 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 (T, ">$DIR/$LONG_TXT")  || die "Cannot open '$DIR/$LONG_TXT'";
open (H, ">$DIR/$LONG_HTML") || die "Cannot open '$DIR/$LONG_HTML'";


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


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


# Print headers

print_text_headers();
print_html_headers();


# Directory Listing

line: while (<I>) {
    chop;

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

    if (/\s*(\S+)\s+(.+)/) {
	($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
	 $atime,$mtime,$ctime,$blksize,$blocks)
		= stat("$DIR/$1");
($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   ";
	}

	if (length($1) < 8) {
	    $s = "\t";
	}
	else {
	    $s = "";
	}


	if ( -f _) {
	    printf T ("\n%s%s\t%02d-%s-%d %02d:%02d     Size %5.1f %s\n", $1, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string);

	    printf H ("<P><B><A NAME=%s></A><A HREF=%s><IMG ALIGN=ABSBOTTOM ALT=\"[FILE]\" BORDER=0 SRC=\"internal-gopher-unknown\">  %s</A></B>%s\t%02d-%s-%d %02d:%02d     Size %5.1f %s\n",
	    $1, $1, $1, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string);

	    ++$n_files;				# Count files in directory.
	    copy_long_description ($1, $2);
	}

	elsif ( -d _) {
	    printf T ("\n%s%s\t%02d-%s-%d %02d:%02d     Size %5.1f %s\n", $1, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string);

	    printf H ("<P><B><A NAME=%s></A><A HREF=%s><IMG ALIGN=ABSBOTTOM ALT=\"[DIR]\" BORDER=0 SRC=\"internal-gopher-unknown\">  %s</A></B>%s\t%02d-%s-%d %02d:%02d    Size %5.1f %s\t%s\n",
	    $1, $1, $1, $s, $mday, $mons[$mon], $year, $hour, $min, $size, $size_string, $2);
	}

	else {
	    print T "$_\n";
	    print H "$_\n";
	    printf STDERR "BAD LINE >>$_<<\n";
	    ++$errs;
	}

    }  # line
}  # while


# Print footers

print T "\n------------------------------------------------------------------------------\n";
print T "Compilation Copyright and Index Files Copyright (C) $YEAR by X2 Support Group\n\n";
print T "LongIndex $ProgVersion by X2 Support Group\n";


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

print H "Compilation Copyright and Index Files Copyright (C) $YEAR by X2 Support Group";
print H "<P><HR>";
print H "<ADDRESS>";
print H "LongIndex $ProgVersion by X2 Support Group (x2ftp+\@\@+x2ftp.oulu.fi.NO.SPAM)";
print H "</ADDRESS>";
print H "</BODY>";
printf H ("</HTML>\n");


close T;
close H;
chmod 0664, "$DIR/$LONG_TXT";
chmod 0664, "$DIR/$LONG_HTML";

printf "\nDirectory %-10s [ %2d files]:  %2d Ok, %2d failed.\n", $DIR, $n_files, $n_ok, $n_fail;

printf STDERR "done\n";
exit $errs;


# -----------------------------------------------------------------------------
#
##  TXT File Handling Subroutines
#
# -----------------------------------------------------------------------------

# Find matching txtfile on the INDEXDIR. This routine is called if
# the pre-formatted txtfile was not named specificly.
#

sub get_txt_name {
   local ($mainfile, $seekpath) = @_;
   local $txtfile = '';

   ($hum, $foo) = split (/\./, $mainfile, 2);

 print "\nSearch $mainfile => $hum (DIR $seekpath)";

   if ($hum) {
	$hum .= ".*";
	while(<"$seekpath/${hum}">) {
	    if (/(\S+\.)(txt|lsm|doc|diz|nfo|readme)/i) {
		$txtfile = $_;
	    }
	}
    }  # hum

    return $txtfile;
}


# -----------------------------------------------------------------------------
#
# Copy the Long Description Form from database directory.
#
# NOTE: The output does not comply with LSM rules any more.
#


sub copy_long_description {
    local ($mainfile, $short) = @_;

