#!/usr/bin/perl
# Vyakarana/Dhatupatha/mdhvindx/disp/mdhv-html.pl Mar 4, 2009
#reads filter option from command line.
# reads txt lines from STDIN, writes output to STDOUT
# Input should be like 'mdhv-sandhi-sort.txt' (only 1st fld used)
#Output file name should end in '.html'
# also reads 'filter' argument, which determines how to
# construct the output. filter should be one of these:
# X='SktRomanUnicode SktDevaUnicode SLP2HK' or
# SLP2SLP if no transliteration desired.
use lib qw(/Volumes/Sanskrit-RAID/jfunderburk/app/cvtskt);
use Sanskrit::EmbedUni;
#use Sanskrit::Embed;
my $index = $ARGV[0];
my $filter= $ARGV[1];
if (not($index)) {$index="index";}
if (!$filter) {$filter="SLP2SLP";}
print STDERR "index = $index, filter = $filter\n";
my $FILTER_DIR = "/Volumes/Sanskrit-Raid/jfunderburk/util/translit_filter";
my @FILTERS = qw(SktRomanUnicode SktDevaUnicode SLP2HK SLP2SLP SLP2ITRANS SktRomanManjushreeCSX SktRomanCSX);
{my $n;
my $found = 0;
foreach(@FILTERS) {
if ($_ eq $filter) {$found = 1};
}
if ($found eq 0) {die "invalid filter. valid are @FILTERS\n";}
}
# print header material for this file
my $HEADER="\n";
print "$HEADER\n";
print "\n";
print "\n";
print "Madhaviya index: $filter\n";
print '';
print "\n";
print "\n";
if (($filter =~ /UTF8|Unicode/) or ($filter eq "SLP2HK") or
($filter eq "SLP2SLP") or ($filter eq "SLP2ITRANS")){
print '' . "\n";
}else {
print '' . "\n";
}
print "\n";
print "\n";
my $table = "";
$table .= '
\n";
print STDERR "$nout lines written from $n lines processed.\n";
if ($filter ne "SLP2SLP") {
open FILTER, "| $FILTER_DIR/$filter";
print FILTER $table;
close FILTER;
}else {
$table =~ s/<\/?SA>//g;
print $table;
}
print "\n";
print "\n";
exit;
sub process {
my $s = shift;
($s1) = split(' : ',$s); # get first field.
my ($fullDAtu,$root,$marker,$sense,$gaRaNum,$gaRa_name,$sUtraNum,$page,$pa) = split(' ; ',$s1);
if ($index ne "index") {
$gaRaNum = "";
}
my $char = substr($root,0,1);
my $line = "";
$line .= "
\n";
# a few senses are very long, though most are not.
# trim the size of $sense
if (length ($sense) > 30) { # tried 40 first
$sense = substr($sense,0,30);
$sense =~ s/[kKgGNcCjJYwWqQRtTdDnpPbBmyrlvSzsh]+$//;
$sense =~ s/\(/ (/;
$sense =~ s/\[/ [/;
if (length($sense) <= 28) {
$line .= "
