#!/usr/local/bin/perl
#
# A. de la Fuente, VILSPA, 12-Jan-1995 
# Translation from dtsplit awk program
#
 
$Usage = 'Usage: dtsplit.pl  style=vilspa <input_file.html' ;

eval "exec /usr/local/bin/perl -S $0 $*"
    if $running_under_some_shell;
			# this emulates #! processing on NIH machines.
			# (remove #! line above if indigestible)

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

#######!/usr/local/bin/gawk -f
#
# Use:    dtsplit [-v style={vilspa}] <input_file.html
#
# This program copies <DT> resource records into separate files by category.
#
# D.Wells, NRAO-CV, 03Feb-22Mar94,31Jly,23Sep.
#    1/5/95: added support for mailing_list category
#    1/6/95: every category is now in a separate page.
# To-Do:
# -produce a page with a list of active CATEGORY codes
# -set up main page as though subordinate page to eliminate special code
# --------------------------------------------------------------------
$[ = 1;			# set array base to 1
$, = ' ';		# set output field separator
$\ = "\n";		# set output record separator

# bc() specifies 'big' categories, ch() specifies ordinary categories;
#
if ($style eq 'vilspa') {
    $link_address = 'Sergio Paoli spaoli@fcaglp.fcaglp.unlp.edu.ar';
    $main_page = 'astronomy';
    $astroweb_path = './';
    #
    &bc('observing', 'Observing resources');
    &ch('telescope', 'Telescopes');
    &ch('survey', 'Astronomical survey projects');

    &ch('schedule', 'Telescope observing schedules');
    &ch('weather', 'Metereological information');
    #
    &bc('data', 'Data Resources');
    &ch('center', 'Data and Archive Centers');

    &ch('infosys', 'Astronomy Information Systems');

    #
    &bc('library', 'Publication-related Resources');
    &ch('preprint', 'Astronomy &amp astrophysics preprints &amp abstracts');
    &ch('abstracts', 'Abstracts of Astronomical Publications');
    &ch('full_text', 'Full-texts of Astronomical Publications');
    &ch('bibliography', 'Astronomical Bibliographical Services');
    &ch('library', 'Astronomy-related Libraries');

    &ch('other_library', 'Other library resources');

    #
    &bc('people', 'People-related Resources');
    &ch('people', 'People');
    &ch('personal', 'Personal Web pages');
    &ch('jobs', 'Jobs');
    &ch('meetings', 'Conferences and Meetings');
    &ch('newsgroup', 'Newsgroups');
    &ch('mailing_list', 'Mailing Lists');
    #
    &bc('organization', 'Organizations');
    &ch('dept', 'Astronomy Departments');

    &ch('society', 'Astronomical Societies');

    &ch('agency', 'Space Agencies');

    #
    &bc('software', 'Software Resources');
    &ch('software', 'Astronomy software servers');

    &ch('document_tools', 'Document Preparation Tools (TeX,etc)');
    #
    &bc('areas', 'Research areas of Astronomy');
    &ch('radio', 'Radio Astronomy');
    &ch('optical', 'Optical Astronomy');

    &ch('space', 'Space Astronomy');

    &ch('solar', 'Solar Astronomy');
    &ch('planetary', 'Planetary Astronomy');
    &ch('high_energy', 'High Energy Astronomy');
    #
    &bc('astroweb', 'Various Lists of Astronomy Resources');
    &ch('astroweb', 'Primary Lists of Astronomy Resources');
    &ch('other_astronomy', 'Other lists of astronomy resources');

    #
    &bc('pictures', 'Astronomical Imagery');
    &ch('pictures', 'Pictures');

    #
    &bc('education', 'Education');
    &ch('education', 'Educational Resources in Astronomy');

    #
    &bc('history', 'History');
    &ch('history', 'History of Astronomy');

    #
    &bc('misc', 'Miscellaneous Resources');
    &ch('protocols', 'Overviews &amp technical notes for protocols');
    &ch('misc', 'Miscellaneous Resources');
    &ch('earth', 'Atmosphere,Ocean,Geophys.,SpaceSci.');
    &ch('physics', 'Physics-related resources');
    &ch('comp_sci', 'Computer Science-related resources');
    #

    ;
}
else {
    printf("$Usage \n");
    $ExitValue = (17);
    exit;
    # -----------------------------------------------------------------------
    # Prefix for header files:
    ;
}
$header_prefix = 'head';
# Prefix for temporary files:
$temp_prefix = 'temp';
# Prefix for output pages:
$output_prefix = 'yp';
# Filename for the main page:
$header_main = $header_prefix . '_' . $main_page . '.html';
$output_main = $output_prefix . '_' . $main_page . '.html';