    $LSMmode	= 0;
    $Category	= "";
    $Filename	= "";
    $Title	= "";
    $Version	= "";
    $EnterDate	= "";
    $Description= "";
    $Keywords	= "";
    $Source	= "";
    $FormatsList= "";
    $Author	= "";
    $Maintainer	= "";
    $Site	= "";
    $Platforms	= "";
    $Requires	= "";
    $CopyingPolicy="";


#    $txtfile = $TextName{$mainfile};
#    $txtfile = get_txt_name ($mainfile, $TXTDIR);

    if ( ($keysfile = $IndexName{$mainfile})) {
	$keysfile = $INDEXDIR . "/" . $keysfile;
    }
    else {
	$keysfile = get_txt_name ($mainfile, $INDEXDIR);
    }


    if (!open (D, $keysfile)) {
	printf STDERR "Cannot read '$keysfile'\n";

	print T "\t$short\n";

	print H "<UL>\n";
	print H "$short<BR>\n";
	if (! $mainfile =~ /index/) {
	    print H "No further information available.\n";
	}
	print H "</UL>\n\n";
	return -1;
    }


    # Check that correct fields are copied. Skip the rest.
    # The text file being read here is completely re-formatted by
    # 'dosadm' to ensure correct keyword syntax troughout the file.
    #

    while (<D>) {
	chop;
	if (/^\s*$/) { next; } 

	last if (/^-----------------------/);		# Unused feature

#	if (/^Begin3$/) {
#	    ++$LSMmode;
#	    next;
#	}


	### Scan for Required fields ###

	if (/^Type\s*:\s*(.+)/i || /^Category\s*:\s*(.+)/i) {
	    $Category = $1;
	    next;
	}
	elsif (/^\s*Program\s*:\s*(.+)/i) {
	    $Filename = $1;
	    next;
	}
	elsif (/^\s*Title\s*:\s*(.+)/i) {
	    $Title = $1;
	    next;
	}
	elsif (/^\s*Version\s*:\s*V?\s*(.+)/i) {
	    $Version = $1;
	    next;
	}
	elsif (/^\s*Entered-date\s*:\s*(.+)/i) {
	    $EnterDate = $1;
	    next;
	}
	elsif (/^\s*Description\s*:\s*(.+)/i) {
	    $Description = $1;
	    next;
	}
	elsif (/^\s*Keywords\s*:\s*(.+)/i) {
	    $Keywords = $1;
	    next;
	}
	elsif (/^\s*Source-code\s*:\s*(.+)/i) {
	    $Source = $1;
	    next;
	}
	elsif (/^\s*Formats\s*:\s*(.+)/i) {
	    $FormatsList = $1;
	    next;
	}
	elsif (/^\s*Authors?\s*:\s*(.+)/i) {
	    $Author = $1;
	    $Author =~ tr/\<\>/\(\)/;
	    next;
	}
	elsif (/^\s*Maintained-by\s*:\s*(.+)/i) {
	    $Maintainer = $1;
	    $Maintainer =~ tr/\<\>/\(\)/;
	    next;
	}
#	elsif (/^\s*Primary-site\s*:\s*(.+)/i) {
#	    $Site = $1;
#	    next;
#	}
	elsif (/^\s*Platforms\s*:\s*(.+)/i) {
	    $Platforms = $1;
	    next;
	}
	elsif (/^\s*Requires\s*:\s*(.+)/i) {
	    $Requires = $1;
	    next;
	}
	elsif (/^\s*Copying-policy\s*:\s*(.+)/i) {
	    $CopyingPolicy = $1;
	    $CopyingPolicy =~ tr/\<\>/\(\)/;
	    next;
	}
	else {
	    print STDERR "Unrecognised keyword near >>$_<<\n";
	    next;
	}

    }  # while


    # -----------------------------------------------------------------
    # Checking.

    if (! $Filename) {
	$Filename = $mainfile;
    }

    if (! $Description) {
	$Description = $short;
    }


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

#   printf T "Category:       %s\n", $Category;

	printf T "Program:        %s\n", $Filename;
	printf T "Title:          %s\n", $Title;
	printf T "Version:        %s\n", $Version;

    ($LSMmode || $EnterDate) &&
	printf T "Entered-date:   %s\n", $EnterDate;

	printf T "Description:    %s\n", $Description;
	printf T "Keywords:       %s\n", $Keywords;
# ---
# X2 Format additions
	printf T "Source-code:    %s\n", $Source;
    $FormatsList &&
	printf T "Formats:        %s\n", $FormatsList;
# ---
	printf T "Author:         %s\n", $Author;

    if ($LSMmode) {
	printf T "Maintained-by:  %s\n", $Maintainer;
#	printf T "Primary-site:   %s\n", "";
#	printf T "Alternate-site: %s\n", "";
#	printf T "Original-site:  %s\n", "";
    }  # LSM Format

	printf T "Platforms:      %s\n", $Platforms;
    $Requires &&
	printf T "Requires:       %s\n", $Requires;
	printf T "Copying-policy: %s\n", $CopyingPolicy;

    print T "\n";


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

    # After printing text version, protect any HTML metacharacters.

    $Description = html_encode($Description);
    $Requires    = html_encode($Requires);


    print H "<UL>\n";

#       printf H "<EM>Category:</EM>       %s\n", $Category;
	printf H "<EM>Program:</EM>        %s<BR>\n", $Filename;
	printf H "<EM>Title:</EM>          %s<BR>\n", $Title;
	printf H "<EM>Version:</EM>        %s<BR>\n", $Version;

    ($LSMmode || $EnterDate) &&
	printf H "<EM>Entered-date:</EM>   %s<BR>\n", $EnterDate;

	printf H "<EM>Description:</EM>    %s<BR>\n", $Description;
	printf H "<EM>Keywords:</EM>       %s<BR>\n", $Keywords;
# ---
# X2 Format additions
	printf H "<EM>Source-code:</EM>    %s<BR>\n", $Source;
    $FormatsList &&
	printf H "<EM>Formats:</EM>        %s<BR>\n", $FormatsList;
# ---
	printf H "<EM>Author:</EM>         %s<BR>\n", $Author;

    if ($LSMmode) {
	printf H "<EM>Maintained-by:</EM>  %s<BR>\n", $Maintainer;
#	printf H "<EM>Primary-site:</EM>   %s<BR>\n", "";
#	printf H "<EM>Alternate-site:</EM> %s<BR>\n", "";
#	printf H "<EM>Original-site:</EM>  %s<BR>\n", "";
    }  # LSM Format

	printf H "<EM>Platforms:</EM>      %s<BR>\n", $Platforms;
    $Requires &&
	printf H "<EM>Requires:</EM>       %s<BR>\n", $Requires;
	printf H "<EM>Copying-policy:</EM> %s<BR>\n", $CopyingPolicy;

    print H "</UL>\n\n";

}  # copy_long_description


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

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

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

    return $value;

}  # html_encode


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

sub print_text_headers {

print T "\n\t\t\tx2ftp.oulu.fi\n";
print T "\tX2 Support Group Game Development Archives\n";
print T "\nDirectory: /pub/msdos/programming/${DIR}\n";
print T "\n------------------------------------------------------------------------------\n";
print T "\n";
print T "Index updated:  $date\n";
print T "This is a new index file format. Please, tell us what you think about it.\n";
 print T "\n------------------------------------------------------------------------------\n\n";

}  # print_text_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><HR><P>\nThis is a new index file format. Please, tell us what you think about it.\nAny patches welcome. ;-)\n";
print H "<P><HR>";

}  # print_html_headers