# Filename for the tail file:

$tailfile = $temp_prefix . '_tail.html';
open (tailfile,">$tailfile");
($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = gmtime(time);
$mon = $mon + 1 ;
$timeupdate = "$year/$mon/$mday $hour:$min GMT" ;
print tailfile "<ADDRESS>" ;
print tailfile "Updated on $timeupdate by $link_address " ;
print tailfile "</ADDRESS>\n";
close (tailfile) ;
#
$category_table = $output_prefix . '_categories.html';
# -----------------------------------------------------------------------
# initialize the output files with header lines:
foreach $i (keys %h) {
    if ($h{$i} ne '=') {
	# printf("<A NAME=\"%s\">\n", c[i]) > temp_file(i);
	&Pick('>>', &temp_file($i)) &&
	    (printf $fh "<H1>\n%s\n</H1>\n<DL>\n\n", $h{$i});
    }
}

#
# --------------------------------------------------------------------
# Resource records begin with <DT>. The lines of the record are
# concatenated into one string, delimited by code '||'. The record
# is then appended to the appropriate file based on CATEGORY code(s)
# defined in the configuration information in the BEGIN section.

line: while (<>) {
    chop;	# strip record separator
    @Fld = split(' ', $_, 9999);
    if ($_ =~ /^[<][Dd][Tt][>]/) {
	$e = $_ . '||';

	# read from stdin, appending lines to string, until encounter blank line:
	do {
	    $_ = &Getline0();

	    $e = $e . $_ . '||';
	}
	 while length > 0 ;

	# locate CATEGORY line and parse it:
	if (($u1 = index($e, '<!-- CATEGORY ')) == 0) {
	    print 'no CATEGORY!?! e[30]=' . substr($e, 1, 30);
	}
	else {
	    $u2 = index(substr($e, $u1, 999999), ' -->') + $u1 - 1;
	    $ec = substr($e, $u1, $u2 + 5 - $u1);
	    $nf = (@cats = split(' ', $ec, 9999));
	    # the strings of the CATEGORY definition have been parsed into cats[nf]:
	    foreach $i (keys %done_cat) {
		delete $done_cat{$i};
	    }
	    for ($i = 3; $i < $nf; $i++) {	#???
		$nomatch = 1;

		foreach $j (keys %c) {
		    if ($cats[$i] eq $c{$j}) {	#???
			$nomatch = 0;
			if (!(defined $done_cat{&temp_file($j)})) {
			    &print_entry($e, $j);
			    $ncat{$j}++;
			    $done_cat{&temp_file($j)} = 1;
 			    last;
			}
		    }
		}
		if ($nomatch) {
		    $unmatch{$cats[$i]}++;
		}
	    }
	    $nrec++;
	}
    }
    # --------------------------------------------------------------------
    # --------------------------------------------------------------------
    # Function to append resource record 'e' to temp_file(i):
    # --------------------------------------------------------------------
    # ------------------------------------------------------------------
    # Category code configuration for this set of pages:
    #    the c[] are the actual codes which appear in <DT> files
    #    the h[] are the <H2> strings for listings of the categories
    #    the t[] are the names of the pages to which categories are appended
    #    the h1[] are <TITLE>/<H1> strings for multiple-category pages
    #
    # Function to compute "temp" filenames. A name is formed from category code
    # c[i] unless f[i] is set to an alternate name. If a[i] is set to the
    # subscript of a different code, then the filename will be formed from
    # the c[] or a[] entries of that subscript. This scheme enables codes
    # to be aliased. 
    # ------------------------------------------------------------------
    # Function to compute "header" filenames.
    # Function to compute "output" filenames.
}

printf "%4d resource records from input\n", $nrec;
# initialize the category table:
&Pick('>', $category_table) &&
    (print $fh '<LISTING>');
&Pick('>>', $category_table) &&
    (printf $fh ('The AstroWeb master database currently'));
&Pick('>>', $category_table) &&
    (printf $fh "contains %d resource records,\n", $nrec);
&Pick('>>', $category_table) &&
    (printf $fh ("classified into the following categories:\n\n"));
&Pick('>>', $category_table) &&
    (print $fh '#Records  Category_codes');
&Pick('>>', $category_table) &&
    (print $fh '--------  --------------');
# terminate the "temp" output files:
for ($i = 1; $i <= $k; $i++) {	#???
    if ($h{$i} ne '=') {
	&Pick('>>', $category_table) &&
	    (printf $fh "%8d  %-20s %s\n", $ncat{$i}, $c{$i}, $h{$i});
	&Pick('>>', &temp_file($i)) &&
	    (print $fh "</DL>\n");
	delete $opened{&temp_file($i)} && close(&temp_file($i));
    }
}
$num_unmatch = 0;
foreach $i (keys %unmatch) {
    $num_unmatch++;
}
printf "%4d unrecognized category codes:\n", $num_unmatch;
foreach $i (keys %unmatch) {
    printf "\t%4d cases of '%s'\n", $unmatch{$i}, $i;
    # initialize output files and copy header files to them:
    ;
}
for ($i = 1; $i <= $k; $i++) {	#???
    if ($initialized{&output_file($i)} != 1) {
	&Pick('>', &output_file($i)) &&
	    (printf $fh ("<HTML>\n<HEAD>\n<TITLE>\n"));
	&Pick('>>', &output_file($i)) &&
	    (printf $fh "%s\n", $h{$i});
	&Pick('>>', &output_file($i)) &&
	    (printf $fh "</TITLE>\n<LINK REV=\"made\" HREF=\"%s\">\n", 
	$link_address);
	&Pick('>>', &output_file($i)) &&
	    (printf $fh ("</HEAD>\n<BODY>\n"));

	while  ((($_ = &Getline2(&header_file($i)),$getline_ok)) > 0) {
	    &Pick('>>', &output_file($i)) &&
		(print $fh $_);
	} 
	&Pick('>>', &output_file($i)) &&
	    (printf $fh ("\n"));
	$initialized{&output_file($i)} = 1;
    }
}
if (!(-f  $header_main)){
   system('touch ' . $header_main );
} 
system('cp ' . $header_main . ' ' . $output_main);
&Pick('>>', $output_main) &&
    (printf $fh

      "The AstroWeb <A HREF=\"./newmaster.html\">master database</A>\ncurrently contains %d distinct resource records:</FORM><UL><UL>\n",

      
$nrec);
# concatenate the "temp" files into the "output" files:
for ($i = 1; $i <= $k; $i++) {	#???
    if ($h{$i} ne '=') {
	while (($_ = &Getline2(&temp_file($i)),$getline_ok) > 0) {
	    &Pick('>>', &output_file($i)) &&
		(print $fh $_);
	}
	delete $opened{&temp_file($i)} && close(&temp_file($i));
	&Pick('>>', &output_file($i)) &&
	    (print $fh "<HR>\n");
	system('rm -f ' . &temp_file($i));
	#
	if ($h1t{$bch{$i}} ne 'done') {
	    &Pick('>>', $output_main) &&
		(print $fh ("\n</UL><LI><B>" . $h1{$bch{$i}} . "</B><UL>\n"));
	    $h1t{$bch{$i}} = 'done';
	}
	&Pick('>>', $output_main) &&
	    (print $fh ("<LI><A HREF=\"" . $astroweb_path . &output_file($i) .

	      "\">"));

	&Pick('>>', $output_main) &&
	    (printf $fh "%s</A> (%d records)\n", $h{$i}, $ncat{$i});
    }
}
foreach $i (keys %c) {
    if ($done{&output_file($i)}!= 1) {
	while ((($_ = &Getline2($tailfile),$getline_ok)) > 0) {
	    &Pick('>>', &output_file($i)) &&
		(print $fh $_);
	}
	delete $opened{$tailfile} && close($tailfile);
	&Pick('>>', &output_file($i)) &&
	    (print $fh "</BODY>\n</HTML>");
	delete $opened{&output_file($i)} && close(&output_file($i));
	$done{&output_file($i)} = 1;
    }
}
&Pick('>>', $output_main) &&
    (print $fh "\n</UL></UL>");

while ((($_ = &Getline2($tailfile),$getline_ok)) > 0) {
    &Pick('>>', &output_file($i)) &&
       (print $fh $_);
    }
   

delete $opened{$tailfile} && close($tailfile);
&Pick('>>', $output_main) &&
    (print $fh "\n</BODY>\n</HTML>\n");

exit $ExitValue;

sub print_entry {
    local($e, $i) = @_;
    $tfi = &temp_file($i);
    while (length($e) > 0) {
	$p = index($e, '||');
	# find end of this line
	&Pick('>>', $tfi) &&
	    (printf $fh "%s\n", substr($e, 1, $p - 1));
	$e = substr($e, $p + 2, 999999);
    }
}

sub bc {
    local($big_category_code, $big_category_header) = @_;
    $mc = $big_category_code;
    $h1{$mc} = $big_category_header;
}

sub ch {
    local($category_code, $category_header) = @_;
    $c{++$k} = $category_code;

    $h{$k} = $category_header;

    if ($h{$k} eq '=') {
	if ($k == 1) {
	    print 'ABORT! cannot determine category alias!';
	    $ExitValue = (13); last line;
	}
	$a{$k} = (defined $a{($k - 1)}) ? $a{$k - 1} : ($k - 1);
    }
    $bch{$k} = $mc;
    # mc is set by bc() above 
    $t{$k} = $c{$k};
    # produce separate files for categories
}

sub temp_file {
    local($i) = @_;
    $j = (defined $a{$i}) ? $a{$i} : $i;

    $filename = (defined $f{$j}) ? $f{$j} : $c{$j};
    $temp_prefix . '_' . $filename . '.html';
}

sub header_file {
    local($i) = @_;
    $filename = $t{$i};
    $header_prefix . '_' . $filename . '.html';
}

sub output_file {
    local($i) = @_;
    $filename = $t{$i};
    $output_prefix . '_' . $filename . '.html';
}

sub Getline0 {
    if ($getline_ok = (($_ = <>) ne '')) {
	chop;	# strip record separator
	@Fld = split(' ', $_, 9999);
    }
    $_;
}

sub Getline2 {
    &Pick('',@_);
    if ($getline_ok = (($_ = <$fh>) ne '')) {
	chop;	# strip record separator
	@Fld = split(' ', $_, 9999);
    }
    $_;
}

sub Pick {
    local($mode,$name,$pipe) = @_;
    $fh = $name;
    open($name,$mode.$name.$pipe) unless $opened{$name}++;
}